An operating system release is considered "supported" if the vendor continues to provide security patches for the product. With an unsupported release, it will not be possible to resolve security issues discovered in the system software.

Red Hat offers the Extended Update Support (EUS) add-on to a Red Hat Enterprise Linux subscription, for a fee, for those customers who wish to standardize on a specific minor release for an extended period.

Verify that the version or RHEL 9 is vendor supported with the following command:

$ cat /etc/redhat-release 

Red Hat Enterprise Linux release 9.2 (Plow)

If the installed version of RHEL 9 is not supported, this is a finding.

Upgrade to a supported version of RHEL 9.

Installing software updates is a fundamental mitigation against the exploitation of publicly known vulnerabilities. If the most recent security patches and updates are not installed, unauthorized users may take advantage of weaknesses in the unpatched software. The lack of prompt attention to patching could result in a system compromise.

Verify RHEL 9 security patches and updates are installed and up to date. Updates are required to be applied with a frequency determined by organizational policy.

Obtain the list of available package security updates from Red Hat. The URL for updates is https://access.redhat.com/errata-search/. It is important to note that updates provided by Red Hat may not be present on the system if the underlying packages are not installed.

Check that the available package security updates have been installed on the system with the following command:

$ dnf history list | more

    ID | Command line | Date and time | Action(s) | Altered    
-------------------------------------------------------------------------------    
   70 | install aide | 2023-03-05 10:58 | Install | 1    
   69 | update -y | 2023-03-04 14:34 | Update | 18 EE    
   68 | install vlc | 2023-02-21 17:12 | Install | 21   
   67 | update -y | 2023-02-21 17:04 | Update | 7 EE 

Typical update frequency may be overridden by Information Assurance Vulnerability Alert (IAVA) notifications from CYBERCOM.

If the system is in noncompliance with the organizational patching policy, this is a finding.

Install RHEL 9 security patches and updates at the organizationally defined frequency. If system updates are installed via a centralized repository that is configured on the system, all updates can be installed with the following command:

$ sudo dnf update

Display of a standardized and approved use notification before granting access to the operating system ensures privacy and security notification verbiage used is consistent with applicable federal laws, Executive Orders, directives, policies, regulations, standards, and guidance.

System use notifications are required only for access via login interfaces with human users and are not required when such human interfaces do not exist.

Satisfies: SRG-OS-000023-GPOS-00006, SRG-OS-000228-GPOS-00088

Verify RHEL 9 displays the Standard Mandatory DOD Notice and Consent Banner before granting access to the operating system via a command line user logon.

Check that a banner is displayed at the command line login screen with the following command:

$ sudo cat /etc/issue

If the banner is set correctly it will return the following text:

"You are accessing a U.S. Government (USG) Information System (IS) that is provided for USG-authorized use only.

By using this IS (which includes any device attached to this IS), you consent to the following conditions:

-The USG routinely intercepts and monitors communications on this IS for purposes including, but not limited to, penetration testing, COMSEC monitoring, network operations and defense, personnel misconduct (PM), law enforcement (LE), and counterintelligence (CI) investigations.

-At any time, the USG may inspect and seize data stored on this IS.

-Communications using, or data stored on, this IS are not private, are subject to routine monitoring, interception, and search, and may be disclosed or used for any USG-authorized purpose.

-This IS includes security measures (e.g., authentication and access controls) to protect USG interests--not for your personal benefit or privacy.

-Notwithstanding the above, using this IS does not constitute consent to PM, LE or CI investigative searching or monitoring of the content of privileged communications, or work product, related to personal representation or services by attorneys, psychotherapists, or clergy, and their assistants. Such communications and work product are private and confidential. See User Agreement for details."

If the banner text does not match the Standard Mandatory DOD Notice and Consent Banner exactly, or the line is commented out, this is a finding.

Configure RHEL 9 to display the Standard Mandatory DOD Notice and Consent Banner before granting access to the system via command line logon.

Edit the "/etc/issue" file to replace the default text with the Standard Mandatory DOD Notice and Consent Banner. The DOD-required text is:

"You are accessing a U.S. Government (USG) Information System (IS) that is provided for USG-authorized use only.

By using this IS (which includes any device attached to this IS), you consent to the following conditions:

-The USG routinely intercepts and monitors communications on this IS for purposes including, but not limited to, penetration testing, COMSEC monitoring, network operations and defense, personnel misconduct (PM), law enforcement (LE), and counterintelligence (CI) investigations.

-At any time, the USG may inspect and seize data stored on this IS.

-Communications using, or data stored on, this IS are not private, are subject to routine monitoring, interception, and search, and may be disclosed or used for any USG-authorized purpose.

-This IS includes security measures (e.g., authentication and access controls) to protect USG interests -- not for your personal benefit or privacy.

-Notwithstanding the above, using this IS does not constitute consent to PM, LE or CI investigative searching or monitoring of the content of privileged communications, or work product, related to personal representation or services by attorneys, psychotherapists, or clergy, and their assistants. Such communications and work product are private and confidential. See User Agreement for details."

Unnecessary service packages must not be installed to decrease the attack surface of the system. Graphical display managers have a long history of security vulnerabilities and must not be used, unless approved and documented.

Verify that RHEL 9 is configured to boot to the command line:

$ systemctl get-default

multi-user.target

If the system default target is not set to "multi-user.target" and the information system security officer (ISSO) lacks a documented requirement for a graphical user interface, this is a finding.

Document the requirement for a graphical user interface with the ISSO or set the default target to multi-user with the following command:

$ sudo systemctl set-default multi-user.target

The most important characteristic of a random number generator is its randomness, namely its ability to deliver random numbers that are impossible to predict. Entropy in computer security is associated with the unpredictability of a source of randomness. The random source with high entropy tends to achieve a uniform distribution of random values. Random number generators are one of the most important building blocks of cryptosystems.  

The rngd service feeds random data from hardware device to kernel random device. Quality (nonpredictable) random number generation is important for several security functions (i.e., ciphers).

Note: For RHEL 9 systems running with kernel FIPS mode enabled as specified by RHEL-09-671010, this requirement is Not Applicable.

Verify that RHEL 9 has enabled the hardware random number generator entropy gatherer service with the following command:

$ systemctl is-active rngd

active

If the "rngd" service is not active, this is a finding.

Install the rng-tools package with the following command:

$ sudo dnf install rng-tools

Then enable the rngd service run the following command:

$ sudo systemctl enable --now rngd

In the event of a system failure, RHEL 9 must preserve any information necessary to determine cause of failure and any information necessary to return to operations with least disruption to system processes.

Verify that "systemd-journald" is active with the following command:

$ systemctl is-active systemd-journald

active

If the systemd-journald service is not active, this is a finding.

To enable the systemd-journald service, run the following command:

$ sudo systemctl enable --now systemd-journald

A locally logged-on user who presses Ctrl-Alt-Delete when at the console can reboot the system. If accidentally pressed, as could happen in the case of a mixed OS environment, this can create the risk of short-term loss of availability of systems due to unintentional reboot. In a graphical user environment, risk of unintentional reboot from the Ctrl-Alt-Delete sequence is reduced because the user will be prompted before any action is taken.

Satisfies: SRG-OS-000324-GPOS-00125, SRG-OS-000480-GPOS-00227

Verify RHEL 9 is configured to not reboot the system when Ctrl-Alt-Delete is pressed seven times within two seconds with the following command:

$ grep -i ctrl /etc/systemd/system.conf

CtrlAltDelBurstAction=none

If the "CtrlAltDelBurstAction" is not set to "none", commented out, or is missing, this is a finding.

Configure the system to disable the CtrlAltDelBurstAction by added or modifying the following line in the "/etc/systemd/system.conf" configuration file:

CtrlAltDelBurstAction=none

Reload the daemon for this change to take effect.

$ sudo systemctl daemon-reload

A locally logged-on user who presses Ctrl-Alt-Delete when at the console can reboot the system. If accidentally pressed, as could happen in the case of a mixed OS environment, this can create the risk of short-term loss of availability of systems due to unintentional reboot. In a graphical user environment, risk of unintentional reboot from the Ctrl-Alt-Delete sequence is reduced because the user will be prompted before any action is taken.

Satisfies: SRG-OS-000324-GPOS-00125, SRG-OS-000480-GPOS-00227

Verify RHEL 9 is not configured to reboot the system when Ctrl-Alt-Delete is pressed with the following command:

$ sudo systemctl status ctrl-alt-del.target

ctrl-alt-del.target
Loaded: masked (Reason: Unit ctrl-alt-del.target is masked.)
Active: inactive (dead)

If the "ctrl-alt-del.target" is loaded and not masked, this is a finding.

Configure RHEL 9 to disable the ctrl-alt-del.target with the following command:

$ sudo systemctl disable --now ctrl-alt-del.target
$ sudo systemctl mask --now ctrl-alt-del.target

The debug-shell requires no authentication and provides root privileges to anyone who has physical access to the machine. While this feature is disabled by default, masking it adds an additional layer of assurance that it will not be enabled via a dependency in systemd. This also prevents attackers with physical access from trivially bypassing security on the machine through valid troubleshooting configurations and gaining root access when the system is rebooted.

Satisfies: SRG-OS-000324-GPOS-00125, SRG-OS-000480-GPOS-00227

Verify RHEL 9 is configured to mask the debug-shell systemd service with the following command:

$ sudo systemctl status debug-shell.service

debug-shell.service
Loaded: masked (Reason: Unit debug-shell.service is masked.)
Active: inactive (dead)

If the "debug-shell.service" is loaded and not masked, this is a finding.

Configure RHEL 9 to mask the debug-shell systemd service with the following command:

$ sudo systemctl disable --now debug-shell.service
$ sudo systemctl mask --now debug-shell.service

To mitigate the risk of unauthorized access to sensitive information by entities that have been issued certificates by DOD-approved PKIs, all DOD systems (e.g., web servers and web portals) must be properly configured to incorporate access control methods that do not rely solely on the possession of a certificate for access. Successful authentication must not automatically give an entity access to an asset or security boundary. Authorization procedures and controls must be implemented to ensure each authenticated entity also has a validated and current authorization. Authorization is the process of determining whether an entity, once authenticated, is permitted to access a specific asset. Information systems use access control policies and enforcement mechanisms to implement this requirement.

Password protection on the boot loader configuration ensures users with physical access cannot trivially alter important bootloader settings. These include which kernel to use, and whether to enter single-user mode.

Verify the boot loader superuser password has been set and run the following command:

$ sudo grep "superusers" /etc/grub2.cfg 

password_pbkdf2  superusers-account   ${GRUB2_PASSWORD}  

To verify the boot loader superuser account password has been set, and the password encrypted, run the following command:

$ sudo cat /boot/grub2/user.cfg 

GRUB2_PASSWORD=grub.pbkdf2.sha512.10000.C4E08AC72FBFF7E837FD267BFAD7AEB3D42DDC
2C99F2A94DD5E2E75C2DC331B719FE55D9411745F82D1B6CFD9E927D61925F9BBDD1CFAA0080E0
916F7AB46E0D.1302284FCCC52CD73BA3671C6C12C26FF50BA873293B24EE2A96EE3B57963E6D7
0C83964B473EC8F93B07FE749AA6710269E904A9B08A6BBACB00A2D242AD828 

If a "GRUB2_PASSWORD" is not set, this is a finding.

Configure RHEL 9 to require a grub bootloader password for the grub superuser account.

Generate an encrypted grub2 password for the grub superuser account with the following command:

$ sudo grub2-setpassword
Enter password:
Confirm password:

Using interactive or recovery boot, the console user could disable auditing, firewalls, or other services, weakening system security.

Verify that GRUB 2 is configured to disable interactive boot.

Check that the current GRUB 2 configuration disables the ability of systemd to spawn an interactive boot process with the following command:

$ sudo grubby --info=ALL | grep args | grep 'systemd.confirm_spawn'

If any output is returned, this is a finding.

Configure RHEL 9 to allocate sufficient audit_backlog_limit to disable the ability of systemd to spawn an interactive boot process with the following command:

$ sudo grubby --update-kernel=ALL --remove-args="systemd.confirm_spawn"

Having a nondefault grub superuser username makes password-guessing attacks less effective.

Verify the boot loader superuser account has been set with the following command:

$ sudo grep -A1 "superusers" /etc/grub2.cfg 

set superusers=""
export superusers
 
The  is the actual account name different from common names like root, admin, or administrator.

If superusers contains easily guessable usernames, this is a finding.

Configure RHEL 9 to have a unique username for the grub superuser account.

Edit the "/etc/grub.d/01_users" file and add or modify the following lines with a nondefault username for the superusers account:

set superusers=""
export superusers

Once the superuser account has been added, update the grub.cfg file by running:

$ sudo grubby --update-kernel=ALL

The "root" group is a highly privileged group. Furthermore, the group-owner of this file should not have any access privileges anyway.

Verify the group ownership of the "/boot/grub2/grub.cfg" file with the following command:

$ sudo stat -c "%G %n" /boot/grub2/grub.cfg 

root /boot/grub2/grub.cfg

If "/boot/grub2/grub.cfg" file does not have a group owner of "root", this is a finding.

Change the group of the file /boot/grub2/grub.cfg to root by running the following command:

$ sudo chgrp root /boot/grub2/grub.cfg

The " /boot/grub2/grub.cfg" file stores sensitive system configuration. Protection of this file is critical for system security.

Verify the ownership of the "/boot/grub2/grub.cfg" file with the following command:

$ sudo stat -c "%U %n" /boot/grub2/grub.cfg 

root /boot/grub2/grub.cfg 

If "/boot/grub2/grub.cfg" file does not have an owner of "root", this is a finding.

Change the owner of the file /boot/grub2/grub.cfg to root by running the following command:

$ sudo chown root /boot/grub2/grub.cfg

System calls are special routines in the Linux kernel, which userspace applications ask to do privileged tasks. Invoking a system call is an expensive operation because the processor must interrupt the currently executing task and switch context to kernel mode and then back to userspace after the system call completes. Virtual system calls map into user space a page that contains some variables and the implementation of some system calls. This allows the system calls to be executed in userspace to alleviate the context switching expense.

Virtual system calls provide an opportunity of attack for a user who has control of the return instruction pointer. Disabling virtual system calls help to prevent return oriented programming (ROP) attacks via buffer overflows and overruns. If the system intends to run containers based on RHEL 6 components, then virtual system calls will need enabled so the components function properly.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000134-GPOS-00068

Verify the current GRUB 2 configuration disables virtual system calls with the following command:

$ sudo grubby --info=ALL | grep args | grep -v 'vsyscall=none'

If any output is returned, this is a finding.

Check that virtual system calls are disabled by default to persist in kernel updates with the following command: 

$ sudo grep vsyscall /etc/default/grub

GRUB_CMDLINE_LINUX="vsyscall=none"

If "vsyscall" is not set to "none", is missing or commented out, and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Document the use of virtual system calls with the ISSO as an operational requirement or disable them with the following command:

$ sudo grubby --update-kernel=ALL --args="vsyscall=none"

Add or modify the following line in "/etc/default/grub" to ensure the configuration survives kernel updates:

GRUB_CMDLINE_LINUX="vsyscall=none"

Poisoning writes an arbitrary value to freed pages, so any modification or reference to that page after being freed or before being initialized will be detected and prevented. This prevents many types of use-after-free vulnerabilities at little performance cost. Also prevents leak of data and detection of corrupted memory.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000134-GPOS-00068

Verify that GRUB 2 is configured to enable page poisoning to mitigate use-after-free vulnerabilities.

Check that the current GRUB 2 configuration has page poisoning enabled  with the following command:

$ sudo grubby --info=ALL | grep args | grep -v 'page_poison=1'

If any output is returned, this is a finding.

Check that page poisoning is enabled by default to persist in kernel updates with the following command: 

$ sudo grep page_poison /etc/default/grub

GRUB_CMDLINE_LINUX="page_poison=1"

If "page_poison" is not set to "1", is missing or commented out, this is a finding.

Configure RHEL 9 to enable page poisoning with the following commands:

$ sudo grubby --update-kernel=ALL --args="page_poison=1"

Add or modify the following line in "/etc/default/grub" to ensure the configuration survives kernel updates:

GRUB_CMDLINE_LINUX="page_poison=1"

Some adversaries launch attacks with the intent of executing code in nonexecutable regions of memory or in memory locations that are prohibited. Security safeguards employed to protect memory include, for example, data execution prevention and address space layout randomization. Data execution prevention safeguards can be either hardware-enforced or software-enforced with hardware providing the greater strength of mechanism.

Poisoning writes an arbitrary value to freed pages, so any modification or reference to that page after being freed or before being initialized will be detected and prevented. This prevents many types of use-after-free vulnerabilities at little performance cost. Also prevents leak of data and detection of corrupted memory.

SLAB objects are blocks of physically contiguous memory. SLUB is the unqueued SLAB allocator.

Satisfies: SRG-OS-000433-GPOS-00192, SRG-OS-000134-GPOS-00068

Verify that GRUB 2 is configured to enable poisoning of SLUB/SLAB objects to mitigate use-after-free vulnerabilities with the following commands:

Check that the current GRUB 2 configuration has poisoning of SLUB/SLAB objects enabled:

$ sudo grubby --info=ALL | grep args | grep -v 'slub_debug=P' 

If any output is returned, this is a finding.

Check that poisoning of SLUB/SLAB objects is enabled by default to persist in kernel updates: 

$ sudo grep slub_debug /etc/default/grub

GRUB_CMDLINE_LINUX="slub_debug=P"

If "slub_debug" does not contain "P", is missing, or is commented out, this is a finding.

Configure RHEL  to enable poisoning of SLUB/SLAB objects with the following commands:

$ sudo grubby --update-kernel=ALL --args="slub_debug=P"

Add or modify the following line in "/etc/default/grub" to ensure the configuration survives kernel updates:

GRUB_CMDLINE_LINUX="slub_debug=P"

Kernel page-table isolation is a kernel feature that mitigates the Meltdown security vulnerability and hardens the kernel against attempts to bypass kernel address space layout randomization (KASLR).

Satisfies: SRG-OS-000433-GPOS-00193, SRG-OS-000095-GPOS-00049

Verify RHEL 9 enables kernel page-table isolation with the following command:

$ sudo grubby --info=ALL | grep pti

args="ro crashkernel=auto resume=/dev/mapper/rhel-swap rd.lvm.lv=rhel/root rd.lvm.lv=rhel/swap rhgb quiet fips=1 audit=1 audit_backlog_limit=8192 pti=on 

If the "pti" entry does not equal "on", or is missing, this is a finding.

Check that kernel page-table isolation is enabled by default to persist in kernel updates: 

$ sudo grep pti /etc/default/grub

GRUB_CMDLINE_LINUX="pti=on"

If "pti" is not set to "on", is missing or commented out, this is a finding.

Configure RHEL 9 to enable kernel page-table isolation with the following command:

$ sudo grubby --update-kernel=ALL --args="pti=on"

Add or modify the following line in "/etc/default/grub" to ensure the configuration survives kernel updates:

GRUB_CMDLINE_LINUX="pti=on"

Without the capability to generate audit records, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

If auditing is enabled late in the startup process, the actions of some startup processes may not be audited. Some audit systems also maintain state information only available if auditing is enabled before a given process is created.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000473-GPOS-00218, SRG-OS-000254-GPOS-00095

Verify that GRUB 2 is configured to enable auditing of processes that start prior to the audit daemon with the following commands:

Check that the current GRUB 2 configuration enabled auditing:

$ sudo grubby --info=ALL | grep audit

args="ro crashkernel=auto resume=/dev/mapper/rhel-swap rd.lvm.lv=rhel/root rd.lvm.lv=rhel/swap rhgb quiet fips=1 audit=1 audit_backlog_limit=8192 pti=on 

If "audit" is not set to "1" or is missing, this is a finding.

Check that auditing is enabled by default to persist in kernel updates: 

$ sudo grep audit /etc/default/grub

GRUB_CMDLINE_LINUX="audit=1"

If "audit" is not set to "1", is missing, or is commented out, this is a finding.

Enable auditing of processes that start prior to the audit daemon with the following command:

$ sudo grubby --update-kernel=ALL --args="audit=1"

Add or modify the following line in "/etc/default/grub" to ensure the configuration survives kernel updates:

GRUB_CMDLINE_LINUX="audit=1"

Preventing unauthorized information transfers mitigates the risk of information, including encrypted representations of information, produced by the actions of prior users/roles (or the actions of processes acting on behalf of prior users/roles) from being available to any current users/roles (or current processes) that obtain access to shared system resources (e.g., registers, main memory, hard disks) after those resources have been released back to information systems. The control of information in shared resources is also commonly referred to as object reuse and residual information protection.

This requirement generally applies to the design of an information technology product, but it can also apply to the configuration of particular information system components that are, or use, such products. This can be verified by acceptance/validation processes in DOD or other government agencies.

There may be shared resources with configurable protections (e.g., files in storage) that may be assessed on specific information system components.

Restricting access to the kernel message buffer limits access to only root. This prevents attackers from gaining additional system information as a nonprivileged user.

Satisfies: SRG-OS-000132-GPOS-00067, SRG-OS-000138-GPOS-00069

Verify RHEL 9 is configured to restrict access to the kernel message buffer with the following commands:

Check the status of the kernel.dmesg_restrict kernel parameter.

$ sudo sysctl kernel.dmesg_restrict

kernel.dmesg_restrict = 1

If "kernel.dmesg_restrict" is not set to "1" or is missing, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F kernel.dmesg_restrict | tail -1

kernel.dmesg_restrict = 1

If "kernel.dmesg_restrict" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to restrict access to the kernel message buffer.

Add or edit the following line in a system configuration file, in the "/etc/sysctl.d/" directory:

kernel.dmesg_restrict = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Preventing unauthorized information transfers mitigates the risk of information, including encrypted representations of information, produced by the actions of prior users/roles (or the actions of processes acting on behalf of prior users/roles) from being available to any current users/roles (or current processes) that obtain access to shared system resources (e.g., registers, main memory, hard disks) after those resources have been released back to information systems. The control of information in shared resources is also commonly referred to as object reuse and residual information protection.

This requirement generally applies to the design of an information technology product, but it can also apply to the configuration of particular information system components that are, or use, such products. This can be verified by acceptance/validation processes in DOD or other government agencies.

There may be shared resources with configurable protections (e.g., files in storage) that may be assessed on specific information system components.

Setting the kernel.perf_event_paranoid kernel parameter to "2" prevents attackers from gaining additional system information as a nonprivileged user.

Satisfies: SRG-OS-000132-GPOS-00067, SRG-OS-000138-GPOS-00069

Verify RHEL 9 is configured to prevent kernel profiling by nonprivileged users with the following commands:

Check the status of the kernel.perf_event_paranoid kernel parameter.

$ sudo sysctl kernel.perf_event_paranoid

kernel.perf_event_paranoid = 2

If "kernel.perf_event_paranoid" is not set to "2" or is missing, this is a finding.
Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config  | egrep -v '^(#|;)' | grep -F kernel.perf_event_paranoid | tail -1

kernel.perf_event_paranoid = 2

If "kernel.perf_event_paranoid" is not set to "2" or is missing, this is a finding.

Configure RHEL 9 to prevent kernel profiling by nonprivileged users.

Add or edit the following line in a system configuration file, in the "/etc/sysctl.d/" directory:

kernel.perf_event_paranoid = 2

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Changes to any software components can have significant effects on the overall security of the operating system. This requirement ensures the software has not been tampered with and that it has been provided by a trusted vendor.

Disabling kexec_load prevents an unsigned kernel image (that could be a windows kernel or modified vulnerable kernel) from being loaded. Kexec can be used subvert the entire secureboot process and should be avoided at all costs especially since it can load unsigned kernel images.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000366-GPOS-00153

Verify RHEL 9 is configured to disable kernel image loading.

Check the status of the kernel.kexec_load_disabled kernel parameter with the following command:

$ sudo sysctl kernel.kexec_load_disabled

kernel.kexec_load_disabled = 1

If "kernel.kexec_load_disabled" is not set to "1" or is missing, this is a finding.

Check that the configuration files are present to enable this kernel parameter with the following command:

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F kernel.kexec_load_disabled | tail -1

kernel.kexec_load_disabled = 1

If "kernel.kexec_load_disabled" is not set to "1" or is missing, this is a finding.

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:

kernel.kexec_load_disabled = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Exposing kernel pointers (through procfs or "seq_printf()") exposes kernel writeable structures, which may contain functions pointers. If a write vulnerability occurs in the kernel, allowing write access to any of this structure, the kernel can be compromised. This option disallows any program without the CAP_SYSLOG capability to get the addresses of kernel pointers by replacing them with "0".

Satisfies: SRG-OS-000132-GPOS-00067, SRG-OS-000433-GPOS-00192, SRG-OS-000480-GPOS-00227

Verify the runtime status of the kernel.kptr_restrict kernel parameter with the following command:

$ sudo sysctl kernel.kptr_restrict 

kernel.kptr_restrict = 1

Verify the configuration of the kernel.kptr_restrict kernel parameter with the following command:

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F kernel.kptr_restrict | tail -1

kernel.kptr_restrict =1

If "kernel.kptr_restrict" is not set to "1" or is missing, this is a finding.

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:

kernel.kptr_restrict = 1

Reload settings from all system configuration files with the following command:

$ sudo sysctl --system

By enabling the fs.protected_hardlinks kernel parameter, users can no longer create soft or hard links to files they do not own. Disallowing such hardlinks mitigates vulnerabilities based on insecure file system accessed by privileged programs, avoiding an exploitation vector exploiting unsafe use of open() or creat().

Satisfies: SRG-OS-000312-GPOS-00123, SRG-OS-000324-GPOS-00125

Verify RHEL 9 is configured to enable DAC on hardlinks.

Check the status of the fs.protected_hardlinks kernel parameter with the following command:

$ sudo sysctl fs.protected_hardlinks

fs.protected_hardlinks = 1

If "fs.protected_hardlinks" is not set to "1" or is missing, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F fs.protected_hardlinks | tail -1

fs.protected_hardlinks = 1

If "fs.protected_hardlinks" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to enable DAC on hardlinks with the following:

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:

fs.protected_hardlinks = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

By enabling the fs.protected_symlinks kernel parameter, symbolic links are permitted to be followed only when outside a sticky world-writable directory, or when the user identifier (UID) of the link and follower match, or when the directory owner matches the symlink's owner. Disallowing such symlinks helps mitigate vulnerabilities based on insecure file system accessed by privileged programs, avoiding an exploitation vector exploiting unsafe use of open() or creat().

Satisfies: SRG-OS-000312-GPOS-00123, SRG-OS-000324-GPOS-00125

Verify RHEL 9 is configured to enable DAC on symlinks.

Check the status of the fs.protected_symlinks kernel parameter with the following command:

$ sudo sysctl fs.protected_symlinks

fs.protected_symlinks = 1

If "fs.protected_symlinks " is not set to "1" or is missing, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F fs.protected_symlinks | tail -1

fs.protected_symlinks = 1

If "fs.protected_symlinks" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to enable DAC on symlinks with the following:

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:

fs.protected_symlinks = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

A core dump includes a memory image taken at the time the operating system terminates an application. The memory image could contain sensitive data and is generally useful only for developers trying to debug problems.

Verify RHEL 9 disables storing core dumps with the following commands:

$ sudo sysctl kernel.core_pattern

kernel.core_pattern = |/bin/false

If the returned line does not have a value of "|/bin/false", or a line is not returned and the need for core dumps is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Check that the configuration files are present to disable core dump storage.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F kernel.core_pattern | tail -1

kernel.core_pattern = |/bin/false

If "kernel.core_pattern" is not set to "|/bin/false" and is not documented with the ISSO as an operational requirement, or is missing, this is a finding.

Configure RHEL 9 to disable storing core dumps.

Add or edit the following line in a system configuration file, in the "/etc/sysctl.d/" directory:

kernel.core_pattern = |/bin/false

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

Disabling Asynchronous Transfer Mode (ATM) protects the system against exploitation of any flaws in its implementation.

Verify that RHEL 9 disables the ability to load the ATM kernel module with the following command:

$ sudo grep -r atm /etc/modprobe.conf /etc/modprobe.d/* 

blacklist atm

If the command does not return any output, or the line is commented out, and use of ATM is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the atm kernel module from being loaded, add the following line to the file  /etc/modprobe.d/atm.conf (or create atm.conf if it does not exist):

install atm /bin/false
blacklist atm

Disabling Controller Area Network (CAN) protects the system against exploitation of any flaws in its implementation.

Verify that RHEL 9 disables the ability to load the CAN kernel module with the following command:

$ sudo grep -r can /etc/modprobe.conf /etc/modprobe.d/* 

blacklist can

If the command does not return any output, or the line is commented out, and use of CAN is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the can kernel module from being loaded, add the following line to the file  /etc/modprobe.d/can.conf (or create atm.conf if it does not exist):

install can /bin/false
blacklist can

Disabling firewire protects the system against exploitation of any flaws in its implementation.

Verify that RHEL 9 disables the ability to load the firewire-core kernel module with the following command:

$ sudo grep -r firewire-core /etc/modprobe.conf /etc/modprobe.d/* 

blacklist firewire-core

If the command does not return any output, or the line is commented out, and use of firewire-core is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the firewire-core kernel module from being loaded, add the following line to the file /etc/modprobe.d/firewire-core.conf (or create firewire-core.conf if it does not exist):

install firewire-core /bin/false
blacklist firewire-core

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Failing to disconnect unused protocols can result in a system compromise.

The Stream Control Transmission Protocol (SCTP) is a transport layer protocol, designed to support the idea of message-oriented communication, with several streams of messages within one connection. Disabling SCTP protects the system against exploitation of any flaws in its implementation.

Verify that RHEL 9 disables the ability to load the sctp kernel module with the following command:

$ sudo grep -r sctp /etc/modprobe.conf /etc/modprobe.d/* 

blacklist sctp

If the command does not return any output, or the line is commented out, and use of sctp is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the sctp kernel module from being loaded, add the following line to the file  /etc/modprobe.d/sctp.conf (or create sctp.conf if it does not exist):

install sctp /bin/false
blacklist sctp

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Failing to disconnect unused protocols can result in a system compromise.

The Transparent Inter Process Communication (TIPC) is a protocol that is specially designed for intra-cluster communication. It can be configured to transmit messages either on UDP or directly across Ethernet. Message delivery is sequence guaranteed, loss free and flow controlled. Disabling TIPC protects the system against exploitation of any flaws in its implementation.

Verify that RHEL 9 disables the ability to load the tipc kernel module with the following command:

$ sudo grep -r tipc /etc/modprobe.conf /etc/modprobe.d/* 

blacklist tipc

If the command does not return any output, or the line is commented out, and use of tipc is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the tipc kernel module from being loaded, add the following line to the file  /etc/modprobe.d/tipc.conf (or create tipc.conf if it does not exist):

install tipc /bin/false
blacklist tipc

Address space layout randomization (ASLR) makes it more difficult for an attacker to predict the location of attack code they have introduced into a process' address space during an attempt at exploitation. Additionally, ASLR makes it more difficult for an attacker to know the location of existing code in order to repurpose it using return oriented programming (ROP) techniques.

Satisfies: SRG-OS-000433-GPOS-00193, SRG-OS-000480-GPOS-00227

Verify RHEL 9 is implementing ASLR with the following command:

$ sudo sysctl kernel.randomize_va_space

kernel.randomize_va_space = 2

Check that the configuration files are present to enable this kernel parameter.
Verify the configuration of the kernel.kptr_restrict kernel parameter with the following command:

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F kernel.randomize_va_space | tail -1

kernel.randomize_va_space = 2

If "kernel.randomize_va_space" is not set to "2" or is missing, this is a finding.

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:

kernel.randomize_va_space = 2

Reload settings from all system configuration files with the following command:

$ sudo sysctl --system

Loading and accessing the packet filters programs and maps using the bpf() system call has the potential of revealing sensitive information about the kernel state.

Satisfies: SRG-OS-000132-GPOS-00067, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 prevents privilege escalation through the kernel by disabling access to the bpf system call with the following commands:

$ sudo sysctl kernel.unprivileged_bpf_disabled

kernel.unprivileged_bpf_disabled = 1

If the returned line does not have a value of "1", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F kernel.unprivileged_bpf_disabled | tail -1
kernel.unprivileged_bpf_disabled = 1

If the network parameter "kernel.unprivileged_bpf_disabled" is not equal to "1", or nothing is returned, this is a finding.

Configure RHEL 9 to prevent privilege escalation thru the kernel by disabling access to the bpf syscall by adding the following line to a file, in the "/etc/sysctl.d" directory:

kernel.unprivileged_bpf_disabled = 1

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

Unrestricted usage of ptrace allows compromised binaries to run ptrace on other processes of the user. Like this, the attacker can steal sensitive information from the target processes (e.g., SSH sessions, web browser, etc.) without any additional assistance from the user (i.e., without resorting to phishing).

Satisfies: SRG-OS-000132-GPOS-00067, SRG-OS-000480-GPOS-00227

Verify RHEL 9 restricts usage of ptrace to descendant processes with the following commands:

$ sudo sysctl kernel.yama.ptrace_scope

kernel.yama.ptrace_scope = 1

If the returned line does not have a value of "1", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F kernel.yama.ptrace_scope| tail -1
kernel.yama.ptrace_scope = 1

If the network parameter "kernel.yama.ptrace_scope" is not equal to "1", or nothing is returned, this is a finding.

Configure RHEL 9 to restrict usage of ptrace to descendant processes by adding the following line to a file, in the "/etc/sysctl.d" directory:

kernel.yama.ptrace_scope = 1

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

A core dump includes a memory image taken at the time the operating system terminates an application. The memory image could contain sensitive data and is generally useful only for developers or system operators trying to debug problems.

Enabling core dumps on production systems is not recommended; however, there may be overriding operational requirements to enable advanced debugging. Permitting temporary enablement of core dumps during such situations must be reviewed through local needs and policy.

Verify RHEL 9 disables core dump backtraces by issuing the following command:

$ grep -i process /etc/systemd/coredump.conf

ProcessSizeMax=0

If the "ProcessSizeMax" item is missing, commented out, or the value is anything other than "0" and the need for core dumps is not documented with the information system security officer (ISSO) as an operational requirement for all domains that have the "core" item assigned, this is a finding.

Configure the operating system to disable core dump backtraces.

Add or modify the following line in /etc/systemd/coredump.conf:

ProcessSizeMax=0

A core dump includes a memory image taken at the time the operating system terminates an application. The memory image could contain sensitive data and is generally useful only for developers or system operators trying to debug problems. Enabling core dumps on production systems is not recommended; however, there may be overriding operational requirements to enable advanced debugging. Permitting temporary enablement of core dumps during such situations must be reviewed through local needs and policy.

Verify RHEL 9 disables storing core dumps for all users by issuing the following command:

$ grep -i storage /etc/systemd/coredump.conf

Storage=none

If the "Storage" item is missing, commented out, or the value is anything other than "none" and the need for core dumps is not documented with the information system security officer (ISSO) as an operational requirement for all domains that have the "core" item assigned, this is a finding.

Configure the operating system to disable storing core dumps for all users.

Add or modify the following line in /etc/systemd/coredump.conf:

Storage=none

A core dump includes a memory image taken at the time the operating system terminates an application. The memory image could contain sensitive data and is generally useful only for developers trying to debug problems.

Verify RHEL 9 disables core dumps for all users by issuing the following command:

$ grep -r -s core /etc/security/limits.conf /etc/security/limits.d/*.conf

/etc/security/limits.conf:* hard core 0

This can be set as a global domain (with the * wildcard) but may be set differently for multiple domains.

If the "core" item is missing, commented out, or the value is anything other than "0" and the need for core dumps is not documented with the information system security officer (ISSO) as an operational requirement for all domains that have the "core" item assigned, this is a finding.

Configure the operating system to disable core dumps for all users.

Add the following line to the top of the /etc/security/limits.conf or in a single ".conf" file defined in /etc/security/limits.d/:

* hard core 0

A core dump includes a memory image taken at the time the operating system terminates an application. The memory image could contain sensitive data and is generally useful only for developers trying to debug problems.

Verify RHEL 9 is not configured to acquire, save, or process core dumps with the following command:

$ sudo systemctl status systemd-coredump.socket

systemd-coredump.socket
Loaded: masked (Reason: Unit systemd-coredump.socket is masked.)
Active: inactive (dead)

If the "systemd-coredump.socket" is loaded and not masked and the need for core dumps is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Configure the system to disable the systemd-coredump.socket with the following command:

$ sudo systemctl mask --now systemd-coredump.socket

Created symlink /etc/systemd/system/systemd-coredump.socket -> /dev/null

Reload the daemon for this change to take effect.

$ sudo systemctl daemon-reload

User namespaces are used primarily for Linux containers. The value "0" disallows the use of user namespaces.

Verify RHEL 9 disables the use of user namespaces with the following commands:

$ sudo sysctl user.max_user_namespaces

user.max_user_namespaces = 0

If the returned line does not have a value of "0", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F user.max_user_namespaces | tail -1
user.max_user_namespaces = 0

If the network parameter "user.max_user_namespaces" is not equal to "0", or nothing is returned, this is a finding.

If the use of namespaces is operationally required and documented with the information system security manager (ISSM), this is not a finding.

Configure RHEL 9 to disable the use of user namespaces by adding the following line to a file, in the "/etc/sysctl.d" directory:

user.max_user_namespaces = 0

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

ExecShield uses the segmentation feature on all x86 systems to prevent execution in memory higher than a certain address. It writes an address as a limit in the code segment descriptor, to control where code can be executed, on a per-process basis. When the kernel places a process's memory regions such as the stack and heap higher than this address, the hardware prevents execution in that address range. This is enabled by default on the latest Red Hat and Fedora systems if supported by the hardware.

Verify ExecShield is enabled on 64-bit RHEL 9 systems with the following command:

$ sudo dmesg | grep '[NX|DX]*protection' 

[ 0.000000] NX (Execute Disable) protection: active

If "dmesg" does not show "NX (Execute Disable) protection" active, this is a finding.

Update the GRUB 2 bootloader configuration.

Run the following command:

$ sudo grubby --update-kernel=ALL --remove-args=noexec

Kernel core dumps may contain the full contents of system memory at the time of the crash. Kernel core dumps consume a considerable amount of disk space and may result in denial of service by exhausting the available space on the target file system partition. Unless the system is used for kernel development or testing, there is little need to run the kdump service.

Verify that the kdump service is disabled in system boot configuration with the following command:

$ systemctl is-enabled  kdump  

disabled 

Verify that the kdump service is not active (i.e., not running) through current runtime configuration with the following command:

$ systemctl is-active kdump 

inactive 

Verify that the kdump service is masked with the following command:

$ sudo systemctl show  kdump  | grep "LoadState\|UnitFileState" 

LoadState=masked 

UnitFileState=masked 

If the "kdump" service is loaded or active, and is not masked, this is a finding.

Disable and mask the kdump service on RHEL 9.

To disable the kdump service run the following command:

$ sudo systemctl disable --now kdump

To mask the kdump service run the following command:

$ sudo systemctl mask --now kdump

Cryptographic verification of vendor software packages ensures that all software packages are obtained from a valid source and protects against spoofing that could lead to installation of malware on the system. Red Hat cryptographically signs all software packages, which includes updates, with a GPG key to verify that they are valid.

Confirm Red Hat package-signing keys are installed on the system and verify their fingerprints match vendor values.

Note: For RHEL 9 software packages, Red Hat uses GPG keys labeled "release key 2" and "auxiliary key 3". The keys are defined in key file "/etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release" by default.

List Red Hat GPG keys installed on the system:

$ sudo rpm -q --queryformat "%{SUMMARY}\n" gpg-pubkey | grep -i "red hat"

Red Hat, Inc. (release key 2)  public key
Red Hat, Inc. (auxiliary key 3)  public key

If Red Hat GPG keys "release key 2" and "auxiliary key 3" are not installed, this is a finding.

List key fingerprints of installed Red Hat GPG keys:

$ sudo gpg -q --keyid-format short --with-fingerprint /etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release

If key file "/etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release" is missing, this is a finding.

Example output:

pub   rsa4096/FD431D51 2009-10-22 [SC]
      Key fingerprint = 567E 347A D004 4ADE 55BA  8A5F 199E 2F91 FD43 1D51
uid                   Red Hat, Inc. (release key 2) 
pub   rsa4096/5A6340B3 2022-03-09 [SC]
      Key fingerprint = 7E46 2425 8C40 6535 D56D  6F13 5054 E4A4 5A63 40B3
uid                   Red Hat, Inc. (auxiliary key 3) 

Compare key fingerprints of installed Red Hat GPG keys with fingerprints listed for RHEL 9 on Red Hat "Product Signing Keys" webpage at https://access.redhat.com/security/team/key.

If key fingerprints do not match, this is a finding.

Install Red Hat package-signing keys on the system and verify their fingerprints match vendor values.

Insert RHEL 9 installation disc or attach RHEL 9 installation image to the system. Mount the disc or image to make the contents accessible inside the system.

Assuming the mounted location is "/media/cdrom", use the following command to copy Red Hat GPG key file onto the system:

$ sudo cp /media/cdrom/RPM-GPG-KEY-redhat-release /etc/pki/rpm-gpg/
	 
Import Red Hat GPG keys from key file into system keyring:

$ sudo rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release
	 
Using the steps listed in the Check Text, confirm the newly imported keys show as installed on the system and verify their fingerprints match vendor values.

Changes to any software components can have significant effects on the overall security of the operating system. This requirement ensures the software has not been tampered with and that it has been provided by a trusted vendor.

All software packages must be signed with a cryptographic key recognized and approved by the organization.

Verifying the authenticity of software prior to installation validates the integrity of the software package received from a vendor. This verifies the software has not been tampered with and that it has been provided by a trusted vendor.

Verify that dnf always checks the GPG signature of software packages originating from external software repositories before installation:

$ grep gpgcheck /etc/dnf/dnf.conf

gpgcheck=1

If "gpgcheck" is not set to "1", or if the option is missing or commented out, ask the system administrator how the GPG signatures of software packages are being verified.

If there is no process to verify GPG signatures that is approved by the organization, this is a finding.

Configure dnf to always check the GPG signature of software packages originating from external software repositories before installation.

Add or update the following line in the [main] section of the /etc/dnf/dnf.conf file:

gpgcheck=1

Changes to any software components can have significant effects on the overall security of the operating system. This requirement ensures the software has not been tampered with and that it has been provided by a trusted vendor.

All software packages must be signed with a cryptographic key recognized and approved by the organization.

Verifying the authenticity of software prior to installation validates the integrity of the software package received from a vendor. This verifies the software has not been tampered with and that it has been provided by a trusted vendor.

Verify that dnf always checks the GPG signature of locally installed software packages before installation:

$ grep localpkg_gpgcheck /etc/dnf/dnf.conf 

localpkg_gpgcheck=1 

If "localpkg_gpgcheck" is not set to "1", or if the option is missing or commented out, ask the system administrator how the GPG signatures of local software packages are being verified.

If there is no process to verify GPG signatures that is approved by the organization, this is a finding.

Configure dnf to always check the GPG signature of local software packages before installation.

Add or update the following line in the [main] section of the /etc/dnf/dnf.conf file:

localpkg_gpgcheck=1

Changes to any software components can have significant effects on the overall security of the operating system. This requirement ensures the software has not been tampered with and that it has been provided by a trusted vendor.

All software packages must be signed with a cryptographic key recognized and approved by the organization.

Verifying the authenticity of software prior to installation validates the integrity of the software package received from a vendor. This verifies the software has not been tampered with and that it has been provided by a trusted vendor.

Verify that all software repositories defined in "/etc/yum.repos.d/" have been configured with "gpgcheck" enabled:

$ grep gpgcheck /etc/yum.repos.d/*.repo | more

gpgcheck = 1

If "gpgcheck" is not set to "1" for all returned lines, this is a finding.

Configure all software repositories defined in "/etc/yum.repos.d/" to have "gpgcheck" enabled:

$ sudo sed -i 's/gpgcheck\s*=.*/gpgcheck=1/g' /etc/yum.repos.d/*

The hashes of important files like system executables should match the information given by the RPM database. Executables with erroneous hashes could be a sign of nefarious activity on the system.

The following command will list which files on the system have file hashes different from what is expected by the RPM database:

 $ rpm -Va --noconfig | awk '$1 ~ /..5/ && $2 != "c"' 

If there is output, this is a finding.

Given output from the check command, identify the package that provides the output and reinstall it. The following trimmed example output shows a package that has failed verification, been identified, and been reinstalled:

$ rpm -Va --noconfig | awk '$1 ~ /..5/ && $2 != "c"'
S.5....T.    /usr/bin/znew
$ sudo dnf provides /usr/bin/znew
[...]
gzip-1.10-8.el9.x86_64 : The GNU data compression program
[...]
$ sudo dnf reinstall gzip
[...]
$ rpm -Va --noconfig | awk '$1 ~ /..5/ && $2 != "c"'
[no output]

Previous versions of software components that are not removed from the information system after updates have been installed may be exploited by some adversaries.

Verify RHEL 9 removes all software components after updated versions have been installed with the following command:

$ grep clean /etc/dnf/dnf.conf 

clean_requirements_on_remove=1 

If "clean_requirements_on_remove" is not set to "1", this is a finding.

Configure RHEL 9 to remove all software components after updated versions have been installed.

Edit the file /etc/dnf/dnf.conf by adding or editing the following line:

 clean_requirements_on_remove=1

The Red Hat Subscription Manager application manages software subscriptions and software repositories for installed software products on the local system. It communicates with backend servers, such as the Red Hat Customer Portal or an on-premise instance of Subscription Asset Manager, to register the local system and grant access to software resources determined by the subscription entitlement.

Verify that RHEL 9 subscription-manager package is installed with the following command:

$ sudo dnf list --installed subscription-manager

Example output:

subscription-manager.x86_64          1.29.26-3.el9_0

If the "subscription-manager" package is not installed, this is a finding.

The  subscription-manager package can be installed with the following command:
 
$ sudo dnf install subscription-manager

The FTP service provides an unencrypted remote access that does not provide for the confidentiality and integrity of user passwords or the remote session. If a privileged user were to log on using this service, the privileged user password could be compromised. SSH or other encrypted file transfer methods must be used in place of this service.

Removing the "vsftpd" package decreases the risk of accidental activation.

Satisfies: SRG-OS-000074-GPOS-00042, SRG-OS-000095-GPOS-00049, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 does not have a File Transfer Protocol (FTP) server package installed with the following command:

$ sudo dnf list --installed | grep ftp 

If the "ftp" package is installed, this is a finding.

The ftp package can be removed with the following command (using vsftpd as an example):

$ sudo dnf remove vsftpd

The sendmail software was not developed with security in mind, and its design prevents it from being effectively contained by SELinux. Postfix must be used instead.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000095-GPOS-00049

Verify that the sendmail package is not installed with the following command:

$ sudo dnf list --installed sendmail

Error: No matching Packages to list

If the "sendmail" package is installed, this is a finding.

Remove the sendmail package with the following command:

$ sudo dnf remove sendmail

"nfs-utils" provides a daemon for the kernel NFS server and related tools. This package also contains the "showmount" program. "showmount" queries the mount daemon on a remote host for information about the Network File System (NFS) server on the remote host. For example, "showmount" can display the clients that are mounted on that host.

Verify that the nfs-utils package is not installed with the following command:

$ sudo dnf list --installed nfs-utils

Error: No matching Packages to list

If the "nfs-utils" package is installed, this is a finding.

Remove the nfs-utils package with the following command:

$ sudo dnf remove nfs-utils

The NIS service provides an unencrypted authentication service, which does not provide for the confidentiality and integrity of user passwords or the remote session.

Removing the "ypserv" package decreases the risk of the accidental (or intentional) activation of NIS or NIS+ services.

Verify that the ypserv package is not installed with the following command:

$ sudo dnf list --installed ypserv

Error: No matching Packages to list

If the "ypserv" package is installed, this is a finding.

Remove the ypserv package with the following command:

$ sudo dnf remove ypserv

The "rsh-server" service provides unencrypted remote access service, which does not provide for the confidentiality and integrity of user passwords or the remote session and has very weak authentication. If a privileged user were to login using this service, the privileged user password could be compromised. The "rsh-server" package provides several obsolete and insecure network services. Removing it decreases the risk of accidental (or intentional) activation of those services.

Verify that the rsh-server package is not installed with the following command:

$ sudo dnf list --installed rsh-server

Error: No matching Packages to list

If the "rsh-server" package is installed, this is a finding.

Remove the rsh-server package with the following command:

$ sudo dnf remove rsh-server

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities are often overlooked and therefore, may remain unsecure. They increase the risk to the platform by providing additional attack vectors.

The telnet service provides an unencrypted remote access service, which does not provide for the confidentiality and integrity of user passwords or the remote session. If a privileged user were to login using this service, the privileged user password could be compromised.

Removing the "telnet-server" package decreases the risk of accidental (or intentional) activation of the telnet service.

Verify that the telnet-server package is not installed with the following command:

$ sudo dnf list --installed telnet-server

Error: No matching Packages to list

If the "telnet-server" package is installed, this is a finding.

Remove the telnet-server package with the following command:

$ sudo dnf remove telnet-server

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore, may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Operating systems are capable of providing a wide variety of functions and services. Some of the functions and services provided by default may not be necessary to support essential organizational operations (e.g., key missions, functions).

The gssproxy package is a proxy for GSS API credential handling and could expose secrets on some networks. It is not needed for normal function of the OS.

Satisfies: SRG-OS-000095-GPOS-00049, SRG-OS-000480-GPOS-00227

Verify that the gssproxy package is not installed with the following command:

$ sudo dnf list --installed gssproxy

Error: No matching Packages to list

If the "gssproxy" package is installed and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Remove the gssproxy package with the following command:

$ sudo dnf remove gssproxy

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Operating systems are capable of providing a wide variety of functions and services. Some of the functions and services, provided by default, may not be necessary to support essential organizational operations (e.g., key missions, functions).

The iprutils package provides a suite of utilities to manage and configure SCSI devices supported by the ipr SCSI storage device driver.

Satisfies: SRG-OS-000095-GPOS-00049, SRG-OS-000480-GPOS-00227

Verify that the iprutils package is not installed with the following command:

$ sudo dnf list --installed iprutils

Error: No matching Packages to list

If the "iprutils" package is installed and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Remove the iprutils package with the following command:

$ sudo dnf remove iprutils

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Operating systems are capable of providing a wide variety of functions and services. Some of the functions and services, provided by default, may not be necessary to support essential organizational operations (e.g., key missions, functions).

The tuned package contains a daemon that tunes the system settings dynamically. It does so by monitoring the usage of several system components periodically. Based on that information, components will then be put into lower or higher power savings modes to adapt to the current usage. The tuned package is not needed for normal OS operations.

Satisfies: SRG-OS-000095-GPOS-00049, SRG-OS-000480-GPOS-00227

Verify that the tuned package is not installed with the following command:

$ sudo dnf list --installed tuned

Error: No matching Packages to list

If the "tuned" package is installed and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Remove the tuned package with the following command:

$ sudo dnf remove tuned

Removing the "tftp-server" package decreases the risk of the accidental (or intentional) activation of tftp services.

If TFTP is required for operational support (such as transmission of router configurations), its use must be documented with the information systems security manager (ISSM), restricted to only authorized personnel, and have access control rules established.

Verify that RHEL 9 does not have a "tftp-server" package installed with the following command:

$ sudo dnf list --installed | grep tftp-server

If the "tftp-server" package is installed, this is a finding.

The "tftp-server" package can be removed with the following command:

$ sudo dnf remove tftp-server

Quagga is a network routing software suite providing implementations of Open Shortest Path First (OSPF), Routing Information Protocol (RIP), Border Gateway Protocol (BGP) for Unix and Linux platforms.

If there is no need to make the router software available, removing it provides a safeguard against its activation.

Verify that the quagga package is not installed with the following command:

$ sudo dnf list --installed quagga

Error: No matching Packages to list

If the "quagga" package is installed, and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Remove the quagga package with the following command:

$ sudo dnf remove quagga

Unnecessary service packages must not be installed to decrease the attack surface of the system. Graphical display managers have a long history of security vulnerabilities and must not be used, unless approved and documented.

Verify that a graphical user interface is not installed with the following command:

$ sudo dnf list --installed "xorg*common"

Error: No matching Packages to list

If the "x11-server-common" package is installed, and the use of a graphical user interface has not been documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Document the requirement for a graphical user interface with the ISSO or remove all xorg packages with the following command:

Warning: If you are accessing the system through the graphical user interface, change to the multi-user.target with the following command:

$ sudo systemctl isolate multi-user.target

Warning: Removal of the graphical user interface will immediately render it useless. The following commands must not be run from a virtual terminal emulator in the graphical interface.

$ sudo dnf remove "xorg*"
$ sudo systemctl set-default multi-user.target

Without the use of multifactor authentication, the ease of access to privileged functions is greatly increased. Multifactor authentication requires using two or more factors to achieve authentication. A privileged account is defined as an information system account with authorizations of a privileged user. The DOD common access card (CAC) with DOD-approved PKI is an example of multifactor authentication.

Satisfies: SRG-OS-000105-GPOS-00052, SRG-OS-000375-GPOS-00160, SRG-OS-000376-GPOS-00161, SRG-OS-000377-GPOS-00162

Verify that RHEL 9 has the openssl-pkcs11 package installed with the following command:

$ sudo dnf list --installed openssl-pkcs11

Example output:

openssl-pkcs.i686          0.4.11-7.el9
openssl-pkcs.x86_64          0.4.11-7.el9

If the "openssl-pkcs11" package is not installed, this is a finding.

Note: If the system administrator demonstrates the use of an approved alternate multifactor authentication method, this requirement is not applicable.

The openssl-pkcs11 package can be installed with the following command:
 
$ sudo dnf install openssl-pkcs11

GnuTLS is a secure communications library implementing the SSL, TLS and DTLS protocols and technologies around them. It provides a simple C language application programming interface (API) to access the secure communications protocols as well as APIs to parse and write X.509, PKCS #12, OpenPGP and other required structures. This package contains command line TLS client and server and certificate manipulation tools.

Verify that RHEL 9 has the gnutls-utils package installed with the following command:

$ dnf list --installed gnutls-utils

Example output:

gnutls-utils.x86_64          3.7.3-9.el9

If the "gnutls-utils" package is not installed, this is a finding.

The gnutls-utils package can be installed with the following command:
 
$ sudo dnf install gnutls-utils

Network Security Services (NSS) is a set of libraries designed to support cross-platform development of security-enabled client and server applications. Install the "nss-tools" package to install command-line tools to manipulate the NSS certificate and key database.

Verify that RHEL 9 has the nss-tools package installed with the following command:

$ dnf list --installed nss-tools

Example output:

nss-tools.x86_64          3.71.0-7.el9

If the "nss-tools" package is not installed, this is a finding.

The nss-tools package can be installed with the following command:
 
$ sudo dnf install nss-tools

"rng-tools" provides hardware random number generator tools, such as those used in the formation of x509/PKI certificates.

Verify that RHEL 9 has the rng-tools package installed with the following command:

$ sudo dnf list --installed rng-tools

Example output:

rng-tools.x86_64          6.14-2.git.b2b7934e.el9

If the "rng-tools" package is not installed, this is a finding.

The rng-tools package can be installed with the following command:
 
$ sudo dnf install rng-tools

The "s-nail" package provides the mail command required to allow sending email notifications of unauthorized configuration changes to designated personnel.

Verify that RHEL 9 is configured to allow sending email notifications.

Note: The "s-nail" package provides the "mail" command that is used to send email messages.

Verify that the "s-nail" package is installed on the system:

$ sudo dnf list --installed s-nail

s-nail.x86_64          14.9.22-6.el9
	 
If "s-nail" package is not installed, this is a finding.

The s-nail package can be installed with the following command:

$ sudo dnf install s-nail

Ensuring that "/home" is mounted on its own partition enables the setting of more restrictive mount options, and also helps ensure that users cannot trivially fill partitions used for log or audit data storage.

Verify that a separate file system/partition has been created for "/home" with the following command:

$ mount | grep /home

UUID=fba5000f-2ffa-4417-90eb-8c54ae74a32f on /home type ext4 (rw,nodev,nosuid,noexec,seclabel)

If a separate entry for "/home" is not in use, this is a finding.

Migrate the "/home" directory onto a separate file system/partition.

The "/tmp" partition is used as temporary storage by many programs. Placing "/tmp" in its own partition enables the setting of more restrictive mount options, which can help protect programs that use it.

Verify that a separate file system/partition has been created for "/tmp" with the following command:

$ mount | grep /tmp 

tmpfs /tmp tmpfs noatime,mode=1777 0 0

If a separate entry for "/tmp" is not in use, this is a finding.

Migrate the "/tmp" path onto a separate file system.

Ensuring that "/var" is mounted on its own partition enables the setting of more restrictive mount options. This helps protect system services such as daemons or other programs which use it. It is not uncommon for the "/var" directory to contain world-writable directories installed by other software packages.

Verify that a separate file system/partition has been created for "/var" with the following command:

$ mount | grep /var

UUID=c274f65f-c5b5-4481-b007-bee96feb8b05 /var xfs noatime 1 2

If a separate entry for "/var" is not in use, this is a finding.

Migrate the "/var" path onto a separate file system.

Placing "/var/log" in its own partition enables better separation between log files and other files in "/var/".

Verify that a separate file system/partition has been created for "/var/log" with the following command:

$ mount | grep /var/log

UUID=c274f65f-c5b5-4486-b021-bee96feb8b21 /var/log xfs noatime 1 2

If a separate entry for "/var/log" is not in use, this is a finding.

Migrate the "/var/log" path onto a separate file system.

Placing "/var/log/audit" in its own partition enables better separation between audit files and other system files, and helps ensure that auditing cannot be halted due to the partition running out of space.

Satisfies: SRG-OS-000341-GPOS-00132, SRG-OS-000480-GPOS-00227

Verify that a separate file system/partition has been created for the system audit data path with the following command:

Note: /var/log/audit is used as the example as it is a common location.

$ mount | grep /var/log/audit 

UUID=2efb2979-45ac-82d7-0ae632d11f51 on /var/log/home type xfs  (rw,realtime,seclabel,attr2,inode64)

If no line is returned, this is a finding.

Migrate the system audit data path onto a separate file system.

The "/var/tmp" partition is used as temporary storage by many programs. Placing "/var/tmp" in its own partition enables the setting of more restrictive mount options, which can help protect programs that use it.

Verify that a separate file system/partition has been created for "/var/tmp" with the following command:

$ mount | grep /var/tmp

UUID=c274f65f-c5b5-4379-b017-bee96feb7a34 /var/log xfs noatime 1 2

If a separate entry for "/var/tmp" is not in use, this is a finding.

Migrate the "/var/tmp" path onto a separate file system.

An authentication process resists replay attacks if it is impractical to achieve a successful authentication by recording and replaying a previous authentication message.

Satisfies: SRG-OS-000114-GPOS-00059, SRG-OS-000378-GPOS-00163, SRG-OS-000480-GPOS-00227

Note: If the autofs service is not installed, this requirement is not applicable.

Verify that RHEL 9 file system automount function has been disabled with the following command:

$ sudo systemctl is-enabled  autofs

masked

If the returned value is not "masked", "disabled", or "Failed to get unit file state for autofs.service for autofs", and is not documented as operational requirement with the information system security officer ISSO, this is a finding.

Configure RHEL 9 to disable the ability to automount devices.

The autofs service can be disabled with the following command:

$ sudo systemctl mask --now autofs.service

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/home" is mounted with the "nodev" option with the following command:

Note: If a separate file system has not been created for the user home directories (user home directories are mounted under "/"), this is automatically a finding, as the "nodev" option cannot be used on the "/" system.

$ mount | grep /home

tmpfs on /home type tmpfs (rw,nodev,nosuid,noexec,seclabel)

If the "/home" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/home" directory.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Satisfies: SRG-OS-000368-GPOS-00154, SRG-OS-000480-GPOS-00227

Verify "/home" is mounted with the "nosuid" option with the following command:

Note: If a separate file system has not been created for the user home directories (user home directories are mounted under "/"), this is automatically a finding, as the "nosuid" option cannot be used on the "/" system.

$ mount | grep /home

tmpfs on /home type tmpfs (rw,nodev,nosuid,noexec,seclabel)

If the "/home" file system is mounted without the "nosuid" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/home" directory.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/home" is mounted with the "noexec" option with the following command:

Note: If a separate file system has not been created for the user home directories (user home directories are mounted under "/"), this is automatically a finding, as the "noexec" option cannot be used on the "/" system.

$ mount | grep /home

tmpfs on /home type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/home" file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/home" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify RHEL 9 has the "nodev" option configured for all NFS mounts with the following command:

$ cat /etc/fstab | grep nfs

192.168.22.2:/mnt/export /data nfs4 rw,nosuid,nodev,noexec,sync,soft,sec=krb5:krb5i:krb5p

Note: If no NFS mounts are configured, this requirement is Not Applicable.

If the system is mounting file systems via NFS and the "nodev" option is missing, this is a finding.

Update each NFS mounted file system to use the "nodev" option on file systems that are being imported via NFS.

The "noexec" mount option causes the system not to execute binary files. This option must be used for mounting any file system not containing approved binary as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify RHEL 9 has the "noexec" option configured for all NFS mounts with the following command:

$ cat /etc/fstab | grep nfs

192.168.22.2:/mnt/export /data nfs4 rw,nosuid,nodev,noexec,sync,soft,sec=krb5:krb5i:krb5p

If no NFS mounts are configured, this requirement is Not Applicable.

If the system is mounting file systems via NFS and the "noexec" option is missing, this is a finding.

Update each NFS mounted file system to use the "noexec" option on file systems that are being imported via NFS.

The "nosuid" mount option causes the system not to execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify RHEL 9 has the "nosuid" option configured for all NFS mounts with the following command:

Note: If no NFS mounts are configured, this requirement is Not Applicable.

$ cat /etc/fstab | grep nfs

192.168.22.2:/mnt/export /data nfs4 rw,nosuid,nodev,noexec,sync,soft,sec=krb5:krb5i:krb5p

If the system is mounting file systems via NFS and the "nosuid" option is missing, this is a finding.

Update each NFS mounted file system to use the "nosuid" option on file systems that are being imported via NFS.

The "noexec" mount option causes the system not to execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify file systems that are used for removable media are mounted with the "noexec" option with the following command:

$ more /etc/fstab

UUID=2bc871e4-e2a3-4f29-9ece-3be60c835222 /mnt/usbflash vfat noauto,owner,ro,nosuid,nodev,noexec 0 0

If a file system found in "/etc/fstab" refers to removable media and it does not have the "noexec" option set, this is a finding.

Configure the "/etc/fstab" to use the "noexec" option on file systems that are associated with removable media.

The "nodev" mount option causes the system not to interpret character or block special devices. Executing character or blocking special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify file systems that are used for removable media are mounted with the "nodev" option with the following command:

$ more /etc/fstab

UUID=2bc871e4-e2a3-4f29-9ece-3be60c835222 /mnt/usbflash vfat noauto,owner,ro,nosuid,nodev,noexec 0 0

If a file system found in "/etc/fstab" refers to removable media and it does not have the "nodev" option set, this is a finding.

Configure the "/etc/fstab" to use the "nodev" option on file systems that are associated with removable media.

The "nosuid" mount option causes the system not to execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify file systems that are used for removable media are mounted with the "nosuid" option with the following command:

$ more /etc/fstab

UUID=2bc871e4-e2a3-4f29-9ece-3be60c835222 /mnt/usbflash vfat noauto,owner,ro,nosuid,nodev,noexec 0 0

If a file system found in "/etc/fstab" refers to removable media and it does not have the "nosuid" option set, this is a finding.

Configure the "/etc/fstab" to use the "nosuid" option on file systems that are associated with removable media.

The only legitimate location for device files is the "/dev" directory located on the root partition. The only exception to this is chroot jails.

Verify that the "/boot" mount point has the "nodev" option is with the following command:
  
$ sudo mount | grep '\s/boot\s'

/dev/sda1 on /boot type xfs (rw,nodev,relatime,seclabel,attr2)

If the "/boot" file system does not have the "nodev" option set, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/boot" directory.

The "nosuid" mount option causes the system not to execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Satisfies: SRG-OS-000368-GPOS-00154, SRG-OS-000480-GPOS-00227

Verify the /boot directory is mounted with the "nosuid" option with the following command:

$ mount | grep '\s/boot\s'

/dev/sda1 on /boot type xfs (rw,nosuid,relatime,seclabe,attr2,inode64,noquota)

If the /boot file system does not have the "nosuid" option set, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/boot" directory.

The "nosuid" mount option causes the system not to execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Satisfies: SRG-OS-000368-GPOS-00154, SRG-OS-000480-GPOS-00227

Note: For systems that use BIOS, this requirement is Not Applicable.

Verify the /boot/efi directory is mounted with the "nosuid" option with the following command:

$ mount | grep '\s/boot/efi\s'

/dev/sda1 on /boot/efi type vfat (rw,nosuid,relatime,fmask=0077,dmask=0077,codepage=437,iocharset=ascii,shortname=winnt,errors=remount-ro)

If the /boot/efi file system does not have the "nosuid" option set, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/boot/efi" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/dev/shm" is mounted with the "nodev" option with the following command:

$ mount | grep /dev/shm

tmpfs on /dev/shm type tmpfs (rw,nodev,nosuid,noexec,seclabel)

If the /dev/shm file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/dev/shm" file system.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/dev/shm" is mounted with the "noexec" option with the following command:

$ mount | grep /dev/shm

tmpfs on /dev/shm type tmpfs (rw,nodev,nosuid,noexec,seclabel)

If the /dev/shm file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/dev/shm" file system.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/dev/shm" is mounted with the "nosuid" option with the following command:

$ mount | grep /dev/shm

tmpfs on /dev/shm type tmpfs (rw,nodev,nosuid,noexec,seclabel)

If the /dev/shm file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/dev/shm" file system.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/tmp" is mounted with the "nodev" option:

$ mount | grep /tmp

/dev/mapper/rhel-tmp on /tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/tmp" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/tmp" directory.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/tmp" is mounted with the "noexec" option:

$ mount | grep /tmp

/dev/mapper/rhel-tmp on /tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/tmp" file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/tmp" directory.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/tmp" is mounted with the "nosuid" option:

$ mount | grep /tmp

/dev/mapper/rhel-tmp on /tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/tmp" file system is mounted without the "nosuid" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/tmp" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/var" is mounted with the "nodev" option:

$ mount | grep /var

/dev/mapper/rhel-var on /var type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/var" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/var/log" is mounted with the "nodev" option:

$ mount | grep /var/log

/dev/mapper/rhel-var-log on /var/log type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/var/log" directory.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/log" is mounted with the "noexec" option:

$ mount | grep /var/log

/dev/mapper/rhel-var-log on /var/log type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log" file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/var/log" directory.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/log" is mounted with the "nosuid" option:

$ mount | grep /var/log

/dev/mapper/rhel-var-log on /var/log type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log" file system is mounted without the "nosuid" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/var/log" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/var/log/audit" is mounted with the "nodev" option:

$ mount | grep /var/log/audit

/dev/mapper/rhel-var-log-audit on /var/log/audit type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log/audit" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/var/log/audit" directory.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/log/audit" is mounted with the "noexec" option:

$ mount | grep /var/log/audit

/dev/mapper/rhel-var-log-audit on /var/log/audit type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log/audit" file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/var/log/audit" directory.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/log/audit" is mounted with the "nosuid" option:

$ mount | grep /var/log/audit

/dev/mapper/rhel-var-log-audit on /var/log/audit type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/log/audit" file system is mounted without the "nosuid" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/var/log/audit" directory.

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify "/var/tmp" is mounted with the "nodev" option:

$ mount | grep /var/tmp

/dev/mapper/rhel-var-tmp on /var/tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/tmp" file system is mounted without the "nodev" option, this is a finding.

Modify "/etc/fstab" to use the "nodev" option on the "/var/tmp" directory.

The "noexec" mount option causes the system to not execute binary files. This option must be used for mounting any file system not containing approved binary files, as they may be incompatible. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/tmp" is mounted with the "noexec" option:

$ mount | grep /var/tmp

/dev/mapper/rhel-var-tmp on /var/tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/tmp" file system is mounted without the "noexec" option, this is a finding.

Modify "/etc/fstab" to use the "noexec" option on the "/var/tmp" directory.

The "nosuid" mount option causes the system to not execute "setuid" and "setgid" files with owner privileges. This option must be used for mounting any file system not containing approved "setuid" and "setguid" files. Executing files from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

Verify "/var/tmp" is mounted with the "nosuid" option:

$ mount | grep /var/tmp

/dev/mapper/rhel-var-tmp on /var/tmp type xfs (rw,nodev,nosuid,noexec,seclabel)

If the "/var/tmp" file system is mounted without the "nosuid" option, this is a finding.

Modify "/etc/fstab" to use the "nosuid" option on the "/var/tmp" directory.

RHEL 9 systems handling data requiring "data at rest" protections must employ cryptographic mechanisms to prevent unauthorized disclosure and modification of the information at rest.

Selection of a cryptographic mechanism is based on the need to protect the integrity of organizational information. The strength of the mechanism is commensurate with the security category and/or classification of the information. Organizations have the flexibility to either encrypt all information on storage devices (i.e., full disk encryption) or encrypt specific data structures (e.g., files, records, or fields).

Satisfies: SRG-OS-000405-GPOS-00184, SRG-OS-000185-GPOS-00079, SRG-OS-000404-GPOS-00183

Note: If there is a documented and approved reason for not having data-at-rest encryption at the operating system level, such as encryption provided by a hypervisor or a disk storage array in a virtualized environment, this requirement is not applicable.

Verify RHEL 9 prevents unauthorized disclosure or modification of all information requiring at-rest protection by using disk encryption. 

Note: If there is a documented and approved reason for not having data-at-rest encryption, this requirement is Not Applicable.

Verify all system partitions are encrypted with the following command:

$ blkid

/dev/map per/rhel-root:  UUID="67b7d7fe-de60-6fd0-befb-e6748cf97743" TYPE="crypto_LUKS"

Every persistent disk partition present must be of type "crypto_LUKS". If any partitions other than the boot partition or pseudo file systems (such as /proc or /sys) or temporary file systems (that are tmpfs) are not type "crypto_LUKS", ask the administrator to indicate how the partitions are encrypted. If there is no evidence that these partitions are encrypted, this is a finding.

Configure RHEL 9 to prevent unauthorized modification of all information at rest by using disk encryption.

Encrypting a partition in an already installed system is more difficult, because existing partitions will need to be resized and changed.

To encrypt an entire partition, dedicate a partition for encryption in the partition layout.

It is detrimental for operating systems to provide, or install by default, functionality exceeding requirements or mission objectives. These unnecessary capabilities or services are often overlooked and therefore may remain unsecured. They increase the risk to the platform by providing additional attack vectors.

Removing support for unneeded filesystem types reduces the local attack surface of the server.

Compressed ROM/RAM file system (or cramfs) is a read-only file system designed for simplicity and space-efficiency. It is mainly used in embedded and small-footprint systems.

Verify that RHEL 9 disables the ability to load the cramfs kernel module with the following command:

$ sudo grep -r cramfs /etc/modprobe.conf /etc/modprobe.d/* 

blacklist cramfs

If the command does not return any output, or the line is commented out, and use of cramfs is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the cramfs kernel module from being loaded, add the following line to the file /etc/modprobe.d/blacklist.conf (or create blacklist.conf if it does not exist):

install cramfs /bin/false
blacklist cramfs

The "nodev" mount option causes the system to not interpret character or block special devices. Executing character or block special devices from untrusted file systems increases the opportunity for nonprivileged users to attain unauthorized administrative access.

The only legitimate location for device files is the "/dev" directory located on the root partition, with the exception of chroot jails if implemented.

Verify all non-root local partitions are mounted with the "nodev" option with the following command:

$ sudo mount | grep '^/dev\S* on /\S' | grep --invert-match 'nodev'

If any output is produced, this is a finding.

Configure the "/etc/fstab" to use the "nodev" option on all non-root local partitions.

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system commands contained in the following directories have mode "755" or less permissive with the following command:

$ sudo find -L /bin /sbin /usr/bin /usr/sbin /usr/libexec /usr/local/bin /usr/local/sbin -perm /022 -exec ls -l {} \;

If any system commands are found to be group-writable or world-writable, this is a finding.

Configure the system commands to be protected from unauthorized access.

Run the following command, replacing "[FILE]" with any system command with a mode more permissive than "755".

$ sudo chmod 755 [FILE]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library directories have mode "755" or less permissive with the following command:

$ sudo find -L /lib /lib64 /usr/lib /usr/lib64 -perm /022 -type d -exec ls -l {} \;

If any system-wide shared library file is found to be group-writable or world-writable, this is a finding.

Configure the system-wide shared library directories (/lib, /lib64, /usr/lib and /usr/lib64) to be protected from unauthorized access. 

Run the following command, replacing "[DIRECTORY]" with any library directory with a mode more permissive than 755.

$ sudo chmod 755 [DIRECTORY]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library files contained in the following directories have mode "755" or less permissive with the following command:

$ sudo find -L /lib /lib64 /usr/lib /usr/lib64 -perm /022 -type f -exec ls -l {} \;

If any system-wide shared library file is found to be group-writable or world-writable, this is a finding.

Configure the library files to be protected from unauthorized access. Run the following command, replacing "[FILE]" with any library file with a mode more permissive than 755.

$ sudo chmod 755 [FILE]

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify that the "/var/log" directory has a mode of "0755" or less permissive with the following command:

$ ls -ld /var/log

drwxr-xr-x. 16 root root 4096 July 11 11:34 /var/log

If "/var/log" does not have a mode of "0755" or less permissive, this is a finding.

Configure the "/var/log" directory to a mode of "0755" by running the following command:

$ sudo chmod 0755 /var/log

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify the "/var/log/messages" file has a mode of "0640" or less permissive with the following command:

$ ls -la /var/log/messages

rw-------. 1 root root 564223 July 11 11:34 /var/log/messages

If "/var/log/messages" does not have a mode of "0640" or less permissive, this is a finding.

Configure the "/var/log/messages" file to have a mode of "0640" by running the following command:

$ sudo chmod 0640 /var/log/messages

Protecting audit information also includes identifying and protecting the tools used to view and manipulate log data. Therefore, protecting audit tools is necessary to prevent unauthorized operation on audit information.

RHEL 9 systems providing tools to interface with audit information will leverage user permissions and roles identifying the user accessing the tools, and the corresponding rights the user enjoys, to make access decisions regarding the access to audit tools.

Audit tools include, but are not limited to, vendor-provided and open source audit tools needed to successfully view and manipulate audit information system activity and records. Audit tools include custom queries and report generators.

Verify the audit tools have a mode of "0755" or less with the following command:

$ stat -c "%a %n" /sbin/auditctl /sbin/aureport /sbin/ausearch /sbin/autrace /sbin/auditd /sbin/rsyslogd /sbin/augenrules

755 /sbin/auditctl
755 /sbin/aureport
755 /sbin/ausearch
750 /sbin/autrace
755 /sbin/auditd
755 /sbin/rsyslogd
755 /sbin/augenrules

If any of the audit tool files have a mode more permissive than "0755", this is a finding.

Configure the audit tools to have a mode of "0755" by running the following command:

$ sudo chmod 0755 [audit_tool]

Replace "[audit_tool]" with each audit tool that has a more permissive mode than 0755.

Service configuration files enable or disable features of their respective services that if configured incorrectly can lead to insecure and vulnerable configurations. Therefore, service configuration files should have the correct access rights to prevent unauthorized changes.

Verify the permissions of the cron directories with the following command:

$ find /etc/cron* -type d | xargs stat -c "%a %n"

700 /etc/cron.d
700 /etc/cron.daily
700 /etc/cron.hourly
700 /etc/cron.monthly
700 /etc/cron.weekly

If any cron configuration directory is more permissive than "700", this is a finding.

Configure any RHEL 9 cron configuration directory with a mode more permissive than "0700" as follows:

chmod 0700 [cron configuration directory]

Local initialization files are used to configure the user's shell environment upon logon. Malicious modification of these files could compromise accounts upon logon.

Verify that all local initialization files have a mode of "0740" or less permissive with the following command:

Note: The example will be for the "wadea" user, who has a home directory of "/home/wadea".

$ sudo ls -al /home/wadea/.[^.]* | more

-rwxr-xr-x 1 wadea users 896 Mar 10 2011 .profile
-rwxr-xr-x 1 wadea users 497 Jan 6 2007 .login
-rwxr-xr-x 1 wadea users 886 Jan 6 2007 .something

If any local initialization files have a mode more permissive than "0740", this is a finding.

Set the mode of the local initialization files to "0740" with the following command:

Note: The example will be for the wadea user, who has a home directory of "/home/wadea".

$ sudo chmod 0740 /home/wadea/.

Excessive permissions on local interactive user home directories may allow unauthorized access to user files by other users.

Verify the assigned home directory of all local interactive users has a mode of "0750" or less permissive with the following command:

Note: This may miss interactive users that have been assigned a privileged user identifier (UID). Evidence of interactive use may be obtained from a number of log files containing system logon information.

$ sudo ls -ld $(awk -F: '($3>=1000)&&($7 !~ /nologin/){print $6}' /etc/passwd)

drwxr-x--- 2 wadea admin 4096 Jun 5 12:41 wadea

If home directories referenced in "/etc/passwd" do not have a mode of "0750" or less permissive, this is a finding.

Change the mode of interactive user's home directories to "0750". To change the mode of a local interactive user's home directory, use the following command:

Note: The example will be for the user "wadea".

$ sudo chmod 0750 /home/wadea

The "/etc/group" file contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify that the "/etc/group" file has mode "0644" or less permissive with the following command:

$ sudo stat -c "%a %n" /etc/group

644 /etc/group

If a value of "0644" or less permissive is not returned, this is a finding.

Change the mode of the file "/etc/group" to "0644" by running the following command:

$ sudo chmod 0644 /etc/group

The "/etc/group-" file is a backup file of "/etc/group", and as such, contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify that the "/etc/group-" file has mode "0644" or less permissive with the following command:

$ sudo stat -c "%a %n" /etc/group-

644 /etc/group-

If a value of "0644" or less permissive is not returned, this is a finding.

Change the mode of the file "/etc/group-" to "0644" by running the following command:

$ sudo chmod 0644 /etc/group-

The "/etc/gshadow" file contains group password hashes. Protection of this file is critical for system security.

Verify that the "/etc/gshadow" file has mode "0000" with the following command:

$ sudo stat -c "%a %n" /etc/gshadow

0 /etc/gshadow

If a value of "0" is not returned, this is a finding.

Change the mode of the file "/etc/gshadow" to "0000" by running the following command:

$ sudo chmod 0000 /etc/gshadow

The "/etc/gshadow-" file is a backup of "/etc/gshadow", and as such, contains group password hashes. Protection of this file is critical for system security.

Verify that the "/etc/gshadow-" file has mode "0000" with the following command:

$ sudo stat -c "%a %n" /etc/gshadow-

0 /etc/gshadow-

If a value of "0" is not returned, this is a finding.

Change the mode of the file "/etc/gshadow-" to "0000" by running the following command:

$ sudo chmod 0000 /etc/gshadow-

If the "/etc/passwd" file is writable by a group-owner or the world the risk of its compromise is increased. The file contains the list of accounts on the system and associated information, and protection of this file is critical for system security.

Verify that the "/etc/passwd" file has mode "0644" or less permissive with the following command:

$ sudo stat -c "%a %n" /etc/passwd

644 /etc/passwd

If a value of "0644" or less permissive is not returned, this is a finding.

Change the mode of the file "/etc/passwd" to "0644" by running the following command:

$ sudo chmod 0644 /etc/passwd

The "/etc/passwd-" file is a backup file of "/etc/passwd", and as such, contains information about the users that are configured on the system. Protection of this file is critical for system security.

Verify that the "/etc/passwd-" file has mode "0644" or less permissive with the following command:

$ sudo stat -c "%a %n" /etc/passwd-

644 /etc/passwd-

If a value of "0644" or less permissive is not returned, this is a finding.

Change the mode of the file "/etc/passwd-" to "0644" by running the following command:

$ sudo chmod 0644 /etc/passwd-

The "/etc/shadow-" file is a backup file of "/etc/shadow", and as such, contains the list of local system accounts and password hashes. Protection of this file is critical for system security.

Verify that the "/etc/shadow-" file has mode "0000" with the following command:

$ sudo stat -c "%a %n" /etc/shadow-

0 /etc/shadow-

If a value of "0" is not returned, this is a finding.

Change the mode of the file "/etc/shadow-" to "0000" by running the following command:

$ sudo chmod 0000 /etc/shadow-

The "/etc/group" file contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify the ownership of the "/etc/group" file with the following command:

$ sudo stat -c "%U %n" /etc/group 

root /etc/group 

If "/etc/group" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/group to root by running the following command:

$ sudo chown root /etc/group

The "/etc/group" file contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify the group ownership of the "/etc/group" file with the following command:

$ sudo stat -c "%G %n" /etc/group 

root /etc/group

If "/etc/group" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/group to root by running the following command:

$ sudo chgrp root /etc/group

The "/etc/group-" file is a backup file of "/etc/group", and as such, contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify the ownership of the "/etc/group-" file with the following command:

$ sudo stat -c "%U %n" /etc/group- 

root /etc/group- 

If "/etc/group-" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/group- to root by running the following command:

$ sudo chown root /etc/group-

The "/etc/group-" file is a backup file of "/etc/group", and as such, contains information regarding groups that are configured on the system. Protection of this file is important for system security.

Verify the group ownership of the "/etc/group-" file with the following command:

$ sudo stat -c "%G %n" /etc/group- 

root /etc/group-

If "/etc/group-" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/group- to root by running the following command:

$ sudo chgrp root /etc/group-

The "/etc/gshadow" file contains group password hashes. Protection of this file is critical for system security.

Verify the ownership of the "/etc/gshadow" file with the following command:

$ sudo stat -c "%U %n" /etc/gshadow 

root /etc/gshadow 

If "/etc/gshadow" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/gshadow to root by running the following command:

$ sudo chown root /etc/gshadow

The "/etc/gshadow" file contains group password hashes. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/gshadow" file with the following command:

$ sudo stat -c "%G %n" /etc/gshadow 

root /etc/gshadow

If "/etc/gshadow" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/gshadow to root by running the following command:

$ sudo chgrp root /etc/gshadow

The "/etc/gshadow-" file is a backup of "/etc/gshadow", and as such, contains group password hashes. Protection of this file is critical for system security.

Verify the ownership of the "/etc/gshadow-" file with the following command:

$ sudo stat -c "%U %n" /etc/gshadow- 

root /etc/gshadow- 

If "/etc/gshadow-" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/gshadow- to root by running the following command:

$ sudo chown root /etc/gshadow-

The "/etc/gshadow-" file is a backup of "/etc/gshadow", and as such, contains group password hashes. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/gshadow-" file with the following command:

$ sudo stat -c "%G %n" /etc/gshadow- 

root /etc/gshadow-

If "/etc/gshadow-" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/gshadow- to root by running the following command:

$ sudo chgrp root /etc/gshadow-

The "/etc/passwd" file contains information about the users that are configured on the system. Protection of this file is critical for system security.

Verify the ownership of the "/etc/passwd" file with the following command:

$ sudo stat -c "%U %n" /etc/passwd

root /etc/passwd

If "/etc/passwd" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/passwd to root by running the following command:

$ sudo chown root /etc/passwd

The "/etc/passwd" file contains information about the users that are configured on the system. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/passwd" file with the following command:

$ sudo stat -c "%G %n" /etc/passwd 

root /etc/passwd

If "/etc/passwd" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/passwd to root by running the following command:

$ sudo chgrp root /etc/passwd

The "/etc/passwd-" file is a backup file of "/etc/passwd", and as such, contains information about the users that are configured on the system. Protection of this file is critical for system security.

Verify the ownership of the "/etc/passwd-" file with the following command:

$ sudo stat -c "%U %n" /etc/passwd- 

root /etc/passwd- 

If "/etc/passwd-" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/passwd- to root by running the following command:

$ sudo chown root /etc/passwd-

The "/etc/passwd-" file is a backup file of "/etc/passwd", and as such, contains information about the users that are configured on the system. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/passwd-" file with the following command:

$ sudo stat -c "%G %n" /etc/passwd-

root /etc/passwd-

If "/etc/passwd-" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/passwd- to root by running the following command:

$ sudo chgrp root /etc/passwd-

The "/etc/shadow" file contains the list of local system accounts and stores password hashes. Protection of this file is critical for system security. Failure to give ownership of this file to root provides the designated owner with access to sensitive information, which could weaken the system security posture.

Verify the ownership of the "/etc/shadow" file with the following command:

$ sudo stat -c "%U %n" /etc/shadow

root /etc/shadow

If "/etc/shadow" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/shadow to root by running the following command:

$ sudo chown root /etc/shadow

The "/etc/shadow" file stores password hashes. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/shadow" file with the following command:

$ sudo stat -c "%G %n" /etc/shadow 

root /etc/shadow

If "/etc/shadow" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/shadow to root by running the following command:

$ sudo chgrp root /etc/shadow

The "/etc/shadow-" file is a backup file of "/etc/shadow", and as such, contains the list of local system accounts and password hashes. Protection of this file is critical for system security.

Verify the ownership of the "/etc/shadow-" file with the following command:

$ sudo stat -c "%U %n" /etc/shadow- 

root /etc/shadow- 

If "/etc/shadow-" file does not have an owner of "root", this is a finding.

Change the owner of the file /etc/shadow- to root by running the following command:

$ sudo chown root /etc/shadow-

The "/etc/shadow-" file is a backup file of "/etc/shadow", and as such, contains the list of local system accounts and password hashes. Protection of this file is critical for system security.

Verify the group ownership of the "/etc/shadow-" file with the following command:

$ sudo stat -c "%G %n" /etc/shadow-

root /etc/shadow-

If "/etc/shadow-" file does not have a group owner of "root", this is a finding.

Change the group of the file /etc/shadow- to root by running the following command:

$ sudo chgrp root /etc/shadow-

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify the "/var/log" directory is owned by root with the following command:

$ ls -ld /var/log

drwxr-xr-x. 16 root root 4096 July 11 11:34 /var/log

If "/var/log" does not have an owner of "root", this is a finding.

Configure the owner of the directory "/var/log" to "root" by running the following command:

$ sudo chown root /var/log

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify the "/var/log" directory is group-owned by root with the following command:

$ ls -ld /var/log

drwxr-xr-x. 16 root root 4096 July 11 11:34 /var/log

If "/var/log" does not have a group owner of "root", this is a finding.

Configure the group owner of the directory "/var/log" to "root" by running the following command:

$ sudo chgrp root /var/log

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify the "/var/log/messages" file is owned by root with the following command:

$ ls -la /var/log/messages

rw-------. 1 root root 564223 July 11 11:34 /var/log/messages

If "/var/log/messages" does not have an owner of "root", this is a finding.

Change the owner of the "/var/log/messages" file to "root" by running the following command:

$ sudo chown root /var/log/messages

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, personally identifiable information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Verify the "/var/log/messages" file is group-owned by root with the following command:

$ ls -la /var/log/messages

rw-------. 1 root root 564223 July 11 11:34 /var/log/messages

If "/var/log/messages" does not have a group owner of "root", this is a finding.

Change the group owner of the "/var/log/messages" file to "root" by running the following command:

$ sudo chgrp root /var/log/messages

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system commands contained in the following directories are owned by "root" with the following command:

$ sudo find -L /bin /sbin /usr/bin /usr/sbin /usr/libexec /usr/local/bin /usr/local/sbin ! -user root -exec ls -l {} \;

If any system commands are found to not be owned by root, this is a finding.

Configure the system commands to be protected from unauthorized access.

Run the following command, replacing "[FILE]" with any system command file not owned by "root".

$ sudo chown root [FILE]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system commands contained in the following directories are group-owned by "root", or a required system account, with the following command:

$ sudo find -L /bin /sbin /usr/bin /usr/sbin /usr/local/bin /usr/local/sbin ! -group root -exec ls -l {} \;

If any system commands are returned and is not group-owned by a required system account, this is a finding.

Configure the system commands to be protected from unauthorized access.

Run the following command, replacing "[FILE]" with any system command file not group-owned by "root" or a required system account.

$ sudo chgrp root [FILE]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library files are owned by "root" with the following command:

$ sudo find -L /lib /lib64 /usr/lib /usr/lib64 ! -user root -exec ls -l {} \;

If any system-wide shared library file is not owned by root, this is a finding.

Configure the system-wide shared library files (/lib, /lib64, /usr/lib and /usr/lib64) to be protected from unauthorized access.

Run the following command, replacing "[FILE]" with any library file not owned by "root".

$ sudo chown root [FILE]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library files are group-owned by "root" with the following command:

$ sudo find -L /lib /lib64 /usr/lib /usr/lib64 ! -group root -exec ls -l {} \;

If any system-wide shared library file is returned and is not group-owned by a required system account, this is a finding.

Configure the system-wide shared library files (/lib, /lib64, /usr/lib and /usr/lib64) to be protected from unauthorized access.

Run the following command, replacing "[FILE]" with any library file not group-owned by "root".

$ sudo chgrp root [FILE]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library directories are owned by "root" with the following command:

$ sudo find /lib /lib64 /usr/lib /usr/lib64 ! -user root -type d -exec stat -c "%n %U" '{}' \;

If any system-wide shared library directory is not owned by root, this is a finding.

Configure the system-wide shared library directories within (/lib, /lib64, /usr/lib and /usr/lib64) to be protected from unauthorized access.

Run the following command, replacing "[DIRECTORY]" with any library directory not owned by "root".

$ sudo chown root [DIRECTORY]

If RHEL 9 allowed any user to make changes to software libraries, then those changes might be implemented without undergoing the appropriate testing and approvals that are part of a robust change management process.

This requirement applies to RHEL 9 with software libraries that are accessible and configurable, as in the case of interpreted languages. Software libraries also include privileged programs that execute with escalated privileges.

Verify the system-wide shared library directories are group-owned by "root" with the following command:

$ sudo find /lib /lib64 /usr/lib /usr/lib64 ! -group root -type d -exec stat -c "%n %G" '{}' \;

If any system-wide shared library directory is returned and is not group-owned by a required system account, this is a finding.

Configure the system-wide shared library directories (/lib, /lib64, /usr/lib and /usr/lib64) to be protected from unauthorized access.

Run the following command, replacing "[DIRECTORY]" with any library directory not group-owned by "root".

$ sudo chgrp root [DIRECTORY]

Protecting audit information also includes identifying and protecting the tools used to view and manipulate log data. Therefore, protecting audit tools is necessary to prevent unauthorized operation on audit information.

RHEL 9 systems providing tools to interface with audit information will leverage user permissions and roles identifying the user accessing the tools, and the corresponding rights the user enjoys, to make access decisions regarding the access to audit tools.

Audit tools include, but are not limited to, vendor-provided and open source audit tools needed to successfully view and manipulate audit information system activity and records. Audit tools include custom queries and report generators.

Verify the audit tools are owned by "root" with the following command:

$ sudo stat -c "%U %n" /sbin/auditctl /sbin/aureport /sbin/ausearch /sbin/autrace /sbin/auditd /sbin/rsyslogd /sbin/augenrules

root /sbin/auditctl
root /sbin/aureport
root /sbin/ausearch
root /sbin/autrace
root /sbin/auditd
root /sbin/rsyslogd
root /sbin/augenrules

If any audit tools do not have an owner of "root", this is a finding.

Configure the audit tools to be owned by "root" by running the following command:

$ sudo chown root [audit_tool]

Replace "[audit_tool]" with each audit tool not owned by "root".

Protecting audit information also includes identifying and protecting the tools used to view and manipulate log data; therefore, protecting audit tools is necessary to prevent unauthorized operation on audit information.

RHEL 9 systems providing tools to interface with audit information will leverage user permissions and roles identifying the user accessing the tools, and the corresponding rights the user enjoys, to make access decisions regarding the access to audit tools.

Audit tools include, but are not limited to, vendor-provided and open source audit tools needed to successfully view and manipulate audit information system activity and records. Audit tools include custom queries and report generators.

Verify the audit tools are group owned by "root" with the following command:

$ sudo stat -c "%G %n" /sbin/auditctl /sbin/aureport /sbin/ausearch /sbin/autrace /sbin/auditd /sbin/rsyslogd /sbin/augenrules

root /sbin/auditctl
root /sbin/aureport
root /sbin/ausearch
root /sbin/autrace
root /sbin/auditd
root /sbin/rsyslogd
root /sbin/augenrules

If any audit tools do not have a group owner of "root", this is a finding.

Configure the audit tools to be group-owned by "root" by running the following command:

$ sudo chgrp root [audit_tool]

Replace "[audit_tool]" with each audit tool not group-owned by "root".

Service configuration files enable or disable features of their respective services that if configured incorrectly can lead to insecure and vulnerable configurations; therefore, service configuration files must be owned by the correct group to prevent unauthorized changes.

Verify the ownership of all cron configuration files with the command:

$ stat -c "%U %n" /etc/cron*

root /etc/cron.d
root /etc/cron.daily
root /etc/cron.deny
root /etc/cron.hourly
root /etc/cron.monthly
root /etc/crontab
root /etc/cron.weekly

If any crontab is not owned by root, this is a finding.

Configure any cron configuration not owned by root with the following command:

$ sudo chown root [cron config file]

Service configuration files enable or disable features of their respective services that if configured incorrectly can lead to insecure and vulnerable configurations; therefore, service configuration files should be owned by the correct group to prevent unauthorized changes.

Verify the group ownership of all cron configuration files with the following command:

$ stat -c "%G %n" /etc/cron*

root /etc/cron.d
root /etc/cron.daily
root /etc/cron.deny
root /etc/cron.hourly
root /etc/cron.monthly
root /etc/crontab
root /etc/cron.weekly

If any crontab is not group owned by root, this is a finding.

Configure any cron configuration not group-owned by root with the following command:

$ sudo chgrp root [cron config file]

If a world-writable directory is not owned by root, sys, bin, or an application user identifier (UID), unauthorized users may be able to modify files created by others.

The only authorized public directories are those temporary directories supplied with the system or those designed to be temporary file repositories. The setting is normally reserved for directories used by the system and by users for temporary file storage, (e.g., /tmp), and for directories requiring global read/write access.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000138-GPOS-00069

Verify that world writable directories are owned by root, a system account, or an application account with the following command. It will discover and print world-writable directories that are not owned by root.  Run it once for each local partition [PART]:

$ sudo find  PART  -xdev -type d -perm -0002 -uid +0 -print 

If there is output, this is a finding.

Configure all public directories to be owned by root or a system account to prevent unauthorized and unintended information transferred via shared system resources.

Set the owner of all public directories as root or a system account using the command, replace "[Public Directory]" with any directory path not owned by root or a system account:

$ sudo chown root [Public Directory]

Preventing unauthorized information transfers mitigates the risk of information, including encrypted representations of information, produced by the actions of prior users/roles (or the actions of processes acting on behalf of prior users/roles) from being available to any current users/roles (or current processes) that obtain access to shared system resources (e.g., registers, main memory, hard disks) after those resources have been released back to information systems. The control of information in shared resources is also commonly referred to as object reuse and residual information protection.

This requirement generally applies to the design of an information technology product, but it can also apply to the configuration of particular information system components that are, or use, such products. This can be verified by acceptance/validation processes in DOD or other government agencies.

Verify that all world-writable directories have the sticky bit set.

Determine if all world-writable directories have the sticky bit set by running the following command:

$ sudo find / -type d \( -perm -0002 -a ! -perm -1000 \) -print 2>/dev/null

drwxrwxrwt 7 root root 4096 Jul 26 11:19 /tmp

If any of the returned directories are world-writable and do not have the sticky bit set, this is a finding.

Configure all world-writable directories to have the sticky bit set to prevent unauthorized and unintended information transferred via shared system resources.

Set the sticky bit on all world-writable directories using the command, replace "[World-Writable Directory]" with any directory path missing the sticky bit:

$ chmod a+t [World-Writable Directory]

Files without a valid group owner may be unintentionally inherited if a group is assigned the same Group Identifier (GID) as the GID of the files without a valid group owner.

Verify all local files and directories on RHEL 9 have a valid group with the following command:

$ df --local -P | awk {'if (NR!=1) print $6'} | sudo xargs -I '{}' find '{}' -xdev -nogroup

If any files on the system do not have an assigned group, this is a finding.

Either remove all files and directories from RHEL 9 that do not have a valid group, or assign a valid group to all files and directories on the system with the "chgrp" command:

$ sudo chgrp  

Unowned files and directories may be unintentionally inherited if a user is assigned the same user identifier "UID" as the UID of the unowned files.

Verify all local files and directories on RHEL 9 have a valid owner with the following command:

$ df --local -P | awk {'if (NR!=1) print $6'} | sudo xargs -I '{}' find '{}' -xdev -nouser

If any files on the system do not have an assigned owner, this is a finding.

Either remove all files and directories from the system that do not have a valid user, or assign a valid user to all unowned files and directories on RHEL 9 with the "chown" command:

$ sudo chown  

If an unauthorized or modified device is allowed to exist on the system, there is the possibility the system may perform unintended or unauthorized operations.

Verify that all system device files are correctly labeled to prevent unauthorized modification.

List all device files on the system that are incorrectly labeled with the following commands:

Note: Device files are normally found under "/dev", but applications may place device files in other directories and may necessitate a search of the entire system.

# find /dev -context *:device_t:* \( -type c -o -type b \) -printf "%p %Z\n"

# find /dev -context *:unlabeled_t:* \( -type c -o -type b \) -printf "%p %Z\n"

Note: There are device files, such as "/dev/vmci", that are used when the operating system is a host virtual machine. They will not be owned by a user on the system and require the "device_t" label to operate. These device files are not a finding.

If there is output from either of these commands, other than already noted, this is a finding.

Restore the SELinux policy for the affected device file from the system policy database using the following command:

$ sudo restorecon -v 

Substitute "" with the path to the affected device file (from the output of the previous commands). An example device file path would be "/dev/ttyUSB0". If the output of the above command does not indicate that the device was relabeled to a more specific SELinux type label, then the SELinux policy of the system must be updated with more specific policy for the device class specified. If a package was used to install support for a device class, that package could be reinstalled using the following command:

$ sudo dnf reinstall 

If a package was not used to install the SELinux policy for a given device class, then it must be generated manually and provide specific type labels.

Service configuration files enable or disable features of their respective services that if configured incorrectly can lead to insecure and vulnerable configurations; therefore, service configuration files must have the correct access rights to prevent unauthorized changes.

Verify the permissions of /etc/crontab with the following command:

$ stat -c "%a %n" /etc/crontab

0600

If /etc/crontab does not have a mode of "0600", this is a finding.

Configure the RHEL 9 file /etc/crontab with mode 600.

$ sudo chmod 0600 /etc/crontab

The "/etc/shadow" file contains the list of local system accounts and stores password hashes. Protection of this file is critical for system security. Failure to give ownership of this file to root provides the designated owner with access to sensitive information, which could weaken the system security posture.

Verify that the "/etc/shadow" file has mode "0000" with the following command:

$ sudo stat -c "%a %n" /etc/shadow

0 /etc/shadow

If a value of "0" is not returned, this is a finding.

Change the mode of the file "/etc/shadow" to "0000" by running the following command:

$ sudo chmod 0000 /etc/shadow

"Firewalld" provides an easy and effective way to block/limit remote access to the system via ports, services, and protocols.

Remote access services, such as those providing remote access to network devices and information systems, which lack automated control capabilities, increase risk and make remote user access management difficult at best.

Remote access is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

RHEL 9 functionality (e.g., SSH) must be capable of taking enforcement action if the audit reveals unauthorized activity. Automated control of remote access sessions allows organizations to ensure ongoing compliance with remote access policies by enforcing connection rules of remote access applications on a variety of information system components (e.g., servers, workstations, notebook computers, smartphones, and tablets).

Satisfies: SRG-OS-000096-GPOS-00050, SRG-OS-000297-GPOS-00115, SRG-OS-000298-GPOS-00116, SRG-OS-000480-GPOS-00227, SRG-OS-000480-GPOS-00232

Run the following command to determine if the firewalld package is installed with the following command:

$ sudo dnf list --installed firewalld 

Example output:

firewalld.noarch          1.0.0-4.el9

If the "firewall" package is not installed, this is a finding.

To install the "firewalld" package run the following command:

$ sudo dnf install firewalld

"Firewalld" provides an easy and effective way to block/limit remote access to the system via ports, services, and protocols.

Remote access services, such as those providing remote access to network devices and information systems, which lack automated control capabilities, increase risk and make remote user access management difficult at best.

Remote access is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, non-organization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

RHEL 9 functionality (e.g., RDP) must be capable of taking enforcement action if the audit reveals unauthorized activity. Automated control of remote access sessions allows organizations to ensure ongoing compliance with remote access policies by enforcing connection rules of remote access applications on a variety of information system components (e.g., servers, workstations, notebook computers, smartphones, and tablets).

Satisfies: SRG-OS-000096-GPOS-00050, SRG-OS-000297-GPOS-00115, SRG-OS-000480-GPOS-00227, SRG-OS-000480-GPOS-00232

Verify that "firewalld" is active with the following command:

$ systemctl is-active firewalld 

active

If the firewalld service is not active, this is a finding.

To enable the firewalld service run the following command:

$ sudo systemctl enable --now firewalld

Failure to restrict network connectivity only to authorized systems permits inbound connections from malicious systems. It also permits outbound connections that may facilitate exfiltration of DOD data.

RHEL 9 incorporates the "firewalld" daemon, which allows for many different configurations. One of these configurations is zones. Zones can be utilized to a deny-all, allow-by-exception approach. The default "drop" zone will drop all incoming network packets unless it is explicitly allowed by the configuration file or is related to an outgoing network connection.

Verify the RHEL 9 "firewalld" is configured to employ a deny-all, allow-by-exception policy for allowing connections to other systems with the following commands:

$ sudo  firewall-cmd --state

running

$ sudo firewall-cmd --get-active-zones

public
   interfaces: ens33

$ sudo firewall-cmd --info-zone=public | grep target

   target: DROP

$ sudo firewall-cmd --permanent --info-zone=public | grep target

   target: DROP

If no zones are active on the RHEL 9 interfaces or if runtime and permanent targets are set to a different option other than "DROP", this is a finding.

Configure the "firewalld" daemon to employ a deny-all, allow-by-exception with the following commands:

Start by adding the exceptions that are required for mission functionality to the "drop" zone. If SSH access on port 22 is needed, for example, run the following: "sudo firewall-cmd --permanent --add-service=ssh --zone=drop"

Reload the firewall rules to update the runtime configuration from the "--permanent" changes made above:
$ sudo firewall-cmd --reload

Set the default zone to the drop zone:
$ sudo firewall-cmd --set-default-zone=drop
Note: This is a runtime and permanent change.

Add any interfaces to the newly modified "drop" zone:
$ sudo firewall-cmd --permanent --zone=drop --change-interface=ens33

Reload the firewall rules for changes to take effect:
$ sudo firewall-cmd --reload

DoS is a condition when a resource is not available for legitimate users. When this occurs, the organization either cannot accomplish its mission or must operate at degraded capacity.

This requirement addresses the configuration of RHEL 9 to mitigate the impact of DoS attacks that have occurred or are ongoing on system availability. For each system, known and potential DoS attacks must be identified and solutions for each type implemented. A variety of technologies exists to limit or, in some cases, eliminate the effects of DoS attacks (e.g., limiting processes or establishing memory partitions). Employing increased capacity and bandwidth, combined with service redundancy, may reduce the susceptibility to some DoS attacks.

Verify "nftables" is configured to allow rate limits on any connection to the system with the following command:

$ sudo grep -i firewallbackend /etc/firewalld/firewalld.conf

# FirewallBackend
FirewallBackend=nftables

If the "nftables" is not set as the "FirewallBackend" default, this is a finding.

Configure "nftables" to be the default "firewallbackend" for "firewalld" by adding or editing the following line in "etc/firewalld/firewalld.conf":

FirewallBackend=nftables

Establish rate-limiting rules based on organization-defined types of DoS attacks on impacted network interfaces.

To prevent unauthorized connection of devices, unauthorized transfer of information, or unauthorized tunneling (i.e., embedding of data types within data types), organizations must disable or restrict unused or unnecessary ports, protocols, and services on information systems.

Inspect the firewall configuration and running services to verify it is configured to prohibit or restrict the use of functions, ports, protocols, and/or services that are unnecessary or prohibited.

Check which services are currently active with the following command:

$ sudo firewall-cmd --list-all-zones

custom (active)
target: DROP
icmp-block-inversion: no
interfaces: ens33
sources: 
services: dhcpv6-client dns http https ldaps rpc-bind ssh
ports: 
masquerade: no
forward-ports: 
icmp-blocks: 
rich rules: 

Ask the system administrator for the site or program Ports, Protocols, and Services Management Component Local Service Assessment (PPSM CLSA). Verify the services allowed by the firewall match the PPSM CLSA. 

If there are additional ports, protocols, or services that are not in the PPSM CLSA, or there are ports, protocols, or services that are prohibited by the PPSM Category Assurance List (CAL), this is a finding.

Update the host's firewall settings and/or running services to comply with the PPSM CLSA for the site or program and the PPSM CAL.

Then run the following command to load the newly created rule(s):

$ sudo firewall-cmd --reload

Network interfaces in promiscuous mode allow for the capture of all network traffic visible to the system. If unauthorized individuals can access these applications, it may allow them to collect information such as logon IDs, passwords, and key exchanges between systems.

If the system is being used to perform a network troubleshooting function, the use of these tools must be documented with the information systems security officer (ISSO) and restricted to only authorized personnel.

Verify network interfaces are not in promiscuous mode with the following command:

$ ip link | grep -i promisc

If network interfaces are found on the system in promiscuous mode and their use has not been approved by the ISSO and documented, this is a finding.

Configure network interfaces to turn off promiscuous mode unless approved by the ISSO and documented.

Set the promiscuous mode of an interface to off with the following command:

$ sudo ip link set dev  multicast off promisc off

When hardened, the extended Berkeley Packet Filter (BPF) just-in-time (JIT) compiler will randomize any kernel addresses in the BPF programs and maps, and will not expose the JIT addresses in "/proc/kallsyms".

Verify RHEL 9 enables hardening for the BPF JIT with the following commands:

$ sudo sysctl net.core.bpf_jit_harden

net.core.bpf_jit_harden = 2

If the returned line does not have a value of "2", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.core.bpf_jit_harden | tail -1
net.core.bpf_jit_harden = 2

If the network parameter "net.core.bpf_jit_harden" is not equal to "2" or nothing is returned, this is a finding.

Configure RHEL 9 to enable hardening for the BPF JIT compiler by adding the following line to a file, in the "/etc/sysctl.d" directory:

net.core.bpf_jit_harden = 2

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

Inaccurate time stamps make it more difficult to correlate events and can lead to an inaccurate analysis. Determining the correct time a particular event occurred on a system is critical when conducting forensic analysis and investigating system events. Sources outside the configured acceptable allowance (drift) may be inaccurate.

Verify that RHEL 9 has the chrony package installed with the following command:

$ sudo dnf list --installed chrony

Example output:

chrony.x86_64          4.1-3.el9       

If the "chrony" package is not installed, this is a finding.

The chrony package can be installed with the following command:
 
$ sudo dnf install chrony

Inaccurate time stamps make it more difficult to correlate events and can lead to an inaccurate analysis. Determining the correct time a particular event occurred on a system is critical when conducting forensic analysis and investigating system events. Sources outside the configured acceptable allowance (drift) may be inaccurate.

Synchronizing internal information system clocks provides uniformity of time stamps for information systems with multiple system clocks and systems connected over a network.

Verify the chronyd service is active with the following command:

$ systemctl is-active chronyd

active 

If the chronyd service is not active, this is a finding.

To enable the chronyd service run the following command:

$ sudo systemctl enable --now chronyd

Inaccurate time stamps make it more difficult to correlate events and can lead to an inaccurate analysis. Determining the correct time a particular event occurred on a system is critical when conducting forensic analysis and investigating system events. Sources outside the configured acceptable allowance (drift) may be inaccurate.

Synchronizing internal information system clocks provides uniformity of time stamps for information systems with multiple system clocks and systems connected over a network.

Depending on the infrastructure being used the "pool" directive may not be supported.

Authoritative time sources include the United States Naval Observatory (USNO) time servers, a time server designated for the appropriate DOD network (NIPRNet/SIPRNet), and/or the Global Positioning System (GPS).

Satisfies: SRG-OS-000355-GPOS-00143, SRG-OS-000356-GPOS-00144, SRG-OS-000359-GPOS-00146

Verify RHEL 9 is securely comparing internal information system clocks at least every 24 hours with an NTP server with the following commands:

$ sudo grep maxpoll /etc/chrony.conf

server 0.us.pool.ntp.mil iburst maxpoll 16

If the "maxpoll" option is set to a number greater than 16 or the line is commented out, this is a finding.

Verify the "chrony.conf" file is configured to an authoritative DOD time source by running the following command:

$ sudo grep -i server /etc/chrony.conf
server 0.us.pool.ntp.mil 

If the parameter "server" is not set or is not set to an authoritative DOD time source, this is a finding.

Configure RHEL 9 to securely compare internal information system clocks at least every 24 hours with an NTP server by adding/modifying the following line in the /etc/chrony.conf file.

server [ntp.server.name] iburst maxpoll 16

Minimizing the exposure of the server functionality of the chrony daemon diminishes the attack surface.

Satisfies: SRG-OS-000096-GPOS-00050, SRG-OS-000095-GPOS-00049

Verify RHEL 9 disables the chrony daemon from acting as a server with the following command:

$ grep -w port /etc/chrony.conf

port 0

If the "port" option is not set to "0", is commented out, or is missing, this is a finding.

Configure RHEL 9 to disable the chrony daemon from acting as a server by adding/modifying the following line in the /etc/chrony.conf file:

port 0

Not exposing the management interface of the chrony daemon on the network diminishes the attack space.

Satisfies: SRG-OS-000096-GPOS-00050, SRG-OS-000095-GPOS-00049

Verify RHEL 9 disables network management of the chrony daemon with the following command:

$ grep -w cmdport /etc/chrony.conf

cmdport 0

If the "cmdport" option is not set to "0", is commented out, or is missing, this is a finding.

Configure RHEL 9 to disable network management of the chrony daemon by adding/modifying the following line in the /etc/chrony.conf file:

cmdport 0

To provide availability for name resolution services, multiple redundant name servers are mandated. A failure in name resolution could lead to the failure of security functions requiring name resolution, which may include time synchronization, centralized authentication, and remote system logging.

Verify the name servers used by the system with the following command:

$ grep nameserver /etc/resolv.conf

nameserver 192.168.1.2
nameserver 192.168.1.3

If less than two lines are returned that are not commented out, this is a finding.

Configure the operating system to use two or more name servers for DNS resolution based on the DNS mode of the system.

If the NetworkManager DNS mode is set to "none", then add the following lines to "/etc/resolv.conf":

nameserver [name server 1]
nameserver [name server 2]

Replace [name server 1] and [name server 2] with the IPs of two different DNS resolvers.

If the NetworkManager DNS mode is set to "default" then add two DNS servers to a NetworkManager connection. Using the following commands:

$ sudo nmcli connection modify [connection name] ipv4.dns [name server 1]
$ sudo nmcli connection modify [connection name] ipv4.dns [name server 2]

Replace [name server 1] and [name server 2] with the IPs of two different DNS resolvers. Replace [connection name] with a valid NetworkManager connection name on the system. Replace ipv4 with ipv6 if IPv6 DNS servers are used.

In order to ensure that DNS resolver settings are respected, a DNS mode in Network Manager must be configured.

Verify that RHEL 9 has a DNS mode configured in Network Manager.

$ NetworkManager --print-config
[main]
dns=none

If the dns key under main does not exist or is not set to "none" or "default", this is a finding.

Note: If RHEL 9 is configured to use a DNS resolver other than Network Manager, the configuration must be documented and approved by the information system security officer (ISSO).

Configure NetworkManager in RHEL 9 to use a DNS mode.

In "/etc/NetworkManager/NetworkManager.conf" add the following line in the "[main]" section:

dns = none

NetworkManager must be reloaded for the change to take effect.

$ sudo systemctl reload NetworkManager

IP tunneling mechanisms can be used to bypass network filtering. If tunneling is required, it must be documented with the information system security officer (ISSO).

Verify that RHEL 9 does not have unauthorized IP tunnels configured.

Determine if the "IPsec" service is active with the following command:

$ systemctl status ipsec

ipsec.service - Internet Key Exchange (IKE) Protocol Daemon for IPsec
Loaded: loaded (/usr/lib/systemd/system/ipsec.service; disabled)
Active: inactive (dead)

If the "IPsec" service is active, check for configured IPsec connections ("conn"), with the following command:

$ grep -rni conn /etc/ipsec.conf /etc/ipsec.d/ 

Verify any returned results are documented with the ISSO.

If the IPsec tunnels are active and not approved, this is a finding.

Remove all unapproved tunnels from the system, or document them with the ISSO.

If unrestricted mail relaying is permitted, unauthorized senders could use this host as a mail relay for the purpose of sending spam or other unauthorized activity.

If postfix is not installed, this is Not Applicable.

Verify RHEL 9 is configured to prevent unrestricted mail relaying with the following command:

$ postconf -n smtpd_client_restrictions 

smtpd_client_restrictions = permit_mynetworks,reject 

If the "smtpd_client_restrictions" parameter contains any entries other than "permit_mynetworks" and "reject", and the additional entries have not been documented with the information system security officer (ISSO), this is a finding.

Modify the postfix configuration file to restrict client connections to the local network with the following command:

$ sudo postconf -e 'smtpd_client_restrictions = permit_mynetworks,reject'

It is critical for the appropriate personnel to be aware if a system is at risk of failing to process audit logs as required. Without this notification, the security personnel may be unaware of an impending failure of the audit capability, and system operation may be adversely affected.

Audit processing failures include software/hardware errors, failures in the audit capturing mechanisms, and audit storage capacity being reached or exceeded.

Verify that the administrators are notified in the event of an audit processing failure.

Check that the "/etc/aliases" file has a defined value for "root".

$ sudo grep "postmaster:\s*root$" /etc/aliases

If the command does not return a line, or the line is commented out, ask the system administrator to indicate how they and the information systems security officer (ISSO) are notified of an audit process failure. If there is no evidence of the proper personnel being notified of an audit processing failure, this is a finding.

Configure a valid email address as an alias for the root account.

Append the following line to "/etc/aliases":

postmaster: root

Then, run the following command:

$ sudo newaliases

Providing the ability for remote users or systems to initiate a secure VPN connection protects information when it is transmitted over a wide area network.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000120-GPOS-00061

Verify that RHEL 9 libreswan service package is installed.

Check that the libreswan service package is installed with the following command:

$ sudo dnf list --installed libreswan

Example output:

libreswan.x86_64          4.6-3.el9

If the "libreswan" package is not installed, this is a finding.

Install the libreswan service (if it is not already installed) with the following command:

$ sudo dnf install libreswan

The shosts.equiv files are used to configure host-based authentication for the system via SSH. Host-based authentication is not sufficient for preventing unauthorized access to the system, as it does not require interactive identification and authentication of a connection request, or for the use of two-factor authentication.

Verify there are no "shosts.equiv" files on RHEL 9 with the following command:

$ sudo find / -name shosts.equiv

If a "shosts.equiv" file is found, this is a finding.

Remove any found "shosts.equiv" files from the system.

$ sudo rm /[path]/[to]/[file]/shosts.equiv

The .shosts files are used to configure host-based authentication for individual users or the system via SSH. Host-based authentication is not sufficient for preventing unauthorized access to the system, as it does not require interactive identification and authentication of a connection request, or for the use of two-factor authentication.

Verify there are no ".shosts" files on RHEL 9 with the following command:

$ sudo find / -name .shosts

If a ".shosts" file is found, this is a finding.

Remove any found ".shosts" files from the system.

$ sudo rm /[path]/[to]/[file]/.shosts

Denial of service (DoS) is a condition when a resource is not available for legitimate users. When this occurs, the organization either cannot accomplish its mission or must operate at degraded capacity. 

Managing excess capacity ensures that sufficient capacity is available to counter flooding attacks. Employing increased capacity and service redundancy may reduce the susceptibility to some DoS attacks. Managing excess capacity may include, for example, establishing selected usage priorities, quotas, or partitioning.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000420-GPOS-00186, SRG-OS-000142-GPOS-00071

Verify RHEL 9 is configured to use IPv4 TCP syncookies.

Determine if syncookies are used with the following command:

Check the status of the kernel.perf_event_paranoid kernel parameter.

$ sudo sysctl net.ipv4.tcp_syncookies

net.ipv4.tcp_syncookies = 1

Check that the configuration files are present to enable this kernel parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.tcp_syncookies | tail -1

net.ipv4.tcp_syncookies = 1

If the network parameter "ipv4.tcp_syncookies" is not equal to "1" or nothing is returned, this is a finding.

Configure RHEL 9 to use TCP syncookies.

Add or edit the following line in a system configuration file in the "/etc/sysctl.d/" directory:
 net.ipv4.tcp_syncookies = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

ICMP redirect messages are used by routers to inform hosts that a more direct route exists for a particular destination. These messages modify the host's route table and are unauthenticated. An illicit ICMP redirect message could result in a man-in-the-middle attack.

This feature of the IPv4 protocol has few legitimate uses. It should be disabled unless absolutely required.

Verify RHEL 9 will not accept IPv4 ICMP redirect messages.

Check the value of the all "accept_redirects" variables with the following command:

$ sudo sysctl net.ipv4.conf.all.accept_redirects

net.ipv4.conf.all.accept_redirects = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.all.accept_redirects | tail -1

net.ipv4.conf.all.accept_redirects = 0

If "net.ipv4.conf.all.accept_redirects" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to ignore IPv4 ICMP redirect messages.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.accept_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Source-routed packets allow the source of the packet to suggest routers forward the packet along a different path than configured on the router, which can be used to bypass network security measures. This requirement applies only to the forwarding of source-routerd traffic, such as when IPv4 forwarding is enabled and the system is functioning as a router.

Accepting source-routed packets in the IPv4 protocol has few legitimate uses. It must be disabled unless it is absolutely required.

Verify RHEL 9 will not accept IPv4 source-routed packets.

Check the value of the all "accept_source_route" variables with the following command:

$ sudo sysctl net.ipv4.conf.all.accept_source_route

net.ipv4.conf.all.accept_source_route = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.all.accept_source_route | tail -1

net.ipv4.conf.all.accept_source_route = 0

If "net.ipv4.conf.all.accept_source_route" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to ignore IPv4 source-routed packets.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.accept_source_route = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

The presence of "martian" packets (which have impossible addresses) as well as spoofed packets, source-routed packets, and redirects could be a sign of nefarious network activity. Logging these packets enables this activity to be detected.

Verify RHEL 9 logs IPv4 martian packets.

Check the value of the accept source route variable with the following command:

$ sudo sysctl net.ipv4.conf.all.log_martians

net.ipv4.conf.all.log_martians = 1

If the returned line does not have a value of "1", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.all.log_martians | tail -1

net.ipv4.conf.all.log_martians = 1

If "net.ipv4.conf.all.log_martians" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to log martian packets on IPv4 interfaces.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.log_martians=1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

The presence of "martian" packets (which have impossible addresses) as well as spoofed packets, source-routed packets, and redirects could be a sign of nefarious network activity. Logging these packets enables this activity to be detected.

Verify RHEL 9 logs IPv4 martian packets by default.

Check the value of the accept source route variable with the following command:

$ sudo sysctl net.ipv4.conf.default.log_martians

net.ipv4.conf.default.log_martians = 1

If the returned line does not have a value of "1", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.default.log_martians | tail -1

net.ipv4.conf.default.log_martians = 1

If "net.ipv4.conf.default.log_martians" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to log martian packets on IPv4 interfaces by default.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.default.log_martians=1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Enabling reverse path filtering drops packets with source addresses that should not have been able to be received on the interface on which they were received. It must not be used on systems that are routers for complicated networks, but is helpful for end hosts and routers serving small networks.

Verify RHEL 9 uses reverse path filtering on all IPv4 interfaces with the following commands:

$ sudo sysctl net.ipv4.conf.all.rp_filter

net.ipv4.conf.all.rp_filter = 1

If the returned line does not have a value of "1", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.all.rp_filter | tail -1

net.ipv4.conf.all.rp_filter = 1

If "net.ipv4.conf.all.rp_filter" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to use reverse path filtering on all IPv4 interfaces. 

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.rp_filter = 1

The system configuration files need to be reloaded for the changes to take effect. To reload the contents of the files, run the following command:

$ sudo sysctl --system

ICMP redirect messages are used by routers to inform hosts that a more direct route exists for a particular destination. These messages modify the host's route table and are unauthenticated. An illicit ICMP redirect message could result in a man-in-the-middle attack.

This feature of the IPv4 protocol has few legitimate uses. It must be disabled unless absolutely required.

Verify RHEL 9 will not accept IPv4 ICMP redirect messages.

Check the value of the default "accept_redirects" variables with the following command:

$ sudo sysctl net.ipv4.conf.default.accept_redirects

net.ipv4.conf.default.accept_redirects = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.default.accept_redirects | tail -1

net.ipv4.conf.default.accept_redirects = 0

If "net.ipv4.conf.default.accept_redirects" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to prevent IPv4 ICMP redirect messages from being accepted.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.default.accept_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Source-routed packets allow the source of the packet to suggest routers forward the packet along a different path than configured on the router, which can be used to bypass network security measures.

Accepting source-routed packets in the IPv4 protocol has few legitimate uses. It must be disabled unless it is absolutely required, such as when IPv4 forwarding is enabled and the system is legitimately functioning as a router.

Verify RHEL 9 does not accept IPv4 source-routed packets by default.

Check the value of the accept source route variable with the following command:

$ sudo sysctl net.ipv4.conf.default.accept_source_route

net.ipv4.conf.default.accept_source_route = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.default.accept_source_route | tail -1

net.ipv4.conf.default.accept_source_route = 0

If "net.ipv4.conf.default.accept_source_route" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not forward IPv4 source-routed packets by default.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.default.accept_source_route = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Enabling reverse path filtering drops packets with source addresses that should not have been able to be received on the interface on which they were received. It must not be used on systems that are routers for complicated networks, but is helpful for end hosts and routers serving small networks.

Verify RHEL 9 uses reverse path filtering on IPv4 interfaces with the following commands:

$ sudo sysctl net.ipv4.conf.default.rp_filter

net.ipv4.conf.default.rp_filter = 1

If the returned line does not have a value of "1", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.default.rp_filter | tail -1

net.ipv4.conf.default.rp_filter = 1

If "net.ipv4.conf.default.rp_filter" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to use reverse path filtering on IPv4 interfaces by default.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.default.rp_filter = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Responding to broadcast (ICMP) echoes facilitates network mapping and provides a vector for amplification attacks.

Ignoring ICMP echo requests (pings) sent to broadcast or multicast addresses makes the system slightly more difficult to enumerate on the network.

Verify RHEL 9 does not respond to ICMP echoes sent to a broadcast address.

Check the value of the "icmp_echo_ignore_broadcasts" variable with the following command:

$ sudo sysctl net.ipv4.icmp_echo_ignore_broadcasts

net.ipv4.icmp_echo_ignore_broadcasts = 1

If the returned line does not have a value of "1", a line is not returned, or the retuned line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|$)' | grep -F net.ipv4.icmp_echo_ignore_broadcasts | tail -1

net.ipv4.icmp_echo_ignore_broadcasts = 1

If "net.ipv4.icmp_echo_ignore_broadcasts" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to not respond to IPv4 ICMP echoes sent to a broadcast address.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.icmp_echo_ignore_broadcasts = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Some routers will send responses to broadcast frames that violate RFC-1122, which fills up a log file system with many useless error messages. An attacker may take advantage of this and attempt to flood the logs with bogus error logs. Ignoring bogus ICMP error responses reduces log size, although some activity would not be logged.

The runtime status of the net.ipv4.icmp_ignore_bogus_error_responses kernel parameter can be queried by running the following command:

$ sudo sysctl net.ipv4.icmp_ignore_bogus_error_responses 

net.ipv4.icmp_ignore_bogus_error_responses = 1

If "net.ipv4.icmp_ignore_bogus_error_responses" is not set to "1", this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.icmp_ignore_bogus_error_response | tail -1

net.ipv4.icmp_ignore_bogus_error_response = 1

If "net.ipv4.icmp_ignore_bogus_error_response" is not set to "1" or is missing, this is a finding.

Configure RHEL 9 to not log bogus ICMP errors: 

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.icmp_ignore_bogus_error_responses = 1

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

ICMP redirect messages are used by routers to inform hosts that a more direct route exists for a particular destination. These messages contain information from the system's route table possibly revealing portions of the network topology.

The ability to send ICMP redirects is only appropriate for systems acting as routers.

Verify RHEL 9 does not IPv4 ICMP redirect messages.

Check the value of the "all send_redirects" variables with the following command:

$ sudo sysctl net.ipv4.conf.all.send_redirects

net.ipv4.conf.all.send_redirects = 0

If the returned line does not have a value of "0", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv4.conf.all.send_redirects | tail -1

net.ipv4.conf.all.send_redirects = 0

If "net.ipv4.conf.all.send_redirects" is not set to "0" and is not documented with the information system security officer (ISSO) as an operational requirement or is missing, this is a finding.

Configure RHEL 9 to not allow interfaces to perform IPv4 ICMP redirects.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.send_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

ICMP redirect messages are used by routers to inform hosts that a more direct route exists for a particular destination. These messages contain information from the system's route table possibly revealing portions of the network topology.

The ability to send ICMP redirects is only appropriate for systems acting as routers.

Verify RHEL 9 does not allow interfaces to perform Internet Protocol version 4 (IPv4) ICMP redirects by default.

Check the value of the "default send_redirects" variables with the following command:

$ sudo sysctl net.ipv4.conf.default.send_redirects

net.ipv4.conf.default.send_redirects=0

If the returned line does not have a value of "0", or a line is not returned, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv4.conf.default.send_redirects | tail -1

net.ipv4.conf.default.send_redirects = 0

If "net.ipv4.conf.default.send_redirects" is not set to "0" and is not documented with the information system security officer (ISSO) as an operational requirement or is missing, this is a finding.

Configure RHEL 9 to not allow interfaces to perform Internet Protocol version 4 (IPv4) ICMP redirects by default.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.default.send_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Routing protocol daemons are typically used on routers to exchange network topology information with other routers. If this capability is used when not required, system network information may be unnecessarily transmitted across the network.

Verify RHEL 9 is not performing IPv4 packet forwarding, unless the system is a router.

Check that IPv4 forwarding is disabled using the following command:

$ sudo sysctl net.ipv4.conf.all.forwarding

net.ipv4.conf.all.forwarding = 0

If the IPv4 forwarding value is not "0" and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo (/usr/lib/systemd/systemd-sysctl --cat-config; cat /etc/sysctl.conf) | egrep -v '^(#|$)' | grep net.ipv4.conf.all.forwarding | tail -1

net.ipv4.conf.all.forwarding = 0

If "net.ipv4.conf.all.forwarding" is not set to "0" and is not documented with the ISSO as an operational requirement or is missing, this is a finding.

Configure RHEL 9 to not allow IPv4 packet forwarding, unless the system is a router.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv4.conf.all.forwarding = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

An illicit router advertisement message could result in a man-in-the-middle attack.

Verify RHEL 9 does not accept router advertisements on all IPv6 interfaces, unless the system is a router.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Determine if router advertisements are not accepted by using the following command:

$ sudo sysctl net.ipv6.conf.all.accept_ra

net.ipv6.conf.all.accept_ra = 0

If the "accept_ra" value is not "0" and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv6.conf.all.accept_ra | tail -1

net.ipv6.conf.all.accept_ra = 0

If "net.ipv6.conf.all.accept_ra" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not accept router advertisements on all IPv6 interfaces unless the system is a router.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.all.accept_ra = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

An illicit ICMP redirect message could result in a man-in-the-middle attack.

Verify RHEL 9 ignores IPv6 ICMP redirect messages.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Check the value of the "accept_redirects" variables with the following command:

$ sysctl net.ipv6.conf.all.accept_redirects

net.ipv6.conf.all.accept_redirects = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv6.conf.all.accept_redirects | tail -1

net.ipv6.conf.all.accept_redirects = 0

If "net.ipv6.conf.all.accept_redirects" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to ignore IPv6 ICMP redirect messages.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.all.accept_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Source-routed packets allow the source of the packet to suggest that routers forward the packet along a different path than configured on the router, which can be used to bypass network security measures. This requirement applies only to the forwarding of source-routed traffic, such as when forwarding is enabled and the system is functioning as a router.

Verify RHEL 9 does not accept IPv6 source-routed packets.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Check the value of the accept source route variable with the following command:

$ sudo sysctl net.ipv6.conf.all.accept_source_route

net.ipv6.conf.all.accept_source_route = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv6.conf.all.accept_source_route | tail -1

net.ipv6.conf.all.accept_source_route = 0

If "net.ipv6.conf.all.accept_source_route" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not forward IPv6 source-routed packets.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.all.accept_source_route = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

IP forwarding permits the kernel to forward packets from one network interface to another. The ability to forward packets between two networks is only appropriate for systems acting as routers.

Verify RHEL 9 is not performing IPv6 packet forwarding, unless the system is a router.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Check that IPv6 forwarding is disabled using the following commands:

$ sudo sysctl net.ipv6.conf.all.forwarding

net.ipv6.conf.all.forwarding = 0

If the IPv6 forwarding value is not "0" and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' | grep -F net.ipv6.conf.all.forwarding | tail -1

net.ipv6.conf.all.forwarding = 0

If "net.ipv6.conf.all.forwarding" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not allow IPv6 packet forwarding, unless the system is a router.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.all.forwarding = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

An illicit router advertisement message could result in a man-in-the-middle attack.

Verify RHEL 9 does not accept router advertisements on all IPv6 interfaces by default, unless the system is a router.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Determine if router advertisements are not accepted by default by using the following command:

$ sudo sysctl net.ipv6.conf.default.accept_ra

net.ipv6.conf.default.accept_ra = 0

If the "accept_ra" value is not "0" and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv6.conf.default.accept_ra | tail -1

net.ipv6.conf.default.accept_ra = 0

If "net.ipv6.conf.default.accept_ra" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not accept router advertisements on all IPv6 interfaces by default unless the system is a router.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.default.accept_ra = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

ICMP redirect messages are used by routers to inform hosts that a more direct route exists for a particular destination. These messages modify the host's route table and are unauthenticated. An illicit ICMP redirect message could result in a man-in-the-middle attack.

Verify RHEL 9 will not accept IPv6 ICMP redirect messages.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Check the value of the default "accept_redirects" variables with the following command:

$ sudo sysctl net.ipv6.conf.default.accept_redirects

net.ipv6.conf.default.accept_redirects = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv6.conf.default.accept_redirects | tail -1

net.ipv6.conf.default.accept_redirects = 0

If "net.ipv6.conf.default.accept_redirects" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to prevent IPv6 ICMP redirect messages from being accepted.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.default.accept_redirects = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Source-routed packets allow the source of the packet to suggest that routers forward the packet along a different path than configured on the router, which can be used to bypass network security measures. This requirement applies only to the forwarding of source-routed traffic, such as when forwarding is enabled and the system is functioning as a router.

Accepting source-routed packets in the IPv6 protocol has few legitimate uses. It must be disabled unless it is absolutely required.

Verify RHEL 9 does not accept IPv6 source-routed packets by default.

Note: If IPv6 is disabled on the system, this requirement is Not Applicable.

Check the value of the accept source route variable with the following command:

$ sudo sysctl net.ipv6.conf.default.accept_source_route

net.ipv6.conf.default.accept_source_route = 0

If the returned line does not have a value of "0", a line is not returned, or the line is commented out, this is a finding.

Check that the configuration files are present to enable this network parameter.

$ sudo /usr/lib/systemd/systemd-sysctl --cat-config | egrep -v '^(#|;)' |  grep -F net.ipv6.conf.default.accept_source_route | tail -1

net.ipv6.conf.default.accept_source_route = 0

If "net.ipv6.conf.default.accept_source_route" is not set to "0" or is missing, this is a finding.

Configure RHEL 9 to not forward IPv6 source-routed packets by default.

Add or edit the following line in a single system configuration file, in the "/etc/sysctl.d/" directory:

net.ipv6.conf.default.accept_source_route = 0

Load settings from all system configuration files with the following command:

$ sudo sysctl --system

Without protection of the transmitted information, confidentiality and integrity may be compromised because unprotected communications can be intercepted and either read or altered. 

This requirement applies to both internal and external networks and all types of information system components from which information can be transmitted (e.g., servers, mobile devices, notebook computers, printers, copiers, scanners, and facsimile machines). Communication paths outside the physical protection of a controlled boundary are exposed to the possibility of interception and modification. 

Protecting the confidentiality and integrity of organizational information can be accomplished by physical means (e.g., employing physical distribution systems) or by logical means (e.g., employing cryptographic techniques). If physical means of protection are employed, then logical means (cryptography) do not have to be employed, and vice versa.

Satisfies: SRG-OS-000423-GPOS-00187, SRG-OS-000424-GPOS-00188, SRG-OS-000425-GPOS-00189, SRG-OS-000426-GPOS-00190

Verify that RHEL 9 has the openssh-server package installed with the following command:

$ sudo dnf list --installed openssh-server

Example output:

openssh-server.x86_64          8.7p1-8.el9

If the "openssh-server" package is not installed, this is a finding.

The openssh-server package can be installed with the following command:
 
$ sudo dnf install openssh-server

Without protection of the transmitted information, confidentiality and integrity may be compromised because unprotected communications can be intercepted and either read or altered. 

This requirement applies to both internal and external networks and all types of information system components from which information can be transmitted (e.g., servers, mobile devices, notebook computers, printers, copiers, scanners, and facsimile machines). Communication paths outside the physical protection of a controlled boundary are exposed to the possibility of interception and modification. 

Protecting the confidentiality and integrity of organizational information can be accomplished by physical means (e.g., employing physical distribution systems) or by logical means (e.g., employing cryptographic techniques). If physical means of protection are employed, then logical means (cryptography) do not have to be employed, and vice versa.

Satisfies: SRG-OS-000423-GPOS-00187, SRG-OS-000424-GPOS-00188, SRG-OS-000425-GPOS-00189, SRG-OS-000426-GPOS-00190

Verify that "sshd" is active with the following command:

$ systemctl is-active sshd

active

If the "sshd" service is not active, this is a finding.

To enable the sshd service run the following command:

$ systemctl enable --now sshd

This package includes utilities to make encrypted connections and transfer files securely to SSH servers.

Verify that RHEL 9 has the openssh-clients package installed with the following command:

$ sudo dnf list --installed openssh-clients

Example output:

openssh-clients.x86_64          8.7p1-8.el9

If the "openssh-clients" package is not installed, this is a finding.

The openssh-clients package can be installed with the following command:
 
$ sudo dnf install openssh-clients

The warning message reinforces policy awareness during the logon process and facilitates possible legal action against attackers. Alternatively, systems whose ownership should not be obvious should ensure usage of a banner that does not provide easy attribution.

Satisfies: SRG-OS-000023-GPOS-00006, SRG-OS-000228-GPOS-00088

Verify that any SSH connection to the operating system displays the Standard Mandatory DOD Notice and Consent Banner before granting access to the system.

Check for the location of the banner file being used with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*banner'

banner /etc/issue

This command will return the banner keyword and the name of the file that contains the SSH banner (in this case "/etc/issue").

If the line is commented out, this is a finding.

Configure RHEL 9 to display the Standard Mandatory DOD Notice and Consent Banner before granting access to the system via ssh.

Edit the "/etc/ssh/sshd_config" file to uncomment the banner keyword and configure it to point to a file that will contain the logon banner (this file may be named differently or be in a different location if using a version of SSH that is provided by a third-party vendor).

An example configuration line is:

Banner /etc/issue

SSH provides several logging levels with varying amounts of verbosity. "DEBUG" is specifically not recommended other than strictly for debugging SSH communications since it provides so much data that it is difficult to identify important security information. "INFO" or "VERBOSE" level is the basic level that only records login activity of SSH users. In many situations, such as Incident Response, it is important to determine when a particular user was active on a system. The logout record can eliminate those users who disconnected, which helps narrow the field.

Verify that RHEL 9 logs SSH connection attempts and failures to the server.

Check what the SSH daemon's "LogLevel" option is set to with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*loglevel'

LogLevel VERBOSE

If a value of "VERBOSE" is not returned or the line is commented out or missing, this is a finding.

Configure RHEL 9 to log connection attempts add or modify the following line in "/etc/ssh/sshd_config".

LogLevel VERBOSE

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

Without the use of multifactor authentication, the ease of access to privileged functions is greatly increased. Multifactor authentication requires using two or more factors to achieve authentication. A privileged account is defined as an information system account with authorizations of a privileged user. A DOD common access card (CAC) with DOD-approved PKI is an example of multifactor authentication.

Satisfies: SRG-OS-000105-GPOS-00052, SRG-OS-000106-GPOS-00053, SRG-OS-000107-GPOS-00054, SRG-OS-000108-GPOS-00055

Verify that RHEL 9 SSH daemon accepts public key encryption with the following command:
 
$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*pubkeyauthentication'

PubkeyAuthentication yes
 
If "PubkeyAuthentication" is set to no, the line is commented out, or the line is missing, this is a finding.

Note: If the system administrator demonstrates the use of an approved alternate multifactor authentication method, this requirement is not applicable.

To configure the system add or modify the following line in "/etc/ssh/sshd_config".

PubkeyAuthentication yes

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

If an account has an empty password, anyone could log on and run commands with the privileges of that account. Accounts with empty passwords should never be used in operational environments.

Satisfies: SRG-OS-000106-GPOS-00053, SRG-OS-000480-GPOS-00229, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 remote access using SSH prevents logging on with a blank password with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*permitemptypasswords'

PermitEmptyPasswords no

If the "PermitEmptyPasswords" keyword is set to "yes", is missing, or is commented out, this is a finding.

To configure the system to prevent SSH users from logging on with blank passwords edit the following line in "etc/ssh/sshd_config":

PermitEmptyPasswords no

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

Even though the communications channel may be encrypted, an additional layer of security is gained by extending the policy of not logging directly on as root. In addition, logging in with a user-specific account provides individual accountability of actions performed on the system and also helps to minimize direct attack attempts on root's password.

Satisfies: SRG-OS-000109-GPOS-00056, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 remote access using SSH prevents users from logging on directly as "root" with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*permitrootlogin'

PermitRootLogin no

If the "PermitRootLogin" keyword is set to "yes", is missing, or is commented out, this is a finding.

To configure the system to prevent SSH users from logging on directly as root add or modify the following line in "/etc/ssh/sshd_config".

PermitRootLogin no

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

When UsePAM is set to "yes", PAM runs through account and session types properly. This is important when restricted access to services based off of IP, time, or other factors of the account is needed. Additionally, this ensures users can inherit certain environment variables on login or disallow access to the server.

Verify the RHEL 9 SSHD is configured to allow for the UsePAM interface with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*usepam'

UsePAM yes

If the "UsePAM" keyword is set to "no", is missing, or is commented out, this is a finding.

Configure the RHEL 9 SSHD to use the UsePAM interface add or modify the following line in "/etc/ssh/sshd_config".

UsePAM yes

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

Verify that systemwide crypto policies are in effect with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*include'

/etc/ssh/sshd_config:Include /etc/ssh/sshd_config.d/*.conf
/etc/ssh/sshd_config.d/50-redhat.conf:Include /etc/crypto-policies/back-ends/opensshserver.config

If "Include /etc/ssh/sshd_config.d/*.conf" or "Include /etc/crypto-policies/back-ends/opensshserver.config" are not included in the system sshd config this is a finding. Additionally, if the file /etc/ssh/sshd_config.d/50-redhat.conf is missing, this is a finding.

Configure the RHEL 9 SSH daemon to use system-wide crypto policies by running the following commands:

$ sudo dnf reinstall openssh-server

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, non-organization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

RHEL 9 incorporates system-wide crypto policies by default. The SSH configuration file has no effect on the ciphers, MACs, or algorithms unless specifically defined in the /etc/sysconfig/sshd file. The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/opensshserver.config file.

Verify that system-wide crypto policies are in effect with the following command:

$ sudo grep Include /etc/ssh/sshd_config  /etc/ssh/sshd_config.d/*

/etc/ssh/sshd_config:Include /etc/ssh/sshd_config.d/*.conf
/etc/ssh/sshd_config.d/50-redhat.conf:Include /etc/crypto-policies/back-ends/opensshserver.config

If "Include /etc/ssh/sshd_config.d/*.conf" or "Include /etc/crypto-policies/back-ends/opensshserver.config" are not included in the system sshd config or the file "/etc/ssh/sshd_config.d/50-redhat.conf" is missing, this is a finding.

Configure the RHEL 9 SSH daemon to use system-wide crypto policies by running the following commands:

$ sudo dnf reinstall openssh-clients

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

RHEL 9 incorporates system-wide crypto policies by default. The SSH configuration file has no effect on the ciphers, MACs, or algorithms unless specifically defined in the /etc/sysconfig/sshd file. The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/opensshserver.config file.

Verify the SSH client is configured to use only ciphers employing FIPS 140-3 approved algorithms with the following command:

$ sudo grep -i ciphers /etc/crypto-policies/back-ends/openssh.config 

Ciphers aes256-gcm@openssh.com,chacha20-poly1305@openssh.com,aes256-ctr,aes128-gcm@openssh.com,aes128-ctr

If the cipher entries in the "openssh.config" file have any ciphers other than "aes256-gcm@openssh.com,chacha20-poly1305@openssh.com,aes256-ctr,aes128-gcm@openssh.com,aes128-ctr", the order differs from the example above, they are missing, or commented out, this is a finding.

Configure the RHEL 9 SSH client to use only ciphers employing FIPS 140-3 approved algorithms by updating the "/etc/crypto-policies/back-ends/openssh.config" file with the following line:

Ciphers aes256-gcm@openssh.com,chacha20-poly1305@openssh.com,aes256-ctr,aes128-gcm@openssh.com,aes128-ctr

A reboot is required for the changes to take effect.

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

RHEL 9 incorporates system-wide crypto policies by default. The SSH configuration file has no effect on the ciphers, MACs, or algorithms unless specifically defined in the /etc/sysconfig/sshd file. The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/opensshserver.config file.

Verify SSH server is configured to use only ciphers employing FIPS 140-3 approved algorithms with the following command:

$ sudo grep -i macs /etc/crypto-policies/back-ends/openssh.config
MACs hmac-sha2-256-etm@openssh.com,hmac-sha1-etm@openssh.com,umac-128-etm@openssh.com,hmac-sha2-512-etm@openssh.com,hmac-sha2-256,hmac-sha1,umac-128@openssh.com,hmac-sha2-512

If the MACs entries in the "openssh.config" file have any hashes other than "hmac-sha2-256-etm@openssh.com,hmac-sha2-256,hmac-sha2-512-etm@openssh.com,hmac-sha2-512", the order differs from the example above, or they are missing or commented out, this is a finding.

Configure the RHEL 9 SSH server to use only MACs employing FIPS 140-3 approved algorithms by updating the "/etc/crypto-policies/back-ends/openssh.config" file with the following line:

MACs hmac-sha2-256-etm@openssh.com,hmac-sha1-etm@openssh.com,umac-128-etm@openssh.com,hmac-sha2-512-etm@openssh.com,hmac-sha2-256,hmac-sha1,umac-128@openssh.com,hmac-sha2-512

A reboot is required for the changes to take effect.

SSH trust relationships mean a compromise on one host can allow an attacker to move trivially to other hosts.

Verify the operating system does not allow a noncertificate trusted host SSH logon to the system with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*hostbasedauthentication'

HostbasedAuthentication no

If the "HostbasedAuthentication" keyword is not set to "no", is missing, or is commented out, this is a finding.

If the required value is not set, this is a finding.

To configure RHEL 9 to not allow a noncertificate trusted host SSH logon to the system add or modify the following line in "/etc/ssh/sshd_config".

HostbasedAuthentication no

Restart the SSH daemon for the settings to take effect:

$ sudo systemctl restart sshd.service

SSH environment options potentially allow users to bypass access restriction in some configurations.

Verify that unattended or automatic logon via SSH is disabled with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*permituserenvironment'

PermitUserEnvironment no

If "PermitUserEnvironment" is set to "yes", is missing completely, or is commented out, this is a finding.

If the required value is not set, this is a finding.

Configure the RHEL 9 SSH daemon to not allow unattended or automatic logon to the system.

Add or edit the following line in the "/etc/ssh/sshd_config" file:

PermitUserEnvironment no

Restart the SSH daemon  for the setting to take effect:

$ sudo systemctl restart sshd.service

Without protection of the transmitted information, confidentiality and integrity may be compromised because unprotected communications can be intercepted and either read or altered. 

This requirement applies to both internal and external networks and all types of information system components from which information can be transmitted (e.g., servers, mobile devices, notebook computers, printers, copiers, scanners, and facsimile machines). Communication paths outside the physical protection of a controlled boundary are exposed to the possibility of interception and modification. 

Protecting the confidentiality and integrity of organizational information can be accomplished by physical means (e.g., employing physical distribution systems) or by logical means (e.g., employing cryptographic techniques). If physical means of protection are employed, then logical means (cryptography) do not have to be employed, and vice versa.

Session key regeneration limits the chances of a session key becoming compromised.

Satisfies: SRG-OS-000423-GPOS-00187, SRG-OS-000033-GPOS-00014, SRG-OS-000424-GPOS-00188

Verify the SSH server is configured to force frequent session key renegotiation with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*rekeylimit'

RekeyLimit 1G 1h

If "RekeyLimit" does not have a maximum data amount and maximum time defined, is missing, or is commented out, this is a finding.

Configure RHEL 9 to force a frequent session key renegotiation for SSH connections to the server by adding or modifying the following line in the "/etc/ssh/sshd_config" file:

RekeyLimit 1G 1h

Restart the SSH daemon for the settings to take effect.

$ sudo systemctl restart sshd.service

Terminating an unresponsive SSH session within a short time period reduces the window of opportunity for unauthorized personnel to take control of a management session enabled on the console or console port that has been left unattended. In addition, quickly terminating an idle SSH session will also free up resources committed by the managed network element.

Terminating network connections associated with communications sessions includes, for example, deallocating associated TCP/IP address/port pairs at the operating system level and deallocating networking assignments at the application level if multiple application sessions are using a single operating system-level network connection. This does not mean the operating system terminates all sessions or network access; it only ends the unresponsive session and releases the resources associated with that session.

RHEL 9 utilizes /etc/ssh/sshd_config for configurations of OpenSSH. Within the sshd_config, the product of the values of "ClientAliveInterval" and "ClientAliveCountMax" are used to establish the inactivity threshold. The "ClientAliveInterval" is a timeout interval in seconds, after which if no data has been received from the client, sshd will send a message through the encrypted channel to request a response from the client. The "ClientAliveCountMax" is the number of client alive messages that may be sent without sshd receiving any messages back from the client. If this threshold is met, sshd will disconnect the client. For more information on these settings and others, refer to the sshd_config man pages.

Satisfies: SRG-OS-000163-GPOS-00072, SRG-OS-000279-GPOS-00109

Verify the "ClientAliveCountMax" is set to "1" by performing the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*clientalivecountmax'

ClientAliveCountMax 1

If "ClientAliveCountMax" does not exist, is not set to a value of "1" in "/etc/ssh/sshd_config", or is commented out, this is a finding.

Note: This setting must be applied in conjunction with RHEL-09-255100 to function correctly.

Configure the SSH server to terminate a user session automatically after the SSH client has become unresponsive.

Modify or append the following lines in the "/etc/ssh/sshd_config" file:

ClientAliveCountMax 1

In order for the changes to take effect, the SSH daemon must be restarted.

$ sudo systemctl restart sshd.service

Terminating an unresponsive SSH session within a short time period reduces the window of opportunity for unauthorized personnel to take control of a management session enabled on the console or console port that has been left unattended. In addition, quickly terminating an idle SSH session will also free up resources committed by the managed network element.

Terminating network connections associated with communications sessions includes, for example, deallocating associated TCP/IP address/port pairs at the operating system level and deallocating networking assignments at the application level if multiple application sessions are using a single operating system-level network connection. This does not mean the operating system terminates all sessions or network access; it only ends the unresponsive session and releases the resources associated with that session.

RHEL 9 utilizes /etc/ssh/sshd_config for configurations of OpenSSH. Within the sshd_config, the product of the values of "ClientAliveInterval" and "ClientAliveCountMax" are used to establish the inactivity threshold. The "ClientAliveInterval" is a timeout interval in seconds, after which if no data has been received from the client, sshd will send a message through the encrypted channel to request a response from the client. The "ClientAliveCountMax" is the number of client alive messages that may be sent without sshd receiving any messages back from the client. If this threshold is met, sshd will disconnect the client. For more information on these settings and others, refer to the sshd_config man pages.

Satisfies: SRG-OS-000126-GPOS-00066, SRG-OS-000163-GPOS-00072, SRG-OS-000279-GPOS-00109, SRG-OS-000395-GPOS-00175

Verify the "ClientAliveInterval" variable is set to a value of "600" or less by performing the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*clientaliveinterval'

ClientAliveInterval 600

If "ClientAliveInterval" does not exist, does not have a value of "600" or less in "/etc/ssh/sshd_config", or is commented out, this is a finding.

Note: This setting must be applied in conjunction with RHEL-09-255095 to function correctly.

Configure the SSH server to terminate a user session automatically after the SSH client has been unresponsive for 10 minutes.

Modify or append the following lines in the "/etc/ssh/sshd_config" file:

ClientAliveInterval 600

In order for the changes to take effect, the SSH daemon must be restarted.

$ sudo systemctl restart sshd.service

Service configuration files enable or disable features of their respective services, which if configured incorrectly, can lead to insecure and vulnerable configurations. Therefore, service configuration files must be owned by the correct group to prevent unauthorized changes.

Verify the group ownership of the "/etc/ssh/sshd_config" file with the following command:

$ ls -al /etc/ssh/sshd_config

rw-------. 1 root root 3669 Feb 22 11:34 /etc/ssh/sshd_config

If the "/etc/ssh/sshd_config" file does not have a group owner of "root", this is a finding.

Configure the "/etc/ssh/sshd_config" file to be group-owned by root with the following command:

$ sudo chgrp root /etc/ssh/sshd_config

Service configuration files enable or disable features of their respective services, which if configured incorrectly, can lead to insecure and vulnerable configurations. Therefore, service configuration files must be owned by the correct group to prevent unauthorized changes.

Verify the ownership of the "/etc/ssh/sshd_config" file with the following command:

$ ls -al /etc/ssh/sshd_config

rw-------. 1 root root 3669 Feb 22 11:34 /etc/ssh/sshd_config

If the "/etc/ssh/sshd_config" file does not have an owner of "root", this is a finding.

Configure the "/etc/ssh/sshd_config" file to be owned by root with the following command:

$ sudo chown root /etc/ssh/sshd_config

Service configuration files enable or disable features of their respective services that if configured incorrectly can lead to insecure and vulnerable configurations. Therefore, service configuration files should be owned by the correct group to prevent unauthorized changes.

Verify the permissions of the "/etc/ssh/sshd_config" file with the following command:

$ ls -al /etc/ssh/sshd_config

rw-------. 1 root root 3669 Feb 22 11:34 /etc/ssh/sshd_config

If the "/etc/ssh/sshd_config" permissions are not "0600", this is a finding.

Configure the "/etc/ssh/sshd_config" permissions to be "0600" with the following command:

$ sudo chmod 0600 /etc/ssh/sshd_config

If an unauthorized user obtains the private SSH host key file, the host could be impersonated.

Verify the SSH private host key files have a mode of "0640" or less permissive with the following command:

$ ls -l /etc/ssh/*_key

640 /etc/ssh/ssh_host_dsa_key
640 /etc/ssh/ssh_host_ecdsa_key
640 /etc/ssh/ssh_host_ed25519_key
640 /etc/ssh/ssh_host_rsa_key

If any private host key file has a mode more permissive than "0640", this is a finding.

Configure the mode of SSH private host key files under "/etc/ssh" to "0640" with the following command:

$ sudo chmod 0640 /etc/ssh/ssh_host*key

Restart the SSH daemon for the changes to take effect:

$ sudo systemctl restart sshd.service

If a public host key file is modified by an unauthorized user, the SSH service may be compromised.

Verify the SSH public host key files have a mode of "0644" or less permissive with the following command:

Note: SSH public key files may be found in other directories on the system depending on the installation.

$ sudo stat -c "%a %n" /etc/ssh/*.pub

644 /etc/ssh/ssh_host_dsa_key.pub
644 /etc/ssh/ssh_host_ecdsa_key.pub
644 /etc/ssh/ssh_host_ed25519_key.pub
644 /etc/ssh/ssh_host_rsa_key.pub

If any key.pub file has a mode more permissive than "0644", this is a finding.

Change the mode of public host key files under "/etc/ssh" to "0644" with the following command:

$ sudo chmod 0644 /etc/ssh/*key.pub

Restart the SSH daemon for the changes to take effect:

$ sudo systemctl restart sshd.service

If compression is allowed in an SSH connection prior to authentication, vulnerabilities in the compression software could result in compromise of the system from an unauthenticated connection, potentially with root privileges.

Verify the SSH daemon performs compression after a user successfully authenticates with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*compression'

Compression delayed

If the "Compression" keyword is set to "yes", is missing, or the returned line is commented out, this is a finding.

Configure the SSH daemon to not allow compression.

Uncomment the "Compression" keyword in "/etc/ssh/sshd_config" on the system and set the value to "delayed" or "no":

Compression no

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

Generic Security Service Application Program Interface (GSSAPI) authentication is used to provide additional authentication mechanisms to applications. Allowing GSSAPI authentication through SSH exposes the system's GSSAPI to remote hosts, increasing the attack surface of the system.

Satisfies: SRG-OS-000364-GPOS-00151, SRG-OS-000480-GPOS-00227

Verify the SSH daemon does not allow GSSAPI authentication with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*gssapiauthentication'

GSSAPIAuthentication no

If the value is returned as "yes", the returned line is commented out, no output is returned, and the use of GSSAPI authentication has not been documented with the information system security officer (ISSO), this is a finding.

If the required value is not set, this is a finding.

Configure the SSH daemon to not allow GSSAPI authentication.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "no":

GSSAPIAuthentication no

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

Kerberos authentication for SSH is often implemented using Generic Security Service Application Program Interface (GSSAPI). If Kerberos is enabled through SSH, the SSH daemon provides a means of access to the system's Kerberos implementation. Vulnerabilities in the system's Kerberos implementations may be subject to exploitation.

Satisfies: SRG-OS-000364-GPOS-00151, SRG-OS-000480-GPOS-00227

Verify the SSH daemon does not allow Kerberos authentication with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*kerberosauthentication'

KerberosAuthentication no

If the value is returned as "yes", the returned line is commented out, no output is returned, and the use of Kerberos authentication has not been documented with the information system security officer (ISSO), this is a finding.

Configure the SSH daemon to not allow Kerberos authentication.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "no":

KerberosAuthentication no

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

SSH trust relationships mean a compromise on one host can allow an attacker to move trivially to other hosts.

Verify the SSH daemon does not allow rhosts authentication with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*ignorerhosts'

IgnoreRhosts yes

If the value is returned as "no", the returned line is commented out, or no output is returned, this is a finding.

Configure the SSH daemon to not allow rhosts authentication.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "yes":

IgnoreRhosts yes

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

Configuring the IgnoreUserKnownHosts setting for the SSH daemon provides additional assurance that remote login via SSH will require a password, even in the event of misconfiguration elsewhere.

Verify the SSH daemon does not allow known hosts authentication with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*ignoreuserknownhosts'

IgnoreUserKnownHosts yes

If the value is returned as "no", the returned line is commented out, or no output is returned, this is a finding.

Configure the SSH daemon to not allow known hosts authentication.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "yes":

IgnoreUserKnownHosts yes

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

When X11 forwarding is enabled, there may be additional exposure to the server and client displays if the sshd proxy display is configured to listen on the wildcard address.  By default, sshd binds the forwarding server to the loopback address and sets the hostname part of the DISPLAY environment variable to localhost. This prevents remote hosts from connecting to the proxy display.

Verify the SSH daemon does not allow X11Forwarding with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*x11forwarding'

X11forwarding no

If the value is returned as "yes", the returned line is commented out, or no output is returned, and X11 forwarding is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Configure the SSH daemon to not allow X11 forwarding.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "no":

X11forwarding no

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

If other users have access to modify user-specific SSH configuration files, they may be able to log into the system as another user.

Verify the SSH daemon performs strict mode checking of home directory configuration files with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*strictmodes'

StrictModes yes

If the "StrictModes" keyword is set to "no", the returned line is commented out, or no output is returned, this is a finding.

Configure the SSH daemon to perform strict mode checking of home directory configuration files.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "yes":

StrictModes yes

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

Providing users feedback on when account accesses last occurred facilitates user recognition and reporting of unauthorized account use.

Verify the SSH daemon provides users with feedback on when account accesses last occurred with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*printlastlog'

PrintLastLog yes

If the "PrintLastLog" keyword is set to "no", the returned line is commented out, or no output is returned, this is a finding.

Configure the SSH daemon to provide users with feedback on when account accesses last occurred.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "yes":

PrintLastLog yes

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

When X11 forwarding is enabled, there may be additional exposure to the server and client displays if the sshd proxy display is configured to listen on the wildcard address. By default, sshd binds the forwarding server to the loopback address and sets the hostname part of the "DISPLAY" environment variable to localhost. This prevents remote hosts from connecting to the proxy display.

Verify the SSH daemon prevents remote hosts from connecting to the proxy display with the following command:

$ sudo /usr/sbin/sshd -dd 2>&1 | awk '/filename/ {print $4}' | tr -d '\r' | tr '\n' ' ' | xargs sudo grep -iH '^\s*x11uselocalhost'

X11UseLocalhost yes

If the "X11UseLocalhost" keyword is set to "no", is missing, or is commented out, this is a finding.

Configure the SSH daemon to prevent remote hosts from connecting to the proxy display.

Add the following line in "/etc/ssh/sshd_config", or uncomment the line and set the value to "yes":

X11UseLocalhost yes

The SSH service must be restarted for changes to take effect:

$ sudo systemctl restart sshd.service

Display of a standardized and approved use notification before granting access to the operating system ensures privacy and security notification verbiage used is consistent with applicable federal laws, Executive Orders, directives, policies, regulations, standards, and guidance.

For U.S. Government systems, system use notifications are required only for access via login interfaces with human users and are not required when such human interfaces do not exist.

Satisfies: SRG-OS-000023-GPOS-00006, SRG-OS-000228-GPOS-00088

Verify RHEL 9 displays the Standard Mandatory DOD Notice and Consent Banner before granting access to the operating system via a graphical user logon.

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine if the operating system displays a banner at the logon screen with the following command:

$ gsettings get org.gnome.login-screen banner-message-enable

true

If the result is "false", this is a finding.

Configure RHEL 9 to display the Standard Mandatory DOD Notice and Consent Banner before granting access to the system via a graphical user logon.

Create a database to contain the system-wide graphical user logon settings (if it does not already exist) with the following command:

$ sudo touch /etc/dconf/db/local.d/01-banner-message

Add the following lines to the [org/gnome/login-screen] section of the "/etc/dconf/db/local.d/01-banner-message":

[org/gnome/login-screen]

banner-message-enable=true

Run the following command to update the database:

$ sudo dconf update

Display of a standardized and approved use notification before granting access to the operating system ensures privacy and security notification verbiage used is consistent with applicable federal laws, Executive Orders, directives, policies, regulations, standards, and guidance.

For U.S. Government systems, system use notifications are required only for access via login interfaces with human users and are not required when such human interfaces do not exist.

Satisfies: SRG-OS-000023-GPOS-00006, SRG-OS-000228-GPOS-00088

Verify RHEL 9 prevents a user from overriding settings for graphical user interfaces. 

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that graphical settings are locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ sudo grep banner-message-enable /etc/dconf/db/local.d/* 

/org/gnome/login-screen/banner-message-enable

If the output is not "/org/gnome/login-screen/banner-message-enable", the line is commented out, or the line is missing, this is a finding.

Configure RHEL 9 to prevent a user from overriding the banner setting for graphical user interfaces. 

Create a database to contain the system-wide graphical user logon settings (if it does not already exist) with the following command:

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following setting to prevent nonprivileged users from modifying it:

banner-message-enable

Run the following command to update the database:

$ sudo dconf update

Automatically mounting file systems permits easy introduction of unknown devices, thereby facilitating malicious activity.

Satisfies: SRG-OS-000114-GPOS-00059, SRG-OS-000378-GPOS-00163, SRG-OS-000480-GPOS-00227

Verify RHEL 9 disables the graphical user interface automount function with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.desktop.media-handling automount-open 

false

If "automount-open" is set to "true", and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Configure the GNOME desktop to disable automated mounting of removable media.

The dconf settings can be edited in the /etc/dconf/db/* location.

Update the [org/gnome/desktop/media-handling] section of the "/etc/dconf/db/local.d/00-security-settings" database file and add or update the following lines:

[org/gnome/desktop/media-handling]
automount-open=false

Then update the dconf system databases:

$ sudo dconf update

A nonprivileged account is any operating system account with authorizations of a nonprivileged user.

Satisfies: SRG-OS-000114-GPOS-00059, SRG-OS-000378-GPOS-00163, SRG-OS-000480-GPOS-00227

Verify RHEL 9 disables ability of the user to override the graphical user interface automount setting.

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that the automount setting is locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ grep 'automount-open' /etc/dconf/db/local.d/locks/* 

/org/gnome/desktop/media-handling/automount-open

If the command does not return at least the example result, this is a finding.

Configure the GNOME desktop to not allow a user to change the setting that disables automated mounting of removable media.

Add the following line to "/etc/dconf/db/local.d/locks/00-security-settings-lock" to prevent user modification:

/org/gnome/desktop/media-handling/automount-open

Then update the dconf system databases:

$ sudo dconf update

Allowing autorun commands to execute may introduce malicious code to a system. Configuring this setting prevents autorun commands from executing.

Verify RHEL 9 disables the graphical user interface autorun function with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.desktop.media-handling autorun-never 

true

If "autorun-never" is set to "false", and is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Configure the GNOME desktop to disable the autorun function on removable media.

The dconf settings can be edited in the /etc/dconf/db/* location.

Update the [org/gnome/desktop/media-handling] section of the "/etc/dconf/db/local.d/00-security-settings" database file and add or update the following lines:

[org/gnome/desktop/media-handling]
autorun-never=true

Then update the dconf system databases:

$ sudo dconf update

Techniques used to address this include protocols using nonces (e.g., numbers generated for a specific one-time use) or challenges (e.g., TLS, WS_Security). Additional techniques include time-synchronous or challenge-response one-time authenticators.

Satisfies: SRG-OS-000114-GPOS-00059, SRG-OS-000378-GPOS-00163, SRG-OS-000480-GPOS-00227

Verify RHEL 9 disables ability of the user to override the graphical user interface autorun setting.

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that the automount setting is locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ grep 'autorun-never' /etc/dconf/db/local.d/locks/* 

/org/gnome/desktop/media-handling/autorun-never

If the command does not return at least the example result, this is a finding.

Configure the GNOME desktop to not allow a user to change the setting that disables autorun on removable media.

Add the following line to "/etc/dconf/db/local.d/locks/00-security-settings-lock" to prevent user modification:

/org/gnome/desktop/media-handling/autorun-never

Then update the dconf system databases:

$ sudo dconf update

Failure to restrict system access to authenticated users negatively impacts operating system security.

Verify RHEL 9 does not allow an unattended or automatic logon to the system via a graphical user interface.

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Check for the value of the "AutomaticLoginEnable" in the "/etc/gdm/custom.conf" file with the following command:

$  grep -i automaticlogin /etc/gdm/custom.conf

[daemon]
AutomaticLoginEnable=false

If the value of "AutomaticLoginEnable" is not set to "false", this is a finding.

Configure the GNOME desktop display manager to disable automatic login.

Set AutomaticLoginEnable to false in the [daemon] section in /etc/gdm/custom.conf. For example:

[daemon]
AutomaticLoginEnable=false

A session lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not want to log out because of the temporary nature of the absence.

The session lock is implemented at the point where session activity can be determined. Rather than be forced to wait for a period of time to expire before the user session can be locked, RHEL 9 needs to provide users with the ability to manually invoke a session lock so users can secure their session if it is necessary to temporarily vacate the immediate physical vicinity.

Satisfies: SRG-OS-000028-GPOS-00009, SRG-OS-000030-GPOS-00011

Verify RHEL 9 enables a user's session lock until that user re-establishes access using established identification and authentication procedures with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ grep -R removal-action /etc/dconf/db/*

/etc/dconf/db/distro.d/20-authselect:removal-action='lock-screen'

If the "removal-action='lock-screen'" setting is missing or commented out from the dconf database files, this is a finding.

Configure RHEL 9 to enable a user's session lock until that user re-establishes access using established identification and authentication procedures.

Select or create an authselect profile and incorporate the "with-smartcard-lock-on-removal" feature with the following example:

$ sudo authselect select sssd with-smartcard with-smartcard-lock-on-removal

Alternatively, the dconf settings can be edited in the /etc/dconf/db/* location.

Add or update the [org/gnome/settings-daemon/peripherals/smartcard] section of the /etc/dconf/db/local.d/00-security-settings" database file and add or update the following lines:

[org/gnome/settings-daemon/peripherals/smartcard]
removal-action='lock-screen'

Then update the dconf system databases:

$ sudo dconf update

A session lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not want to log out because of the temporary nature of the absence.

The session lock is implemented at the point where session activity can be determined. Rather than be forced to wait for a period of time to expire before the user session can be locked, RHEL 9 needs to provide users with the ability to manually invoke a session lock so users can secure their session if it is necessary to temporarily vacate the immediate physical vicinity.

Satisfies: SRG-OS-000028-GPOS-00009, SRG-OS-000030-GPOS-00011

Verify RHEL 9 disables ability of the user to override the smart card removal action setting.

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that the removal action setting is locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ grep 'removal-action' /etc/dconf/db/local.d/locks/* 

/org/gnome/settings-daemon/peripherals/smartcard/removal-action

If the command does not return at least the example result, this is a finding.

Add the following line to "/etc/dconf/db/local.d/locks/00-security-settings-lock" to prevent user override of the smart card removal action:

/org/gnome/settings-daemon/peripherals/smartcard/removal-action

Then update the dconf system databases:

$ sudo dconf update

A session lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not want to log out because of the temporary nature of the absence.

The session lock is implemented at the point where session activity can be determined.

Regardless of where the session lock is determined and implemented, once invoked, the session lock must remain in place until the user reauthenticates. No other activity aside from reauthentication must unlock the system.

Satisfies: SRG-OS-000028-GPOS-00009, SRG-OS-000030-GPOS-00011

Verify RHEL 9 enables a user's session lock until that user re-establishes access using established identification and authentication procedures with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.desktop.screensaver lock-enabled

true

If the setting is "false", this is a finding.

Configure RHEL 9 to enable a user's session lock until that user re-establishes access using established identification and authentication procedures.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following example:

$ sudo vi /etc/dconf/db/local.d/00-screensaver

Edit the "[org/gnome/desktop/screensaver]" section of the database file and add or update the following lines:

# Set this to true to lock the screen when the screensaver activates
lock-enabled=true

Update the system databases:

$ sudo dconf update

A session time-out lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not log out because of the temporary nature of the absence. Rather than relying on the user to manually lock their operating system session prior to vacating the vicinity, operating systems need to be able to identify when a user's session has idled and take action to initiate the session lock.

The session lock is implemented at the point where session activity can be determined and/or controlled.

Implementing session settings will have little value if a user is able to manipulate these settings from the defaults prescribed in the other requirements of this implementation guide.

Satisfies: SRG-OS-000028-GPOS-00009, SRG-OS-000030-GPOS-00011

Verify RHEL 9 prevents a user from overriding settings for graphical user interfaces. 

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, Gnome Shell. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that graphical settings are locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ sudo grep -i lock-enabled /etc/dconf/db/local.d/locks/*

/org/gnome/desktop/screensaver/lock-enabled

If the command does not return at least the example result, this is a finding.

Configure RHEL 9 to prevent a user from overriding settings for graphical user interfaces.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command: 

Note: The example below is using the database "local" for the system, so if the system is using another database in "/etc/dconf/profile/user", the file should be created under the appropriate subdirectory.

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following setting to prevent nonprivileged users from modifying it:

/org/gnome/desktop/screensaver/lock-enabled

A session time-out lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not logout because of the temporary nature of the absence. Rather than relying on the user to manually lock their operating system session prior to vacating the vicinity, the GNOME desktop can be configured to identify when a user's session has idled and take action to initiate a session lock.

Satisfies: SRG-OS-000029-GPOS-00010, SRG-OS-000031-GPOS-00012

Verify RHEL 9 initiates a session lock after a 15-minute period of inactivity for graphical user interfaces with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ sudo gsettings get org.gnome.desktop.session idle-delay

uint32 900

If "idle-delay" is set to "0" or a value greater than "900", this is a finding.

Configure RHEL 9 to initiate a screensaver after a 15-minute period of inactivity for graphical user interfaces.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command:

$ sudo touch /etc/dconf/db/local.d/00-screensaver

Edit /etc/dconf/db/local.d/00-screensaver and add or update the following lines:

[org/gnome/desktop/session]
# Set the lock time out to 900 seconds before the session is considered idle
idle-delay=uint32 900

Update the system databases:

$ sudo dconf update

A session time-out lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not logout because of the temporary nature of the absence. Rather than relying on the user to manually lock their operating system session prior to vacating the vicinity, the GNOME desktop can be configured to identify when a user's session has idled and take action to initiate the session lock. As such, users should not be allowed to change session settings.

Satisfies: SRG-OS-000029-GPOS-00010, SRG-OS-000031-GPOS-00012

Verify RHEL 9 prevents a user from overriding settings for graphical user interfaces. 

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that graphical settings are locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ sudo grep -i idle /etc/dconf/db/local.d/locks/*

/org/gnome/desktop/session/idle-delay

If the command does not return at least the example result, this is a finding.

Configure RHEL 9 to prevent a user from overriding settings for graphical user interfaces.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command: 

Note: The example below is using the database "local" for the system, so if the system is using another database in "/etc/dconf/profile/user", the file should be created under the appropriate subdirectory.

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following setting to prevent nonprivileged users from modifying it:

/org/gnome/desktop/session/idle-delay

A session lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not want to logout because of the temporary nature of the absence.

Verify RHEL 9 initiates a session lock for graphical user interfaces when the screensaver is activated with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.desktop.screensaver lock-delay

uint32 5

If the "uint32" setting is not set to "5" or less, or is missing, this is a finding.

Configure RHEL 9 to initiate a session lock for graphical user interfaces when a screensaver is activated.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command: 

Note: The example below is using the database "local" for the system, so if the system is using another database in "/etc/dconf/profile/user", the file should be created under the appropriate subdirectory.

$ sudo touch /etc/dconf/db/local.d/00-screensaver

[org/gnome/desktop/screensaver]
lock-delay=uint32 5

The "uint32" must be included along with the integer key values as shown.

Update the system databases:

$ sudo dconf update

A session time-out lock is a temporary action taken when a user stops work and moves away from the immediate physical vicinity of the information system but does not logout because of the temporary nature of the absence. Rather than relying on the user to manually lock their operating system session prior to vacating the vicinity, the GNOME desktop can be configured to identify when a user's session has idled and take action to initiate the session lock. As such, users should not be allowed to change session settings.

Verify RHEL 9 prevents a user from overriding settings for graphical user interfaces. 

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that graphical settings are locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ sudo grep -i lock-delay /etc/dconf/db/local.d/locks/*

/org/gnome/desktop/screensaver/lock-delay

If the command does not return at least the example result, this is a finding.

Configure RHEL 9 to prevent a user from overriding settings for graphical user interfaces.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command: 

Note: The example below is using the database "local" for the system, so if the system is using another database in "/etc/dconf/profile/user", the file should be created under the appropriate subdirectory.

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following setting to prevent nonprivileged users from modifying it:

/org/gnome/desktop/screensaver/lock-delay

Setting the screensaver mode to blank-only conceals the contents of the display from passersby.

To ensure the screensaver is configured to be blank, run the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.desktop.screensaver picture-uri 

If properly configured, the output should be "''".

To ensure that users cannot set the screensaver background, run the following: 

$ grep picture-uri /etc/dconf/db/local.d/locks/* 

If properly configured, the output should be "/org/gnome/desktop/screensaver/picture-uri".

If it is not set or configured properly, this is a finding.

The dconf settings can be edited in the /etc/dconf/db/* location.

First, add or update the [org/gnome/desktop/screensaver] section of the "/etc/dconf/db/local.d/00-security-settings" database file and add or update the following lines:

[org/gnome/desktop/screensaver]
picture-uri=''

Then, add the following line to "/etc/dconf/db/local.d/locks/00-security-settings-lock" to prevent user modification:

/org/gnome/desktop/screensaver/picture-uri

Finally, update the dconf system databases:

$ sudo dconf update

Unlike text-based keyfiles, the binary database is impossible to check through most automated and all manual means; therefore, in order to evaluate dconf configuration, both have to be true at the same time - configuration files have to be compliant, and the database needs to be more recent than those keyfiles, which gives confidence that it reflects them.

Check the last modification time of the local databases, comparing it to the last modification time of the related keyfiles. The following command will check every dconf database and compare its modification time to the related system keyfiles:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ function dconf_needs_update { for db in $(find /etc/dconf/db -maxdepth 1 -type f); do db_mtime=$(stat -c %Y "$db"); keyfile_mtime=$(stat -c %Y "$db".d/* | sort -n | tail -1); if [ -n "$db_mtime" ] && [ -n "$keyfile_mtime" ] && [ "$db_mtime" -lt "$keyfile_mtime" ]; then echo "$db needs update"; return 1; fi; done; }; dconf_needs_update

If the command has any output, then a dconf database needs to be updated, and this is a finding.

Update the dconf databases by running the following command:

$ sudo dconf update

A user who is at the console can reboot the system at the login screen. If restart or shutdown buttons are pressed at the login screen, this can create the risk of short-term loss of availability of systems due to reboot.

Verify RHEL 9 disables a user's ability to restart the system with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, Gnome Shell. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ grep -R disable-restart-buttons /etc/dconf/db/*

/etc/dconf/db/distro.d/20-authselect:disable-restart-buttons='true'

If the "disable-restart-button" setting is not set to "true", is missing or commented out from the dconf database files, this is a finding.

Configure RHEL 9 to disable a user's ability to restart the system.

Add or update the [org/gnome/settings-daemon/] section of the /etc/dconf/db/local.d/00-security-settings database file and add or update the following lines:

[org/gnome/login-screen]
disable-restart-buttons=true

Then update the dconf system databases:

$ sudo dconf update

A user who is at the console can reboot the system at the login screen. If restart or shutdown buttons are pressed at the login screen, this can create the risk of short-term loss of availability of systems due to reboot.

Verify RHEL 9 prevents a user from overriding the disable-restart-buttons setting for graphical user interfaces. 

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

Determine which profile the system database is using with the following command:

$ sudo grep system-db /etc/dconf/profile/user

system-db:local

Check that graphical settings are locked from nonprivileged user modification with the following command:

Note: The example below is using the database "local" for the system, so the path is "/etc/dconf/db/local.d". This path must be modified if a database other than "local" is being used.

$ grep disable-restart-buttons /etc/dconf/db/local.d/locks/* 

/org/gnome/login-screen/disable-restart-buttons

If the command does not return at least the example result, this is a finding.

Configure RHEL 9 to prevent a user from overriding the disable-restart-buttons setting for graphical user interfaces. 

Create a database to contain the system-wide graphical user logon settings (if it does not already exist) with the following command:

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following line to prevent nonprivileged users from modifying it:

/org/gnome/login-screen/disable-restart-buttons

Run the following command to update the database:

$ sudo dconf update

A locally logged-in user who presses Ctrl-Alt-Del, when at the console, can reboot the system. If accidentally pressed, as could happen in the case of mixed OS environment, this can create the risk of short-term loss of availability of systems due to unintentional reboot.

Verify RHEL 9 is configured to ignore the Ctrl-Alt-Del sequence in the GNOME desktop with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ gsettings get org.gnome.settings-daemon.plugins.media-keys logout 

"['']"

If the GNOME desktop is configured to shut down when Ctrl-Alt-Del is pressed, this is a finding.

Configure RHEL 9 to ignore the Ctrl-Alt-Del sequence in the GNOME desktop.

Add or update the [org/gnome/settings-daemon/plugins/media-keys] section of the /etc/dconf/db/local.d/00-security-settings database file and add or update the following lines:

[org/gnome/settings-daemon/plugins/media-keys]
logout=['']

Run the following command to update the database:

$ sudo dconf update

A locally logged-in user who presses Ctrl-Alt-Del, when at the console, can reboot the system. If accidentally pressed, as could happen in the case of mixed OS environment, this can create the risk of short-term loss of availability of systems due to unintentional reboot.

Verify that users cannot enable the Ctrl-Alt-Del sequence in the GNOME desktop with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ grep logout /etc/dconf/db/local.d/locks/* 

/org/gnome/settings-daemon/plugins/media-keys/logout

If the output is not "/org/gnome/settings-daemon/plugins/media-keys/logout", the line is commented out, or the line is missing, this is a finding.

Configure RHEL 9 to disallow the user changing the Ctrl-Alt-Del sequence in the GNOME desktop.

Create a database to container system-wide graphical user logon settings (if it does not already exist) with the following command:

$ sudo touch /etc/dconf/db/local.d/locks/session

Add the following line to the session locks file to prevent nonprivileged users from modifying the Ctrl-Alt-Del setting:

/org/gnome/settings-daemon/plugins/media-keys/logout

Run the following command to update the database:

$ sudo dconf update

Leaving the user list enabled is a security risk since it allows anyone with physical access to the system to enumerate known user accounts without authenticated access to the system.

Verify that RHEL 9 disables the user logon list for graphical user interfaces with the following command:

Note: This requirement assumes the use of the RHEL 9 default graphical user interface, the GNOME desktop environment. If the system does not have any graphical user interface installed, this requirement is Not Applicable.

$ sudo gsettings get org.gnome.login-screen disable-user-list
true

If the setting is "false", this is a finding.

Configure RHEL 9 to disable the user list at logon for graphical user interfaces.

Create a database to contain the system-wide screensaver settings (if it does not already exist) with the following command:
Note: The example below is using the database "local" for the system, so if the system is using another database in "/etc/dconf/profile/user", the file should be created under the appropriate subdirectory.

$ sudo touch /etc/dconf/db/local.d/02-login-screen

[org/gnome/login-screen]
disable-user-list=true

Update the system databases:

$ sudo dconf update

USB mass storage permits easy introduction of unknown devices, thereby facilitating malicious activity.

Satisfies: SRG-OS-000114-GPOS-00059, SRG-OS-000378-GPOS-00163, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 disables the ability to load the USB Storage kernel module with the following command:

$ sudo grep -r usb-storage /etc/modprobe.conf /etc/modprobe.d/* 

blacklist usb-storage

If the command does not return any output, or the line is commented out, and use of USB Storage is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

To configure the system to prevent the usb-storage kernel module from being loaded, add the following line to the file  /etc/modprobe.d/usb-storage.conf (or create usb-storage.conf if it does not exist):

install usb-storage /bin/false
blacklist usb-storage

The USBguard-daemon is the main component of the USBGuard software framework. It runs as a service in the background and enforces the USB device authorization policy for all USB devices. The policy is defined by a set of rules using a rule language described in the usbguard-rules.conf file. The policy and the authorization state of USB devices can be modified during runtime using the usbguard tool.

The system administrator (SA) must work with the site information system security officer (ISSO) to determine a list of authorized peripherals and establish rules within the USBGuard software framework to allow only authorized devices.

Verify USBGuard is installed on the operating system with the following command:

$ sudo dnf list installed usbguard

Example output:

Installed Packages
usbguard.x86_64          1.0.0-10.el9_1.2          @rhel-9-for-x86_64-appstream-rpms

If the USBGuard package is not installed, ask the SA to indicate how unauthorized peripherals are being blocked.

If there is no evidence that unauthorized peripherals are being blocked before establishing a connection, this is a finding.

If the system is virtual machine with no virtual or physical USB peripherals attached, this is not a finding.

Install the usbguard package with the following command:

$ sudo dnf install usbguard

The USBguard-daemon is the main component of the USBGuard software framework. It runs as a service in the background and enforces the USB device authorization policy for all USB devices. The policy is defined by a set of rules using a rule language described in the usbguard-rules.conf file. The policy and the authorization state of USB devices can be modified during runtime using the usbguard tool.

The system administrator (SA) must work with the site information system security officer (ISSO) to determine a list of authorized peripherals and establish rules within the USBGuard software framework to allow only authorized devices.

Verify RHEL 9 has USBGuard enabled with the following command:

$ systemctl is-active usbguard

active

If usbguard is not active, ask the SA to indicate how unauthorized peripherals are being blocked.

If there is no evidence that unauthorized peripherals are being blocked before establishing a connection, this is a finding.

If the system is virtual machine with no virtual or physical USB peripherals attached, this is not a finding.

To enable the USBGuard service run the following command:

$ sudo systemctl enable --now usbguard

Without the capability to generate audit records, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

If auditing is enabled late in the startup process, the actions of some startup processes may not be audited. Some audit systems also maintain state information only available if auditing is enabled before a given process is created.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

The list of audited events is the set of events for which audits are to be generated. This set of events is typically a subset of the list of all events for which the system is capable of generating audit records.

DOD has defined the list of events for which RHEL 9 will provide an audit record generation capability as the following:

1) Successful and unsuccessful attempts to access, modify, or delete privileges, security objects, security levels, or categories of information (e.g., classification levels);

2) Access actions, such as successful and unsuccessful logon attempts, privileged activities or other system-level access, starting and ending time for user access to the system, concurrent logons from different workstations, successful and unsuccessful accesses to objects, all program initiations, and all direct access to the information system;

3) All account creations, modifications, disabling, and terminations; and 

4) All kernel module load, unload, and restart actions.

To verify that Linux Audit logging is enabled for the USBGuard daemon with the following command:

$ sudo grep AuditBackend /etc/usbguard/usbguard-daemon.conf 

AuditBackend=LinuxAudit 

If "AuditBackend" is not set to "LinuxAudit", this is a finding.

If the system is virtual machine with no virtual or physical USB peripherals attached, this is not a finding.

Configure RHEL 9 USBGuard AuditBackend to use the audit system.

Add or edit the following line in /etc/usbguard/usbguard-daemon.conf 

AuditBackend=LinuxAudit

The USBguard-daemon is the main component of the USBGuard software framework. It runs as a service in the background and enforces the USB device authorization policy for all USB devices. The policy is defined by a set of rules using a rule language described in the usbguard-rules.conf file. The policy and the authorization state of USB devices can be modified during runtime using the usbguard tool.

The system administrator (SA) must work with the site information system security officer (ISSO) to determine a list of authorized peripherals and establish rules within the USBGuard software framework to allow only authorized devices.

Note: If the system is virtual machine with no virtual or physical USB peripherals attached, this is Not Applicable.

Verify the USBGuard has a policy configured with the following command:

$ usbguard list-rules

allow id 1d6b:0001 serial

If the command does not return results or an error is returned, ask the SA to indicate how unauthorized peripherals are being blocked.

If there is no evidence that unauthorized peripherals are being blocked before establishing a connection, this is a finding.

Configure the operating system to enable the blocking of unauthorized peripherals with the following command:

Note: This command must be run from a root shell and will create an allow list for any usb devices currently connect to the system.

# usbguard generate-policy --no-hash > /etc/usbguard/rules.conf

Note: Enabling and starting usbguard without properly configuring it for an individual system will immediately prevent any access over a usb device such as a keyboard or mouse.

This requirement applies to wireless peripheral technologies (e.g., wireless mice, keyboards, displays, etc.) used with RHEL 9 systems. Wireless peripherals (e.g., Wi-Fi/Bluetooth/IR keyboards, mice and pointing devices, and near field communications [NFC]) present a unique challenge by creating an open, unsecured port on a computer. Wireless peripherals must meet DOD requirements for wireless data transmission and be approved for use by the Authorizing Official (AO). Even though some wireless peripherals, such as mice and pointing devices, do not ordinarily carry information that need to be protected, modification of communications with these wireless peripherals may be used to compromise the RHEL 9 operating system.

Satisfies: SRG-OS-000095-GPOS-00049, SRG-OS-000300-GPOS-00118

Verify that RHEL 9 disables the ability to load the Bluetooth kernel module with the following command:

$ sudo grep -r bluetooth /etc/modprobe.conf /etc/modprobe.d/* 

blacklist bluetooth

If the command does not return any output, or the line is commented out, and use of Bluetooth is not documented with the information system security officer (ISSO) as an operational requirement, this is a finding.

Configure RHEL 9 to disable the Bluetooth adapter when not in use.

Create or modify the "/etc/modprobe.d/bluetooth.conf" file with the following line:

install bluetooth /bin/false
blacklist bluetooth

Reboot the system for the settings to take effect.

This requirement applies to wireless peripheral technologies (e.g., wireless mice, keyboards, displays, etc.) used with RHEL 9 systems. Wireless peripherals (e.g., Wi-Fi/Bluetooth/IR keyboards, mice and pointing devices, and near field communications [NFC]) present a unique challenge by creating an open, unsecured port on a computer. Wireless peripherals must meet DOD requirements for wireless data transmission and be approved for use by the Authorizing Official (AO). Even though some wireless peripherals, such as mice and pointing devices, do not ordinarily carry information that need to be protected, modification of communications with these wireless peripherals may be used to compromise the RHEL 9 operating system.

Satisfies: SRG-OS-000299-GPOS-00117, SRG-OS-000300-GPOS-00118, SRG-OS-000424-GPOS-00188, SRG-OS-000481-GPOS-00481

Verify there are no wireless interfaces configured on the system with the following command:

Note: This requirement is Not Applicable for systems that do not have physical wireless network radios.

$ nmcli device status

DEVICE                    TYPE            STATE                    CONNECTION
virbr0                      bridge         connected             virbr0
wlp7s0                    wifi              connected            wifiSSID
enp6s0                    ethernet     disconnected        --
p2p-dev-wlp7s0     wifi-p2p     disconnected        --
lo                             loopback    unmanaged           --
virbr0-nic                tun              unmanaged          --

If a wireless interface is configured and has not been documented and approved by the information system security officer (ISSO), this is a finding.

Configure the system to disable all wireless network interfaces with the following command:

$ nmcli radio all off

Any password, no matter how complex, can eventually be cracked; therefore, passwords need to be changed periodically. If the operating system does not limit the lifetime of passwords and force users to change their passwords, there is the risk that the operating system passwords could be compromised.

Setting the password maximum age ensures users are required to periodically change their passwords. Requiring shorter password lifetimes increases the risk of users writing down the password in a convenient location subject to physical compromise.

Verify that RHEL 9 enforces a 60-day maximum password lifetime for new user accounts by running the following command:

$ grep -i pass_max_days /etc/login.defs

PASS_MAX_DAYS 60

If the "PASS_MAX_DAYS" parameter value is greater than "60", or commented out, this is a finding.

Configure RHEL 9 to enforce a 60-day maximum password lifetime.

Add or modify the following line in the "/etc/login.defs" file:

PASS_MAX_DAYS 60

Any password, no matter how complex, can eventually be cracked; therefore, passwords need to be changed periodically. If RHEL 9 does not limit the lifetime of passwords and force users to change their passwords, there is the risk that RHEL 9 passwords could be compromised.

Check whether the maximum time period for existing passwords is restricted to 60 days with the following commands:

$ sudo awk -F: '$5 > 60 {print $1 "" "" $5}' /etc/shadow

$ sudo awk -F: '$5 <= 0 {print $1 "" "" $5}' /etc/shadow

If any results are returned that are not associated with a system account, this is a finding.

Configure noncompliant accounts to enforce a 60-day maximum password lifetime restriction.

passwd -x 60 [user]

If local interactive users are not assigned a valid home directory, there is no place for the storage and control of files they should own.

Verify all local interactive users on RHEL 9 are assigned a home directory upon creation with the following command:

$ grep -i create_home /etc/login.defs

CREATE_HOME yes

If the value for "CREATE_HOME" parameter is not set to "yes", the line is missing, or the line is commented out, this is a finding.

Configure RHEL 9 to assign home directories to all new local interactive users by setting the "CREATE_HOME" parameter in "/etc/login.defs" to "yes" as follows.

CREATE_HOME yes

The umask controls the default access mode assigned to newly created files. A umask of 077 limits new files to mode 600 or less permissive. Although umask can be represented as a four-digit number, the first digit representing special access modes is typically ignored or required to be "0". This requirement applies to the globally configured system defaults and the local interactive user defaults for each account on the system.

Verify that the default umask for all local interactive users is "077".

Identify the locations of all local interactive user home directories by looking at the "/etc/passwd" file.

Check all local interactive user initialization files for interactive users with the following command:

Note: The example is for a system that is configured to create users home directories in the "/home" directory.

# grep -ri umask /home/

/home/wadea/.bash_history:grep -i umask /etc/bashrc /etc/csh.cshrc /etc/profile
/home/wadea/.bash_history:grep -i umask /etc/login.defs

If any local interactive user initialization files are found to have a umask statement that sets a value less restrictive than "077", this is a finding.

Remove the umask statement from all local interactive user's initialization files. 

If the account is for an application, the requirement for a umask less restrictive than "077" can be documented with the information system security officer, but the user agreement for access to the account must specify that the local interactive user must log on to their account first and then switch the user to the application account with the correct option to gain the account's environment variables.

To ensure accountability and prevent unauthenticated access, interactive users must be identified and authenticated to prevent potential misuse and compromise of the system.

Satisfies: SRG-OS-000104-GPOS-00051, SRG-OS-000121-GPOS-00062, SRG-OS-000042-GPOS-00020

Verify that RHEL 9 contains no duplicate UIDs for interactive users with the following command:

$ sudo awk -F ":" 'list[$3]++{print $1, $3}' /etc/passwd 

If output is produced and the accounts listed are interactive user accounts, this is a finding.

Edit the file "/etc/passwd" and provide each interactive user account that has a duplicate UID with a unique UID.

Ensuring shells are not given to system accounts upon login makes it more difficult for attackers to make use of system accounts.

Verify that system accounts must not have an interactive login shell with the following command:

$ awk -F: '($3<1000){print $1 ":" $3 ":" $7}' /etc/passwd

root:0:/bin/bash
bin:1:/sbin/nologin
daemon:2:/sbin/nologin
adm:3:/sbin/nologin
lp:4:/sbin/nologin

Identify the system accounts from this listing that do not have a nologin shell.

If any system account (other than the root account) has a login shell and it is not documented with the information system security officer (ISSO), this is a finding.

Configure RHEL 9 so that all noninteractive accounts on the system do not have an interactive shell assigned to them.

If the system account needs a shell assigned for mission operations, document the need with the information system security officer (ISSO).

Run the following command to disable the interactive shell for a specific noninteractive user account:

Replace  with the user that has a login shell.

$ sudo usermod --shell /sbin/nologin 

Do not perform the steps in this section on the root account. Doing so will cause the system to become inaccessible.

Temporary accounts are privileged or nonprivileged accounts that are established during pressing circumstances, such as new software or hardware configuration or an incident response, where the need for prompt account activation requires bypassing normal account authorization procedures. If any inactive temporary accounts are left enabled on the system and are not either manually removed or automatically expired within 72 hours, the security posture of the system will be degraded and exposed to exploitation by unauthorized users or insider threat actors.

Temporary accounts are different from emergency accounts. Emergency accounts, also known as "last resort" or "break glass" accounts, are local logon accounts enabled on the system for emergency use by authorized system administrators to manage a system when standard logon methods are failing or not available. Emergency accounts are not subject to manual removal or scheduled expiration requirements.

The automatic expiration of temporary accounts may be extended as needed by the circumstances but it must not be extended indefinitely. A documented permanent account should be established for privileged users who need long-term maintenance accounts.

Satisfies: SRG-OS-000123-GPOS-00064, SRG-OS-000002-GPOS-00002

Verify temporary accounts have been provisioned with an expiration date of 72 hours.

For every existing temporary account, run the following command to obtain its account expiration information:

$ sudo chage -l  | grep -i "account expires"

Verify each of these accounts has an expiration date set within 72 hours. 

If any temporary accounts have no expiration date set or do not expire within 72 hours, this is a finding.

Configure the operating system to expire temporary accounts after 72 hours with the following command:

$ sudo chage -E $(date -d +3days +%Y-%m-%d) 

If a user is assigned the Group Identifier (GID) of a group that does not exist on the system, and a group with the GID is subsequently created, the user may have unintended rights to any files associated with the group.

Verify that all RHEL 9 interactive users have a valid GID.

Check that the interactive users have a valid GID with the following command:
 
$ sudo pwck -qr 
 
If the system has any interactive users with duplicate GIDs, this is a finding.

Configure the system so that all GIDs are referenced in "/etc/passwd" are defined in "/etc/group".

Edit the file "/etc/passwd" and ensure that every user's GID is a valid GID.

Inactive identifiers pose a risk to systems and applications because attackers may exploit an inactive identifier and potentially obtain undetected access to the system.

Disabling inactive accounts ensures that accounts which may not have been responsibly removed are not available to attackers who may have compromised their credentials.

Owners of inactive accounts will not notice if unauthorized access to their user account has been obtained.

Verify that RHEL 9 account identifiers (individuals, groups, roles, and devices) are disabled after 35 days of inactivity with the following command:

Check the account inactivity value by performing the following command:

$ sudo grep -i inactive /etc/default/useradd

INACTIVE=35

If "INACTIVE" is set to "-1", a value greater than "35", or is commented out, this is a finding.

Configure RHEL 9 to disable account identifiers after 35 days of inactivity after the password expiration. 

Run the following command to change the configuration for useradd:

$ sudo useradd -D -f 35

The recommendation is 35 days, but a lower value is acceptable.

The executable search path (typically the PATH environment variable) contains a list of directories for the shell to search to find executables. If this path includes the current working directory (other than the users home directory), executables in these directories may be executed instead of system commands.

This variable is formatted as a colon-separated list of directories. If there is an empty entry, such as a leading or trailing colon or two consecutive colons, this is interpreted as the current working directory. If deviations from the default system search path for the local interactive user are required, they must be documented with the information system security officer (ISSO).

Verify that all local interactive user initialization file executable search path statements do not contain statements that will reference a working directory other than user home directories with the following commands:

$ sudo grep -i path= /home/*/.*

/home/[localinteractiveuser]/.bash_profile:PATH=$PATH:$HOME/.local/bin:$HOME/bin

If any local interactive user initialization files have executable search path statements that include directories outside of their home directory and is not documented with the ISSO as an operational requirement, this is a finding.

Edit the local interactive user initialization files to change any PATH variable statements that reference directories other than their home directory.

If a local interactive user requires path variables to reference a directory owned by the application, it must be documented with the ISSO.

If local interactive users are not assigned a valid home directory, there is no place for the storage and control of files they should own.

Verify that interactive users on the system have a home directory assigned with the following command:
 
$ sudo awk -F: '($3>=1000)&&($7 !~ /nologin/){print $1, $3, $6}' /etc/passwd

smithk:x:1000:1000:smithk:/home/smithk:/bin/bash
scsaustin:x:1001:1001:scsaustin:/home/scsaustin:/bin/bash
djohnson:x:1002:1002:djohnson:/home/djohnson:/bin/bash

Inspect the output and verify that all interactive users (normally users with a user identifier (UID) greater that 1000) have a home directory defined.

If users home directory is not defined, this is a finding.

Create and assign home directories to all local interactive users on RHEL 9 that currently do not have a home directory assigned.

If a local interactive user has a home directory defined that does not exist, the user may be given access to the / directory as the current working directory upon logon. This could create a denial of service because the user would not be able to access their logon configuration files, and it may give them visibility to system files they normally would not be able to access.

Verify the assigned home directories of all interactive users on the system exist with the following command:

$ sudo pwck -r 

user 'mailnull': directory 'var/spool/mqueue' does not exist

The output should not return any interactive users.

If users home directory does not exist, this is a finding.

Create home directories to all local interactive users that currently do not have a home directory assigned. Use the following commands to create the user home directory assigned in "/etc/ passwd":

Note: The example will be for the user wadea, who has a home directory of "/home/wadea", a user identifier (UID) of "wadea", and a Group Identifier (GID) of "users assigned" in "/etc/passwd".

$ sudo mkdir /home/wadea 
$ sudo chown wadea /home/wadea
$ sudo chgrp users /home/wadea
$ sudo chmod 0750 /home/wadea

If the Group Identifier (GID) of a local interactive users home directory is not the same as the primary GID of the user, this would allow unauthorized access to the users files, and users that share the same group may not be able to access files that they legitimately should.

Verify the assigned home directory of all local interactive users is group-owned by that user's primary GID with the following command:

Note: This may miss local interactive users that have been assigned a privileged user identifier (UID). Evidence of interactive use may be obtained from a number of log files containing system logon information. The returned directory "/home/wadea" is used as an example.

$ sudo ls -ld $(awk -F: '($3>=1000)&&($7 !~ /nologin/){print $6}' /etc/passwd)

drwxr-x--- 2 wadea admin 4096 Jun 5 12:41 wadea

Check the user's primary group with the following command:

$ sudo grep $(grep wadea /etc/passwd | awk -F: ‘{print $4}') /etc/group

admin:x:250:wadea,jonesj,jacksons

If the user home directory referenced in "/etc/passwd" is not group-owned by that user's primary GID, this is a finding.

Change the group owner of a local interactive user's home directory to the group found in "/etc/passwd". To change the group owner of a local interactive user's home directory, use the following command:

Note: The example will be for the user "wadea", who has a home directory of "/home/wadea", and has a primary group of users.

$ sudo chgrp users /home/wadea

By limiting the number of failed logon attempts, the risk of unauthorized system access via user password guessing, otherwise known as brute-force attacks, is reduced. Limits are imposed by locking the account.

Satisfies: SRG-OS-000329-GPOS-00128, SRG-OS-000021-GPOS-00005

Verify RHEL 9 is configured to lock an account after three unsuccessful logon attempts with the command:

$ grep 'deny =' /etc/security/faillock.conf

deny = 3

If the "deny" option is not set to "3" or less (but not "0"), is missing or commented out, this is a finding.

Configure RHEL 9 to lock an account when three unsuccessful logon attempts occur.

Add/modify the "/etc/security/faillock.conf" file to match the following line:

deny = 3

By limiting the number of failed logon attempts, the risk of unauthorized system access via user password guessing, also known as brute-forcing, is reduced. Limits are imposed by locking the account.

Satisfies: SRG-OS-000329-GPOS-00128, SRG-OS-000021-GPOS-00005

Verify RHEL 9 is configured to lock the root account after three unsuccessful logon attempts with the command:

$ grep even_deny_root /etc/security/faillock.conf

even_deny_root

If the "even_deny_root" option is not set, is missing or commented out, this is a finding.

Configure RHEL 9 to lock out the "root" account after a number of incorrect login attempts using "pam_faillock.so", first enable the feature using the following command:
 
$ sudo authselect enable-feature with-faillock  

 Then edit the "/etc/security/faillock.conf" file as follows:
 
  add or uncomment the following line:
 even_deny_root

By limiting the number of failed logon attempts the risk of unauthorized system access via user password guessing, otherwise known as brute-forcing, is reduced. Limits are imposed by locking the account.

Satisfies: SRG-OS-000329-GPOS-00128, SRG-OS-000021-GPOS-00005

Verify RHEL 9 locks an account after three unsuccessful logon attempts within a period of 15 minutes with the following command:

Note: If the system administrator demonstrates the use of an approved centralized account management method that locks an account after three unsuccessful logon attempts within a period of 15 minutes, this requirement is Not Applicable.

$ grep fail_interval /etc/security/faillock.conf 

fail_interval = 900

If the "fail_interval" option is not set to "900" or less (but not "0"), the line is commented out, or the line is missing, this is a finding.

To configure RHEL 9 to lock out the "root" account after a number of incorrect login attempts within 15 minutes using "pam_faillock.so", enable the feature using the following command:
 
$ authselect enable-feature with-faillock  

Then edit the "/etc/security/faillock.conf" file as follows:

fail_interval = 900

By limiting the number of failed logon attempts the risk of unauthorized system access via user password guessing, otherwise known as brute-forcing, is reduced. Limits are imposed by locking the account.

Satisfies: SRG-OS-000329-GPOS-00128, SRG-OS-000021-GPOS-00005

Verify RHEL 9 is configured to lock an account until released by an administrator after three unsuccessful logon attempts with the command:

$ grep 'unlock_time =' /etc/security/faillock.conf

unlock_time = 0

If the "unlock_time" option is not set to "0", the line is missing, or commented out, this is a finding.

Configure RHEL 9 to lock an account until released by an administrator after three unsuccessful logon attempts with the command:
 
$ authselect enable-feature with-faillock  

Then edit the "/etc/security/faillock.conf" file as follows:

unlock_time = 0

Accounts providing no operational purpose provide additional opportunities for system compromise. Unnecessary accounts include user accounts for individuals not requiring access to the system and application accounts for applications not installed on the system.

Verify that there are no unauthorized interactive user accounts with the following command:

$ less /etc/passwd  

root:x:0:0:root:/root:/bin/bash
...
games:x:12:100:games:/usr/games:/sbin/nologin
scsaustin:x:1001:1001:scsaustin:/home/scsaustin:/bin/bash
djohnson:x:1002:1002:djohnson:/home/djohnson:/bin/bash

Interactive user account, generally will have a user identifier (UID) of 1000 or greater, a home directory in a specific partition, and an interactive shell.

Obtain the list of interactive user accounts authorized to be on the system from the system administrator or information system security officer (ISSO) and compare it to the list of local interactive user accounts on the system.

If there are unauthorized local user accounts on the system, this is a finding.

Remove unauthorized local interactive user accounts with the following command where  is the unauthorized account:

$ sudo userdel  

An account has root authority if it has a user identifier (UID) of "0". Multiple accounts with a UID of "0" afford more opportunity for potential intruders to guess a password for a privileged account. Proper configuration of sudo is recommended to afford multiple system administrators access to root privileges in an accountable manner.

Verify that only the "root" account has a UID "0" assignment with the following command:

$ awk -F: '$3 == 0 {print $1}' /etc/passwd

root

If any accounts other than "root" have a UID of "0", this is a finding.

Change the UID of any account on the system, other than root, that has a UID of "0". 

If the account is associated with system commands or applications, the UID should be changed to one greater than "0" but less than "1000". Otherwise, assign a UID of greater than "1000" that has not already been assigned.

Having lockouts persist across reboots ensures that account is only unlocked by an administrator. If the lockouts did not persist across reboots, an attacker could simply reboot the system to continue brute force attacks against the accounts on the system.

Verify the "/etc/security/faillock.conf" file is configured use a nondefault faillock directory to ensure contents persist after reboot with the following command:

$ grep 'dir =' /etc/security/faillock.conf

dir = /var/log/faillock

If the "dir" option is not set to a nondefault documented tally log directory, is missing or commented out, this is a finding.

Configure RHEL 9 maintain the contents of the faillock directory after a reboot.

Add/modify the "/etc/security/faillock.conf" file to match the following line:

dir = /var/log/faillock

To ensure accountability and prevent unauthenticated access, groups must be identified uniquely to prevent potential misuse and compromise of the system.

Verify that RHEL 9 contains no duplicate GIDs for interactive users with the following command:
 
 $  cut -d : -f 3 /etc/group | uniq -d
 
If the system has duplicate GIDs, this is a finding.

Edit the file "/etc/group" and provide each group that has a duplicate GID with a unique GID.

If user start-up files execute world-writable programs, especially in unprotected directories, they could be maliciously modified to destroy user files or otherwise compromise the system at the user level. If the system is compromised at the user level, it is easier to elevate privileges to eventually compromise the system at the root and network level.

Verify that local initialization files do not execute world-writable programs with the following command:

Note: The example will be for a system that is configured to create user home directories in the "/home" directory.

$ sudo find /home -perm -002 -type f -name ".[^.]*" -exec ls -ld {} \; 

If any local initialization files are found to reference world-writable files, this is a finding.

Set the mode on files being executed by the local initialization files with the following command:

$ sudo chmod 0755 

Terminating an idle interactive command shell user session within a short time period reduces the window of opportunity for unauthorized personnel to take control of it when left unattended in a virtual terminal or physical console.

Satisfies: SRG-OS-000163-GPOS-00072, SRG-OS-000029-GPOS-00010

Verify RHEL 9 is configured to exit interactive command shell user sessions after 10 minutes of inactivity or less with the following command:

$ sudo grep -i tmout /etc/profile /etc/profile.d/*.sh

/etc/profile.d/tmout.sh:declare -xr TMOUT=600

If "TMOUT" is not set to "600" or less in a script located in the "/etc/'profile.d/ directory, is missing or is commented out, this is a finding.

Configure RHEL 9 to exit interactive command shell user sessions after 10 minutes of inactivity.

Add or edit the following line in "/etc/profile.d/tmout.sh":

#!/bin/bash

declare -xr TMOUT=600

Operating system management includes the ability to control the number of users and user sessions that utilize an operating system. Limiting the number of allowed users and sessions per user is helpful in reducing the risks related to denial-of-service (DoS) attacks.

This requirement addresses concurrent sessions for information system accounts and does not address concurrent sessions by single users via multiple system accounts. The maximum number of concurrent sessions must be defined based on mission needs and the operational environment for each system.

Verify RHEL 9 limits the number of concurrent sessions to "10" for all accounts and/or account types with the following command:

$ grep -r -s maxlogins /etc/security/limits.conf /etc/security/limits.d/*.conf

/etc/security/limits.conf:* hard maxlogins 10

This can be set as a global domain (with the * wildcard) but may be set differently for multiple domains.

If the "maxlogins" item is missing, commented out, or the value is set greater than "10" and is not documented with the information system security officer (ISSO) as an operational requirement for all domains that have the "maxlogins" item assigned, this is a finding.

Configure RHEL 9 to limit the number of concurrent sessions to "10" for all accounts and/or account types.

Add the following line to the top of the /etc/security/limits.conf or in a ".conf" file defined in /etc/security/limits.d/:

* hard maxlogins 10

Without auditing of these events, it may be harder or impossible to identify what an attacker did after an attack.

Verify the "/etc/security/faillock.conf" file is configured to log username information when unsuccessful logon attempts occur with the following command:

$ grep audit /etc/security/faillock.conf

audit

If the "audit" option is not set, is missing, or is commented out, this is a finding.

Configure RHEL 9 to log username information when unsuccessful logon attempts occur.

Add/modify the "/etc/security/faillock.conf" file to match the following line:

audit

Increasing the time between a failed authentication attempt and reprompting to enter credentials helps to slow a single-threaded brute force attack.

Verify RHEL 9 enforces a delay of at least four seconds between console logon prompts following a failed logon attempt with the following command:

$ grep -i fail_delay /etc/login.defs

FAIL_DELAY 4

If the value of "FAIL_DELAY" is not set to "4" or greater, or the line is commented out, this is a finding.

Configure the RHEL 9 to enforce a delay of at least four seconds between logon prompts following a failed console logon attempt.

Modify the "/etc/login.defs" file to set the "FAIL_DELAY" parameter to 4 or greater:

FAIL_DELAY 4

The umask controls the default access mode assigned to newly created files. A umask of 077 limits new files to mode 600 or less permissive. Although umask can be represented as a four-digit number, the first digit representing special access modes is typically ignored or required to be "0". This requirement applies to the globally configured system defaults and the local interactive user defaults for each account on the system.

Satisfies: SRG-OS-000480-GPOS-00228, SRG-OS-000480-GPOS-00227

Verify the "umask" setting is configured correctly in the "/etc/bashrc" file with the following command:

Note: If the value of the "umask" parameter is set to "000" "/etc/bashrc" file, the Severity is raised to a CAT I.

$ grep umask /etc/bashrc 

umask 077
umask 077  

If the value for the "umask" parameter is not "077", or the "umask" parameter is missing or is commented out, this is a finding.

Configure RHEL 9 to define default permissions for all authenticated users using the bash shell.

Add or edit the lines for the "umask" parameter in the "/etc/bashrc" file to "077":

umask 077

The umask controls the default access mode assigned to newly created files. A umask of 077 limits new files to mode 600 or less permissive. Although umask can be represented as a four-digit number, the first digit representing special access modes is typically ignored or required to be "0". This requirement applies to the globally configured system defaults and the local interactive user defaults for each account on the system.

Satisfies: SRG-OS-000480-GPOS-00228, SRG-OS-000480-GPOS-00227

Verify the "umask" setting is configured correctly in the "/etc/csh.cshrc" file with the following command:

Note: If the value of the "umask" parameter is set to "000" "/etc/csh.cshrc" file, the Severity is raised to a CAT I.

$ grep umask /etc/csh.cshrc 

umask 077
umask 077  

If the value for the "umask" parameter is not "077", or the "umask" parameter is missing or is commented out, this is a finding.

Configure RHEL 9 to define default permissions for all authenticated users using the c shell.

Add or edit the lines for the "umask" parameter in the "/etc/csh.cshrc" file to "077":

umask 077

Setting the most restrictive default permissions ensures that when new accounts are created, they do not have unnecessary access.

Verify RHEL 9 defines default permissions for all authenticated users in such a way that the user can only read and modify their own files with the following command:

Note: If the value of the "UMASK" parameter is set to "000" in "/etc/login.defs" file, the Severity is raised to a CAT I.

# grep -i umask /etc/login.defs

UMASK 077

If the value for the "UMASK" parameter is not "077", or the "UMASK" parameter is missing or is commented out, this is a finding.

Configure RHEL 9 to define default permissions for all authenticated users in such a way that the user can only read and modify their own files.

Add or edit the lines for the "UMASK" parameter in the "/etc/login.defs" file to "077":

UMASK 077

The umask controls the default access mode assigned to newly created files. A umask of 077 limits new files to mode 600 or less permissive. Although umask can be represented as a four-digit number, the first digit representing special access modes is typically ignored or required to be "0". This requirement applies to the globally configured system defaults and the local interactive user defaults for each account on the system.

Satisfies: SRG-OS-000480-GPOS-00228, SRG-OS-000480-GPOS-00227

Verify the "umask" setting is configured correctly in the "/etc/profile" file with the following command:

Note: If the value of the "umask" parameter is set to "000" "/etc/profile" file, the Severity is raised to a CAT I.

$ grep umask /etc/profile 

umask 077

If the value for the "umask" parameter is not "077", or the "umask" parameter is missing or is commented out, this is a finding.

Configure RHEL 9 to define default permissions for all authenticated users in such a way that the user can only read and modify their own files.

Add or edit the lines for the "umask" parameter in the "/etc/profile" file to "077":

umask 077

Users need to be aware of activity that occurs regarding their account. Providing users with information regarding the number of unsuccessful attempts that were made to login to their account allows the user to determine if any unauthorized activity has occurred and gives them an opportunity to notify administrators.

Verify users are provided with feedback on when account accesses last occurred with the following command:

$ sudo grep pam_lastlog /etc/pam.d/postlogin

session required pam_lastlog.so showfailed

If "pam_lastlog" is missing from "/etc/pam.d/postlogin" file, or the silent option is present, this is a finding.

Configure RHEL 9 to provide users with feedback on when account accesses last occurred by setting the required configuration options in "/etc/pam.d/postlogin". 

Add the following line to the top of "/etc/pam.d/postlogin":

session required pam_lastlog.so showfailed

Terminating an idle session within a short time period reduces the window of opportunity for unauthorized personnel to take control of a management session enabled on the console or console port that has been left unattended.

Verify that RHEL 9 logs out sessions that are idle for 15 minutes with the following command:

$ sudo grep -i ^StopIdleSessionSec /etc/systemd/logind.conf

StopIdleSessionSec=900

If "StopIdleSessionSec" is not configured to "900" seconds, this is a finding.

Configure RHEL 9 to log out idle sessions by editing the /etc/systemd/logind.conf file with the following line:

StopIdleSessionSec=900

The "logind" service must be restarted for the changes to take effect. To restart the "logind" service, run the following command:

$ sudo systemctl restart systemd-logind

Without verification of the security functions, security functions may not operate correctly and the failure may go unnoticed. Security function is defined as the hardware, software, and/or firmware of the information system responsible for enforcing the system security policy and supporting the isolation of code and data on which the protection is based. Security functionality includes, but is not limited to, establishing system accounts, configuring access authorizations (i.e., permissions, privileges), setting events to be audited, and setting intrusion detection parameters.

This requirement applies to operating systems performing security function verification/testing and/or systems and environments that require this functionality.

Satisfies: SRG-OS-000445-GPOS-00199, SRG-OS-000134-GPOS-00068

Ensure that RHEL 9 verifies correct operation of security functions through the use of SELinux with the following command:

$ getenforce

Enforcing

If SELINUX is not set to "Enforcing", this is a finding.

Verify that SELinux is configured to be enforcing at boot.

grep "SELINUX=" /etc/selinux/config
# SELINUX= can take one of these three values:
# NOTE: In earlier Fedora kernel builds, SELINUX=disabled would also
SELINUX=enforcing

If SELINUX line is missing, commented out, or not set to "enforcing", this is a finding.

Configure RHEL 9 to verify correct operation of security functions.

Edit the file "/etc/selinux/config" and add or modify the following line:

 SELINUX=enforcing 

A reboot is required for the changes to take effect.

Setting the SELinux policy to "targeted" or a more specialized policy ensures the system will confine processes that are likely to be targeted for exploitation, such as network or system services.

Note: During the development or debugging of SELinux modules, it is common to temporarily place nonproduction systems in "permissive" mode. In such temporary cases, SELinux policies should be developed, and once work is completed, the system should be reconfigured to "targeted".

Verify the SELINUX on RHEL 9 is using the targeted policy with the following command:

$ sestatus | grep policy

Loaded policy name:             targeted

If the loaded policy name is not "targeted", this is a finding.

Configure RHEL 9 to use the targetd SELINUX policy.

Edit the file "/etc/selinux/config" and add or modify the following line:

 SELINUXTYPE=targeted 

A reboot is required for the changes to take effect.

Not having the correct SELinux context on the faillock directory may lead to unauthorized access to the directory.

Verify the location of the nondefault tally directory for the pam_faillock module with the following command:

Note: If the system does not have SELinux enabled and enforcing a targeted policy, or if the pam_faillock module is not configured for use, this requirement is Not Applicable.

$ grep 'dir =' /etc/security/faillock.conf

dir = /var/log/faillock

Check the security context type of the nondefault tally directory with the following command:

$ ls -Zd /var/log/faillock

unconfined_u:object_r:faillog_t:s0 /var/log/faillock

If the security context type of the nondefault tally directory is not "faillog_t", this is a finding.

Configure RHEL 9 to allow the use of a nondefault faillock tally directory while SELinux enforces a targeted policy.

Create a nondefault faillock tally directory (if it does not already exist) with the following example:

$ sudo mkdir /var/log/faillock

Update the /etc/selinux/targeted/contexts/files/file_contexts.local with "faillog_t" context type for the nondefault faillock tally directory with the following command:

$ sudo semanage fcontext -a -t faillog_t "/var/log/faillock(/.*)?"

Next, update the context type of the nondefault faillock directory/subdirectories and files with the following command:

$ sudo restorecon -R -v /var/log/faillock

Without verification of the security functions, security functions may not operate correctly and the failure may go unnoticed. Security function is defined as the hardware, software, and/or firmware of the information system responsible for enforcing the system security policy and supporting the isolation of code and data on which the protection is based. Security functionality includes, but is not limited to, establishing system accounts, configuring access authorizations (i.e., permissions, privileges), setting events to be audited, and setting intrusion detection parameters.

Policycoreutils contains the policy core utilities that are required for basic operation of an SELinux-enabled system. These utilities include load_policy to load SELinux policies, setfile to label filesystems, newrole to switch roles, and run_init to run /etc/init.d scripts in the proper context.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000134-GPOS-00068

Verify RHEL 9 has the policycoreutils package installed with the following command:

$ sudo dnf list --installed policycoreutils

Example output:

policycoreutils.x86_64          3.3-6.el9_0                                                 

If the "policycoreutils" package is not installed, this is a finding.

The policycoreutils package can be installed with the following command:
 
$ sudo dnf install policycoreutils

The policycoreutils-python-utils package is required to operate and manage an SELinux environment and its policies. It provides utilities such as semanage, audit2allow, audit2why, chcat, and sandbox.

Verify that RHEL 9 policycoreutils-python-utils service package is installed with the following command:

$ sudo dnf list --installed policycoreutils-python-utils

Example output:

policycoreutils-python-utils.noarch          3.3-6.el9_0

If the "policycoreutils-python-utils" package is not installed, this is a finding.

Install the policycoreutils-python-utils service package (if the policycoreutils-python-utils service is not already installed) with the following command:

$ sudo dnf install policycoreutils-python-utils

"sudo" is a program designed to allow a system administrator to give limited root privileges to users and log root activity. The basic philosophy is to give as few privileges as possible but still allow system users to get their work done.

Verify that RHEL 9 sudo package is installed with the following command:

$ sudo dnf list --installed sudo

Example output:

sudo.x86_64          1.9.5p2-7.el9

If the "sudo" package is not installed, this is a finding.

The  sudo  package can be installed with the following command:
 
$ sudo dnf install sudo

Without reauthentication, users may access resources or perform tasks for which they do not have authorization. 

When operating systems provide the capability to escalate a functional capability, it is critical the organization requires the user to reauthenticate when using the "sudo" command.

If the value is set to an integer less than "0", the user's time stamp will not expire and the user will not have to reauthenticate for privileged actions until the user's session is terminated.

Verify RHEL 9 requires reauthentication when using the "sudo" command to elevate privileges with the following command:

$ sudo grep -ir 'timestamp_timeout' /etc/sudoers /etc/sudoers.d/*

/etc/sudoers:Defaults timestamp_timeout=0

If results are returned from more than one file location, this is a finding.

If "timestamp_timeout" is set to a negative number, is commented out, or no results are returned, this is a finding.

Configure RHEL 9 to reauthenticate "sudo" commands after the specified timeout:

Add the following line to "/etc/sudoers":

Defaults timestamp_timeout=0

If the rootpw, targetpw, or runaspw flags are defined and not disabled, by default the operating system will prompt the invoking user for the "root" user password.

Verify that the sudoers security policy is configured to use the invoking user's password for privilege escalation with the following command:

$ sudo egrep -ir '(!rootpw|!targetpw|!runaspw)' /etc/sudoers /etc/sudoers.d/* | grep -v '#'

/etc/sudoers:Defaults !targetpw
/etc/sudoers:Defaults !rootpw
/etc/sudoers:Defaults !runaspw

If no results are returned, this is a finding.

If results are returned from more than one file location, this is a finding.

If "Defaults !targetpw" is not defined, this is a finding.

If "Defaults !rootpw" is not defined, this is a finding.

If "Defaults !runaspw" is not defined, this is a finding.

Define the following in the Defaults section of the /etc/sudoers file or a single configuration file in the /etc/sudoers.d/ directory:

Defaults !targetpw
Defaults !rootpw
Defaults !runaspw

Without reauthentication, users may access resources or perform tasks for which they do not have authorization.

When operating systems provide the capability to escalate a functional capability, it is critical that the user reauthenticate.

Satisfies: SRG-OS-000373-GPOS-00156, SRG-OS-000373-GPOS-00157, SRG-OS-000373-GPOS-00158

Verify that "/etc/sudoers" has no occurrences of "!authenticate" with the following command:

$ sudo grep -ir '!authenticate' /etc/sudoers /etc/sudoers.d/*

If any occurrences of "!authenticate" are returned, this is a finding.

Configure RHEL 9 to not allow users to execute privileged actions without authenticating.

Remove any occurrence of "!authenticate" found in "/etc/sudoers" file or files in the "/etc/sudoers.d" directory.

$ sudo sed -i '/\!authenticate/ s/^/# /g' /etc/sudoers /etc/sudoers.d/*

If the "sudoers" file is not configured correctly, any user defined on the system can initiate privileged actions on the target system.

Verify RHEL 9 restricts privilege elevation to authorized personnel with the following command:

$ sudo sh -c 'grep -iw ALL /etc/sudoers /etc/sudoers.d/*'

If the either of the following entries are returned, this is a finding:
ALL     ALL=(ALL) ALL
ALL     ALL=(ALL:ALL) ALL

Remove the following entries from the /etc/sudoers file or configuration file under /etc/sudoers.d/:

ALL     ALL=(ALL) ALL
ALL     ALL=(ALL:ALL) ALL

The "su" program allows to run commands with a substitute user and group ID. It is commonly used to run commands as the root user. Limiting access to such commands is considered a good security practice.

Satisfies: SRG-OS-000373-GPOS-00156, SRG-OS-000312-GPOS-00123

Verify that RHEL 9 requires uses to be members of the "wheel" group with the following command:

$ grep pam_wheel /etc/pam.d/su 

auth             required        pam_wheel.so use_uid 

If a line for "pam_wheel.so" does not exist, or is commented out, this is a finding.

Configure RHEL 9 to require users to be in the "wheel" group to run "su" command.

In file "/etc/pam.d/su", uncomment the following line:

"#auth    required    pam_wheel.so use_uid"

$ sed '/^[[:space:]]*#[[:space:]]*auth[[:space:]]\+required[[:space:]]\+pam_wheel\.so[[:space:]]\+use_uid$/s/^[[:space:]]*#//' -i /etc/pam.d/su

If necessary, create a "wheel" group and add administrative users to the group.

The organization must identify authorized software programs and permit execution of authorized software. The process used to identify software programs that are authorized to execute on organizational information systems is commonly referred to as allowlisting.

Utilizing an allowlist provides a configuration management method for allowing the execution of only authorized software. Using only authorized software decreases risk by limiting the number of potential vulnerabilities. Verification of allowlisted software occurs prior to execution or at system startup.

User home directories/folders may contain information of a sensitive nature. Nonprivileged users should coordinate any sharing of information with an SA through shared resources.

RHEL 9 ships with many optional packages. One such package is a file access policy daemon called "fapolicyd". "fapolicyd" is a userspace daemon that determines access rights to files based on attributes of the process and file. It can be used to either blocklist or allowlist processes or file access.

Proceed with caution with enforcing the use of this daemon. Improper configuration may render the system nonfunctional. The "fapolicyd" API is not namespace aware and can cause issues when launching or running containers.

Satisfies: SRG-OS-000370-GPOS-00155, SRG-OS-000368-GPOS-00154

Verify that RHEL 9 fapolicyd package is installed with the following command:

$ sudo dnf list --installed fapolicyd

Example output:

fapolicyd.x86_64          1.1-103.el9_0

If the "fapolicyd" package is not installed, this is a finding.

The  fapolicyd  package can be installed with the following command:
 
$ sudo dnf install fapolicyd

The organization must identify authorized software programs and permit execution of authorized software. The process used to identify software programs that are authorized to execute on organizational information systems is commonly referred to as allowlisting.

Utilizing an allowlist provides a configuration management method for allowing the execution of only authorized software. Using only authorized software decreases risk by limiting the number of potential vulnerabilities. Verification of allowlisted software occurs prior to execution or at system startup.

User home directories/folders may contain information of a sensitive nature. Nonprivileged users should coordinate any sharing of information with an SA through shared resources.

RHEL 9 ships with many optional packages. One such package is a file access policy daemon called "fapolicyd". "fapolicyd" is a userspace daemon that determines access rights to files based on attributes of the process and file. It can be used to either blocklist or allowlist processes or file access.

Proceed with caution with enforcing the use of this daemon. Improper configuration may render the system nonfunctional. The "fapolicyd" API is not namespace aware and can cause issues when launching or running containers.

Satisfies: SRG-OS-000370-GPOS-00155, SRG-OS-000368-GPOS-00154

Verify that RHEL 9 fapolicyd is active with the following command:

$ systemctl is-active fapolicyd

active

If fapolicyd module is not active, this is a finding.

Enable the fapolicyd with the following command:

$ systemctl enable --now fapolicyd

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. "pwquality" enforces complex password construction configuration and has the ability to limit brute-force attacks on the system.

RHEL 9 uses "pwquality" as a mechanism to enforce password complexity. This is set in both:
/etc/pam.d/password-auth
/etc/pam.d/system-auth

By limiting the number of attempts to meet the pwquality module complexity requirements before returning with an error, the system will audit abnormal attempts at password changes.

Verify RHEL 9 is configured to limit the "pwquality" retry option to "3". 

Check for the use of the "pwquality" retry option in the system-auth file with the following command:

$ cat /etc/pam.d/system-auth | grep pam_pwquality

password required pam_pwquality.so retry=3

If the value of "retry" is set to "0" or greater than "3", or is missing, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to limit the "pwquality" retry option to "3".

Add the following line to the "/etc/pam.d/system-auth" file (or modify the line to have the required value):

password required pam_pwquality.so retry=3

If an account has an empty password, anyone could log in and run commands with the privileges of that account. Accounts with empty passwords should never be used in operational environments.

Verify that null passwords cannot be used with the following command:

$ sudo grep -i nullok /etc/pam.d/system-auth /etc/pam.d/password-auth

If output is produced, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Remove any instances of the "nullok" option in the "/etc/pam.d/password-auth" and "/etc/pam.d/system-auth" files to prevent logons with empty passwords.

Note: Manual changes to the listed file may be overwritten by the "authselect" program.

If the pam_faillock.so module is not loaded, the system will not correctly lockout accounts to prevent password guessing attacks.

Verify the pam_faillock.so module is present in the "/etc/pam.d/system-auth" file:

$ grep pam_faillock.so /etc/pam.d/system-auth

auth required pam_faillock.so preauth
auth required pam_faillock.so authfail
account required pam_faillock.so

If the pam_faillock.so module is not present in the "/etc/pam.d/system-auth" file with the "preauth" line listed before pam_unix.so, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to include the use of the pam_faillock.so module in the /etc/pam.d/system-auth file.

Add/modify the appropriate sections of the "/etc/pam.d/system-auth" file to match the following lines:
Note: The "preauth" line must be listed before pam_unix.so.

auth required pam_faillock.so preauth
auth required pam_faillock.so authfail
account required pam_faillock.so
auth required pam_deny.so

If the pam_faillock.so module is not loaded, the system will not correctly lockout accounts to prevent password guessing attacks.

Verify the pam_faillock.so module is present in the "/etc/pam.d/password-auth" file:

$ grep pam_faillock.so /etc/pam.d/password-auth

auth required pam_faillock.so preauth
auth required pam_faillock.so authfail
account required pam_faillock.so

If the pam_faillock.so module is not present in the "/etc/pam.d/password-auth" file with the "preauth" line listed before pam_unix.so, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to include the use of the pam_faillock.so module in the /etc/pam.d/password-auth file.

Add/modify the appropriate sections of the "/etc/pam.d/password-auth" file to match the following lines:
Note: The "preauth" line must be listed before pam_unix.so.

auth required pam_faillock.so preauth
auth required pam_faillock.so authfail
account required pam_faillock.so

Enabling PAM password complexity permits enforcement of strong passwords and consequently makes the system less prone to dictionary attacks.

Satisfies: SRG-OS-000069-GPOS-00037, SRG-OS-000070-GPOS-00038, SRG-OS-000480-GPOS-00227

Verify RHEL 9 uses "pwquality" to enforce the password complexity rules in the password-auth file with the following command:

$ cat /etc/pam.d/password-auth | grep pam_pwquality

password required pam_pwquality.so 

If the command does not return a line containing the value "pam_pwquality.so", or the line is commented out, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to use "pwquality" to enforce password complexity rules.

Add the following line to the "/etc/pam.d/password-auth" file (or modify the line to have the required value):

password required pam_pwquality.so

Enabling PAM password complexity permits enforcement of strong passwords and consequently makes the system less prone to dictionary attacks.

Verify RHEL 9 uses "pwquality" to enforce the password complexity rules in the system-auth file with the following command:

$ cat /etc/pam.d/system-auth | grep pam_pwquality

password required pam_pwquality.so 

If the command does not return a line containing the value "pam_pwquality.so", or the line is commented out, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to use "pwquality" to enforce password complexity rules.

Add the following line to the "/etc/pam.d/system-auth" file(or modify the line to have the required value):

password required pam_pwquality.so

Passwords need to be protected at all times, and encryption is the standard method for protecting passwords. If passwords are not encrypted, they can be plainly read (i.e., clear text) and easily compromised. Passwords that are encrypted with a weak algorithm are no more protected than if they are kept in plain text.

Using more hashing rounds makes password cracking attacks more difficult.

Satisfies: SRG-OS-000073-GPOS-00041, SRG-OS-000120-GPOS-00061

Verify the number of rounds for the password hashing algorithm is configured with the following command:

$ sudo grep rounds /etc/pam.d/password-auth

password sufficient pam_unix.so sha512 rounds=5000

If a matching line is not returned or "rounds" is less than "5000", this a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure Red Hat Enterprise Linux 9 to use 5000 hashing rounds for hashing passwords.

Add or modify the following line in "/etc/pam.d/password-auth" and set "rounds" to "5000".

password sufficient pam_unix.so sha512 rounds=5000

Passwords need to be protected at all times, and encryption is the standard method for protecting passwords. If passwords are not encrypted, they can be plainly read (i.e., clear text) and easily compromised. Passwords that are encrypted with a weak algorithm are no more protected than if they are kept in plain text.

Using more hashing rounds makes password cracking attacks more difficult.

Satisfies: SRG-OS-000073-GPOS-00041, SRG-OS-000120-GPOS-00061

Verify the number of rounds for the password hashing algorithm is configured with the following command:

$ sudo grep rounds /etc/pam.d/system-auth

password sufficient pam_unix.so sha512 rounds=5000

If a matching line is not returned or "rounds" is less than 5000, this a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure Red Hat Enterprise Linux 9 to use 5000 hashing rounds for hashing passwords.

Add or modify the following line in "/etc/pam.d/system-auth" and set "rounds" to 5000.

password sufficient pam_unix.so sha512 rounds=5000'

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks.

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised.

Satisfies: SRG-OS-000072-GPOS-00040, SRG-OS-000071-GPOS-00039, SRG-OS-000070-GPOS-00038, SRG-OS-000266-GPOS-00101, SRG-OS-000078-GPOS-00046, SRG-OS-000480-GPOS-00225, SRG-OS-000069-GPOS-00037

Verify that RHEL 9 enforces password complexity rules for the root account.

Check if root user is required to use complex passwords with the following command:

$ grep enforce_for_root /etc/security/pwquality.conf /etc/security/pwquality.conf/*.conf

/etc/security/pwquality.conf:enforce_for_root

If "enforce_for_root" is commented or missing, this is a finding.

Configure RHEL 9 to enforce password complexity on the root account.

Add or update the following line in /etc/security/pwquality.conf:

enforce_for_root

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. 

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised. Requiring a minimum number of lowercase characters makes password guessing attacks more difficult by ensuring a larger search space.

Verify that RHEL 9 enforces password complexity by requiring at least one lowercase character.

Check the value for "lcredit" with the following command:

$ sudo grep lcredit /etc/security/pwquality.conf /etc/security/pwquality.conf/*.conf 

/etc/security/pwquality.conf:lcredit = -1 

If the value of "lcredit" is a positive number or is commented out, this is a finding.

Configure RHEL 9 to enforce password complexity by requiring at least one lowercase character be used by setting the "lcredit" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

lcredit = -1

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks.

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised. Requiring digits makes password guessing attacks more difficult by ensuring a larger search space.

Verify that RHEL 9 enforces password complexity by requiring at least one numeric character.

Check the value for "dcredit" with the following command:

$ sudo grep dcredit /etc/security/pwquality.conf /etc/security/pwquality.conf/*.conf

/etc/security/pwquality.conf:dcredit = -1 

If the value of "dcredit" is a positive number or is commented out, this is a finding.

Configure RHEL 9 to enforce password complexity by requiring at least one numeric character be used by setting the "dcredit" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

dcredit = -1

Enforcing a minimum password lifetime helps to prevent repeated password changes to defeat the password reuse or history enforcement requirement. If users are allowed to immediately and continually change their password, then the password could be repeatedly changed in a short period of time to defeat the organization's policy regarding password reuse.

Setting the minimum password age protects against users cycling back to a favorite password after satisfying the password reuse requirement.

Verify RHEL 9 enforces 24 hours as the minimum password lifetime for new user accounts.

Check for the value of "PASS_MIN_DAYS" in "/etc/login.defs" with the following command: 

$ grep -i pass_min_days /etc/login.defs

PASS_MIN_DAYS 1

If the "PASS_MIN_DAYS" parameter value is not "1" or greater, or is commented out, this is a finding.

Configure RHEL 9 to enforce 24 hours as the minimum password lifetime.

Add the following line in "/etc/login.defs" (or modify the line to have the required value):

PASS_MIN_DAYS 1

Enforcing a minimum password lifetime helps to prevent repeated password changes to defeat the password reuse or history enforcement requirement. If users are allowed to immediately and continually change their password, the password could be repeatedly changed in a short period of time to defeat the organization's policy regarding password reuse.

Verify that RHEL 9 has configured the minimum time period between password changes for each user account as one day or greater with the following command:

$ sudo awk -F: '$4 < 1 {print $1 " " $4}' /etc/shadow

If any results are returned that are not associated with a system account, this is a finding.

Configure noncompliant accounts to enforce a 24 hour minimum password lifetime:

$ sudo passwd -n 1 [user]

Without reauthentication, users may access resources or perform tasks for which they do not have authorization.

When operating systems provide the capability to escalate a functional capability, it is critical that the user reauthenticate.

Satisfies: SRG-OS-000373-GPOS-00156, SRG-OS-000373-GPOS-00157, SRG-OS-000373-GPOS-00158

Verify that "/etc/sudoers" has no occurrences of "NOPASSWD" with the following command:

$ sudo grep -ri nopasswd /etc/sudoers /etc/sudoers.d/*

If any occurrences of "NOPASSWD" are returned from the command and have not been documented with the information system security officer (ISSO) as an organizationally defined administrative group utilizing MFA, this is a finding.

Configure RHEL 9 to not allow users to execute privileged actions without authenticating with a password.

Remove any occurrence of "NOPASSWD" found in "/etc/sudoers" file or files in the "/etc/sudoers.d" directory.

$ sudo sed -i '/NOPASSWD/ s/^/# /g' /etc/sudoers /etc/sudoers.d/*

The shorter the password, the lower the number of possible combinations that need to be tested before the password is compromised.

Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. Password length is one factor of several that helps to determine strength and how long it takes to crack a password. Use of more characters in a password helps to increase exponentially the time and/or resources required to compromise the password.

RHEL 9 uses "pwquality" as a mechanism to enforce password complexity. Configurations are set in the "etc/security/pwquality.conf" file.

The "minlen", sometimes noted as minimum length, acts as a "score" of complexity based on the credit components of the "pwquality" module. By setting the credit components to a negative value, not only will those components be required, but they will not count toward the total "score" of "minlen". This will enable "minlen" to require a 15-character minimum.

The DOD minimum password requirement is 15 characters.

Verify that RHEL 9 enforces a minimum 15-character password length with the following command:

$ grep minlen /etc/security/pwquality.conf

minlen = 15

If the command does not return a "minlen" value of "15" or greater, does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to enforce a minimum 15-character password length.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

minlen = 15

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised. RHEL 9 utilizes "pwquality" as a mechanism to enforce password complexity. Note that to require special characters without degrading the "minlen" value, the credit value must be expressed as a negative number in "/etc/security/pwquality.conf".

Verify that RHEL 9 enforces password complexity by requiring at least one special character with the following command:

$ sudo grep ocredit /etc/security/pwquality.conf /etc/security/pwquality.conf.d/*.conf 

ocredit = -1 

If the value of "ocredit" is a positive number or is commented out, this is a finding.

Configure RHEL 9 to enforce password complexity by requiring at least one special character be used by setting the "ocredit" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

ocredit = -1

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. If RHEL 9 allows the user to select passwords based on dictionary words, this increases the chances of password compromise by increasing the opportunity for successful guesses, and brute-force attacks.

Verify RHEL 9 prevents the use of dictionary words for passwords with the following command:

$ sudo grep dictcheck /etc/security/pwquality.conf /etc/pwquality.conf.d/*.conf 

/etc/security/pwquality.conf:dictcheck=1 

If "dictcheck" does not have a value other than "0", or is commented out, this is a finding.

Configure RHEL 9 to prevent the use of dictionary words for passwords.

Add or update the following line in the "/etc/security/pwquality.conf" file or a configuration file in the /etc/pwquality.conf.d/ directory to contain the "dictcheck" parameter:

dictcheck=1

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised. Requiring a minimum number of uppercase characters makes password guessing attacks more difficult by ensuring a larger search space.

Verify that RHEL 9 enforces password complexity by requiring that at least one uppercase character.

Check the value for "ucredit" with the following command:

$ sudo grep ucredit /etc/security/pwquality.conf /etc/security/pwquality.conf/*.conf 

ucredit = -1 

If the value of "ucredit" is a positive number or is commented out, this is a finding.

Configure RHEL 9 to enforce password complexity by requiring that at least one uppercase character be used by setting the "ucredit" option.

Add the following line to /etc/security/pwquality.conf (or modify the line to have the required value):

ucredit = -1

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks. 

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex the password, the greater the number of possible combinations that need to be tested before the password is compromised. Requiring a minimum number of different characters during password changes ensures that newly changed passwords will not resemble previously compromised ones. Note that passwords changed on compromised systems will still be compromised.

Verify the value of the "difok" option in "/etc/security/pwquality.conf" with the following command:

$ sudo grep difok /etc/security/pwquality.conf

difok = 8
 
If the value of "difok" is set to less than "8", or is commented out, this is a finding.

Configure RHEL 9 to require the change of at least eight of the total number of characters when passwords are changed by setting the "difok" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

difok = 8

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks.

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex a password, the greater the number of possible combinations that need to be tested before the password is compromised.

Verify the value of the "maxclassrepeat" option in "/etc/security/pwquality.conf" with the following command:

$ grep maxclassrepeat /etc/security/pwquality.conf 

maxclassrepeat = 4

If the value of "maxclassrepeat" is set to "0", more than "4", or is commented out, this is a finding.

Configure RHEL 9 to require the change of the number of repeating characters of the same character class when passwords are changed by setting the "maxclassrepeat" option.

Add the following line to "/etc/security/pwquality.conf" conf (or modify the line to have the required value):

maxclassrepeat = 4

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks.

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex a password, the greater the number of possible combinations that need to be tested before the password is compromised.

Verify the value of the "maxrepeat" option in "/etc/security/pwquality.conf" with the following command:

$ grep maxrepeat /etc/security/pwquality.conf 

maxrepeat = 3

If the value of "maxrepeat" is set to more than "3", or is commented out, this is a finding.

Configure RHEL 9 to require the change of the number of repeating consecutive characters when passwords are changed by setting the "maxrepeat" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

maxrepeat = 3

Use of a complex password helps to increase the time and resources required to compromise the password. Password complexity, or strength, is a measure of the effectiveness of a password in resisting attempts at guessing and brute-force attacks.

Password complexity is one factor of several that determines how long it takes to crack a password. The more complex a password, the greater the number of possible combinations that need to be tested before the password is compromised.

Verify the value of the "minclass" option in "/etc/security/pwquality.conf" with the following command:

$ grep minclass /etc/security/pwquality.conf
 
minclass = 4

If the value of "minclass" is set to less than "4", or is commented out, this is a finding.

Configure RHEL 9 to require the change of at least four character classes when passwords are changed by setting the "minclass" option.

Add the following line to "/etc/security/pwquality.conf" (or modify the line to have the required value):

minclass = 4

Passwords need to be protected at all times, and encryption is the standard method for protecting passwords. If passwords are not encrypted, they can be plainly read (i.e., clear text) and easily compromised. Passwords that are encrypted with a weak algorithm are no more protected than if they are kept in plain text.

This setting ensures user and group account administration utilities are configured to store only encrypted representations of passwords. Additionally, the "crypt_style" configuration option ensures the use of a strong hashing algorithm that makes password cracking attacks more difficult.

Verify the user and group account administration utilities are configured to store only encrypted representations of passwords with the following command:

# grep crypt /etc/libuser.conf 

crypt_style = sha512

If the "crypt_style" variable is not set to "sha512", is not in the defaults section, is commented out, or does not exist, this is a finding.

Configure RHEL 9 to use the SHA-512 algorithm for password hashing.

Add or change the following line in the "[defaults]" section of "/etc/libuser.conf" file:

crypt_style = sha512

Passwords need to be protected at all times, and encryption is the standard method for protecting passwords. If passwords are not encrypted, they can be plainly read (i.e., clear text) and easily compromised. Passwords that are encrypted with a weak algorithm are no more protected than if they are kept in plain text.

This setting ensures user and group account administration utilities are configured to store only encrypted representations of passwords. Additionally, the "crypt_style" configuration option ensures the use of a strong hashing algorithm that makes password cracking attacks more difficult.

Verify the system's shadow file is configured to store only encrypted representations of passwords with a hash value of SHA512 with the following command:

# grep -i encrypt_method /etc/login.defs

ENCRYPT_METHOD SHA512

If "ENCRYPT_METHOD" does not have a value of "SHA512", or the line is commented out, this is a finding.

Configure RHEL 9 to store only SHA512 encrypted representations of passwords.

Add or update the following line in the "/etc/login.defs" file:

ENCRYPT_METHOD SHA512

Without reauthentication, users may access resources or perform tasks for which they do not have authorization. When operating systems provide the capability to escalate a functional capability, it is critical the user reauthenticate.

Satisfies: SRG-OS-000373-GPOS-00156, SRG-OS-000373-GPOS-00157, SRG-OS-000373-GPOS-00158

Verify the operating system is not configured to bypass password requirements for privilege escalation with the following command:

$ sudo grep pam_succeed_if /etc/pam.d/sudo 

If any occurrences of "pam_succeed_if" are returned, this is a finding.

Configure the operating system to require users to supply a password for privilege escalation.

Remove any occurrences of " pam_succeed_if " in the  "/etc/pam.d/sudo" file.

If an account has an empty password, anyone could log in and run commands with the privileges of that account. Accounts with empty passwords should never be used in operational environments.

Verify that null or blank passwords cannot be used with the following command:

$ sudo awk -F: '!$2 {print $1}' /etc/shadow

If the command returns any results, this is a finding.

Configure all accounts on RHEL 9 to have a password or lock the account with the following commands:

Perform a password reset:

$ sudo passwd [username] 

To lock an account:

$ sudo passwd -l [username]

Smart card login provides two-factor authentication stronger than that provided by a username and password combination. Smart cards leverage public key infrastructure to provide and verify credentials. Configuring the smart card driver in use by the organization helps to prevent users from using unauthorized smart cards.

Satisfies: SRG-OS-000104-GPOS-00051, SRG-OS-000106-GPOS-00053, SRG-OS-000107-GPOS-00054, SRG-OS-000109-GPOS-00056, SRG-OS-000108-GPOS-00055, SRG-OS-000112-GPOS-00057, SRG-OS-000113-GPOS-00058

Verify that RHEL loads the CAC driver with the following command:

$ grep card_drivers /etc/opensc.conf

card_drivers = cac; 

If "cac" is not listed as a card driver, or there is no line returned for "card_drivers", this is a finding.

Configure RHEL 9 to load the CAC driver.

Add or modify the following line in the "/etc/opensc.conf" file:

card_drivers = cac;

Without the use of multifactor authentication, the ease of access to privileged functions is greatly increased. Multifactor authentication requires using two or more factors to achieve authentication. A privileged account is defined as an information system account with authorizations of a privileged user. The DOD Common Access Card (CAC) with DOD-approved PKI is an example of multifactor authentication.

Satisfies: SRG-OS-000375-GPOS-00160, SRG-OS-000105-GPOS-00052

Verify that RHEL 9 has smart cards are enabled in System Security Services Daemon (SSSD), run the following command:

$ sudo grep pam_cert_auth /etc/sssd/sssd.conf

pam_cert_auth = True 

If "pam_cert_auth" is not set to "True", the line is commented out, or the line is missing, this is a finding.

Note: If the system administrator (SA) demonstrates the use of an approved alternate multifactor authentication method, this requirement is not applicable.

Edit the file "/etc/sssd/sssd.conf" and add or edit the following line:

pam_cert_auth = True

Using an authentication device, such as a DOD common access card (CAC) or token that is separate from the information system, ensures that even if the information system is compromised, credentials stored on the authentication device will not be affected.

Multifactor solutions that require devices separate from information systems gaining access include, for example, hardware tokens providing time-based or challenge-response authenticators and smart cards such as the U.S. Government Personal Identity Verification (PIV) card and the DOD CAC.

RHEL 9 includes multiple options for configuring certificate status checking, but for this requirement focuses on the System Security Services Daemon (SSSD). By default, SSSD performs Online Certificate Status Protocol (OCSP) checking and certificate verification using a sha256 digest function.

Satisfies: SRG-OS-000375-GPOS-00160, SRG-OS-000377-GPOS-00162

Verify the operating system implements Online Certificate Status Protocol (OCSP) and is using the proper digest value on the system with the following command:

$ sudo grep certificate_verification /etc/sssd/sssd.conf /etc/sssd/conf.d/*.conf | grep -v "^#"

certificate_verification = ocsp_dgst=sha512

If the certificate_verification line is missing from the [sssd] section, or is missing "ocsp_dgst=sha512", ask the administrator to indicate what type of multifactor authentication is being utilized and how the system implements certificate status checking. If there is no evidence of certificate status checking being used, this is a finding.

Note: If the system administrator (SA) demonstrates the use of an approved alternate multifactor authentication method, this requirement is not applicable.

Configure RHEL 9 to implement certificate status checking for multifactor authentication.

Review the "/etc/sssd/conf.d/certificate_verification.conf" file to determine if the system is configured to prevent OCSP or certificate verification.

Add the following line to the "/etc/sssd/conf.d/certificate_verification.conf" file:

certificate_verification = ocsp_dgst=sha512

Set the correct ownership and permissions on the "/etc/sssd/conf.d/certificate_verification.conf" file by running these commands:

$ sudo chown root:root "/etc/sssd/conf.d/certificate_verification.conf"
$ sudo chmod 600 "/etc/sssd/conf.d/certificate_verification.conf"

The "sssd" service must be restarted for the changes to take effect. To restart the "sssd" service, run the following command:

$ sudo systemctl restart sssd.service

The pcsc-lite package must be installed if it is to be available for multifactor authentication using smart cards.

Verify that RHEL 9 has the pcsc-lite package installed with the following command:

$ sudo dnf list --installed pcsc-lite

Example output:

pcsc-lite.x86_64          1.9.4-1.el9

If the "pcsc-lite" package is not installed, this is a finding.

Note: If the system administrator (SA) demonstrates the use of an approved alternate multifactor authentication method, this requirement is not applicable.

The  pcsc-lite  package can be installed with the following command:
 
$ sudo dnf install pcsc-lite

The information system ensures that even if the information system is compromised, that compromise will not affect credentials stored on the authentication device.

The daemon program for pcsc-lite and the MuscleCard framework is pcscd. It is a resource manager that coordinates communications with smart card readers and smart cards and cryptographic tokens that are connected to the system.

Verify that the "pcscd" service is active with the following command:

$ systemctl is-active pcscd

active

If the pcscdservice is not active, this is a finding.

To enable the pcscd service run the following command:

$ sudo systemctl enable --now pcscd

The use of PIV credentials facilitates standardization and reduces the risk of unauthorized access.

The DOD has mandated the use of the common access card (CAC) to support identity management and personal authentication for systems covered under Homeland Security Presidential Directive (HSPD) 12, as well as making the CAC a primary component of layered protection for national security systems.

Satisfies: SRG-OS-000375-GPOS-00160, SRG-OS-000376-GPOS-00161

Verify that RHEL 9 has the opensc package installed with the following command:

$ sudo dnf list --installed opensc

Example output:

opensc.x86_64          0.22.0-2.el9

If the "opensc" package is not installed, this is a finding.

The opensc package can be installed with the following command:
 
$ sudo dnf install opensc

If the private key is discovered, an attacker can use the key to authenticate as an authorized user and gain access to the network infrastructure.

The cornerstone of the PKI is the private key used to encrypt or digitally sign information.

If the private key is stolen, this will lead to the compromise of the authentication and nonrepudiation gained through PKI because the attacker can use the private key to digitally sign documents and pretend to be the authorized user.

Both the holders of a digital certificate and the issuing authority must protect the computers, storage devices, or whatever they use to keep the private keys.

Verify the SSH private key files have a passcode.

For each private key stored on the system, use the following command:

$ sudo ssh-keygen -y -f /path/to/file

If the contents of the key are displayed, this is a finding.

Create a new private and public key pair that utilizes a passcode with the following command:

$ sudo ssh-keygen -n [passphrase]

To mitigate the risk of unauthorized access to sensitive information by entities that have been issued certificates by DOD-approved PKIs, all DOD systems (e.g., web servers and web portals) must be properly configured to incorporate access control methods that do not rely solely on the possession of a certificate for access. Successful authentication must not automatically give an entity access to an asset or security boundary. Authorization procedures and controls must be implemented to ensure each authenticated entity also has a validated and current authorization. Authorization is the process of determining whether an entity, once authenticated, is permitted to access a specific asset. Information systems use access control policies and enforcement mechanisms to implement this requirement.

This requirement prevents attackers with physical access from trivially bypassing security on the machine and gaining root access. Such accesses are further prevented by configuring the bootloader password.

Verify that RHEL 9 requires authentication for emergency mode with the following command:

$ grep sulogin /usr/lib/systemd/system/emergency.service 

ExecStart=-/usr/lib/systemd/systemd-sulogin-shell emergency

If this line is not returned, or is commented out, this is a finding. If the output is different, this is a finding.

Configure RHEL 9 to require authentication for emergency mode.

Add or modify the following line in the "/usr/lib/systemd/system/emergency.service" file:

ExecStart=-/usr/lib/systemd/systemd-sulogin-shell emergency

To mitigate the risk of unauthorized access to sensitive information by entities that have been issued certificates by DOD-approved PKIs, all DOD systems (e.g., web servers and web portals) must be properly configured to incorporate access control methods that do not rely solely on the possession of a certificate for access. Successful authentication must not automatically give an entity access to an asset or security boundary. Authorization procedures and controls must be implemented to ensure each authenticated entity also has a validated and current authorization. Authorization is the process of determining whether an entity, once authenticated, is permitted to access a specific asset. Information systems use access control policies and enforcement mechanisms to implement this requirement.

This requirement prevents attackers with physical access from trivially bypassing security on the machine and gaining root access. Such accesses are further prevented by configuring the bootloader password.

Verify that RHEL 9 requires authentication for single-user mode with the following command:

$ grep sulogin /usr/lib/systemd/system/rescue.service 

ExecStart=-/usr/lib/systemd/systemd-sulogin-shell rescue

If this line is not returned, or is commented out, this is a finding.

Configure RHEL 9 to require authentication for single-user mode.

Add or modify the following line in the "/usr/lib/systemd/system/rescue.service" file:

ExecStart=-/usr/lib/systemd/systemd-sulogin-shell rescue

Unapproved mechanisms used for authentication to the cryptographic module are not verified; therefore, cannot be relied upon to provide confidentiality or integrity and DOD data may be compromised.

RHEL 9 systems utilizing encryption are required to use FIPS-compliant mechanisms for authenticating to cryptographic modules.

The key derivation function (KDF) in Kerberos is not FIPS compatible. Ensuring the system does not have any keytab files present prevents system daemons from using Kerberos for authentication. A keytab is a file containing pairs of Kerberos principals and encrypted keys.

FIPS 140-3 is the current standard for validating that mechanisms used to access cryptographic modules utilize authentication that meets DOD requirements. This allows for Security Levels 1, 2, 3, or 4 for use on a general-purpose computing system.

Verify that RHEL 9 prevents system daemons from using Kerberos for authentication with the following command:

$ sudo ls -al /etc/*.keytab

ls: cannot access '/etc/*.keytab': No such file or directory 

If this command produces any "keytab" file(s), this is a finding.

Configure RHEL 9 to prevent system daemons from using Kerberos for authentication.

Remove any files with the .keytab extension from the operating system.

rm -f /etc/*.keytab

Without path validation, an informed trust decision by the relying party cannot be made when presented with any certificate not already explicitly trusted.

A trust anchor is an authoritative entity represented via a public key and associated data. It is used in the context of public key infrastructures, X.509 digital certificates, and DNSSEC.

When there is a chain of trust, usually the top entity to be trusted becomes the trust anchor; it can be, for example, a certification authority (CA). A certification path starts with the subject certificate and proceeds through a number of intermediate certificates up to a trusted root certificate, typically issued by a trusted CA.

This requirement verifies that a certification path to an accepted trust anchor is used for certificate validation and that the path includes status information. Path validation is necessary for a relying party to make an informed trust decision when presented with any certificate not already explicitly trusted. Status information for certification paths includes certificate revocation lists or online certificate status protocol responses. Validation of the certificate status information is out of scope for this requirement.

Satisfies: SRG-OS-000066-GPOS-00034, SRG-OS-000384-GPOS-00167

Verify RHEL 9 for PKI-based authentication has valid certificates by constructing a certification path (which includes status information) to an accepted trust anchor.

Check that the system has a valid DOD root CA installed with the following command:

$ sudo openssl x509 -text -in /etc/sssd/pki/sssd_auth_ca_db.pem

Example output:

Certificate:
    Data:
        Version: 3 (0x2)
        Serial Number: 1 (0x1)
        Signature Algorithm: sha256WithRSAEncryption
        Issuer: C = US, O = U.S. Government, OU = DoD, OU = PKI, CN = DoD Root CA 3
        Validity
        Not Before: Mar 20 18:46:41 2012 GMT
        Not After: Dec 30 18:46:41 2029 GMT
        Subject: C = US, O = U.S. Government, OU = DoD, OU = PKI, CN = DoD Root CA 3
        Subject Public Key Info:
            Public Key Algorithm: rsaEncryption

If the root CA file is not a DOD-issued certificate with a valid date and installed in the "/etc/sssd/pki/sssd_auth_ca_db.pem" location, this is a finding.

Configure RHEL 9, for PKI-based authentication, to validate certificates by constructing a certification path (which includes status information) to an accepted trust anchor.

Obtain a valid copy of the DOD root CA file from the PKI CA certificate bundle from cyber.mil and copy the DoD_PKE_CA_chain.pem into the following file:
/etc/sssd/pki/sssd_auth_ca_db.pem

Without mapping the certificate used to authenticate to the user account, the ability to determine the identity of the individual user or group will not be available for forensic analysis.

Verify the certificate of the user or group is mapped to the corresponding user or group in the "sssd.conf" file with the following command:

$ sudo cat /etc/sssd/sssd.conf 
 
[certmap/testing.test/rule_name]
matchrule =.*EDIPI@mil
maprule = (userCertificate;binary={cert!bin})
domains = testing.test

If the certmap section does not exist, ask the system administrator (SA) to indicate how certificates are mapped to accounts. If there is no evidence of certificate mapping, this is a finding.

Configure RHEL 9 to map the authenticated identity to the user or group account by adding or modifying the certmap section of the "/etc/sssd/sssd.conf" file based on the following example:

[certmap/testing.test/rule_name]
matchrule = .*EDIPI@mil
maprule = (userCertificate;binary={cert!bin})
domains = testing.test

The "sssd" service must be restarted for the changes to take effect. To restart the "sssd" service, run the following command:

$ sudo systemctl restart sssd.service

If cached authentication information is out-of-date, the validity of the authentication information may be questionable.

Verify that the System Security Services Daemon (SSSD) prohibits the use of cached authentications after one day.

Note: If smart card authentication is not being used on the system, this requirement is Not Applicable.

Check that SSSD allows cached authentications with the following command:

$ sudo grep cache_credentials /etc/sssd/sssd.conf

cache_credentials = true

If "cache_credentials" is set to "false" or missing from the configuration file, this is not a finding and no further checks are required.

If "cache_credentials" is set to "true", check that SSSD prohibits the use of cached authentications after one day with the following command:

$ sudo grep offline_credentials_expiration  /etc/sssd/sssd.conf

offline_credentials_expiration = 1

If "offline_credentials_expiration" is not set to a value of "1", this is a finding.

Configure the SSSD to prohibit the use of cached authentications after one day.

Add or change the following line in "/etc/sssd/sssd.conf" just below the line [pam]:

offline_credentials_expiration = 1

Without verification of the security functions, security functions may not operate correctly, and the failure may go unnoticed. Security function is defined as the hardware, software, and/or firmware of the information system responsible for enforcing the system security policy and supporting the isolation of code and data on which the protection is based. Security functionality includes, but is not limited to, establishing system accounts, configuring access authorizations (i.e., permissions, privileges), setting events to be audited, and setting intrusion detection parameters.

Satisfies: SRG-OS-000363-GPOS-00150, SRG-OS-000445-GPOS-00199

Verify that RHEL 9 has the Advanced Intrusion Detection Environment (AIDE) package installed with the following command:

$ sudo dnf list --installed aide

Example output:

aide.x86_64          0.16.100.el9

If AIDE is not installed, ask the system administrator (SA) how file integrity checks are performed on the system. 

If there is no application installed to perform integrity checks, this is a finding.

If AIDE is installed, check if it has been initialized with the following command:

$ sudo /usr/sbin/aide --check

If the output is "Couldn't open file /var/lib/aide/aide.db.gz for reading", this is a finding.

Install AIDE, initialize it, and perform a manual check.

Install AIDE:

$ sudo dnf install aide

Initialize AIDE:
     
$ sudo /usr/sbin/aide --init

Example output:

Start timestamp: 2023-06-05 10:09:04 -0600 (AIDE 0.16)
AIDE initialized database at /var/lib/aide/aide.db.new.gz

Number of entries:      86833

---------------------------------------------------
The attributes of the (uncompressed) database(s):
---------------------------------------------------

/var/lib/aide/aide.db.new.gz
  MD5      : coZUtPHhoFoeD7+k54fUvQ==
  SHA1     : DVpOEMWJwo0uPgrKZAygIUgSxeM=
  SHA256   : EQiZH0XNEk001tcDmJa+5STFEjDb4MPE
             TGdBJ/uvZKc=
  SHA512   : 86KUqw++PZhoPK0SZvT3zuFq9yu9nnPP
             toei0nENVELJ1LPurjoMlRig6q69VR8l
             +44EwO9eYyy9nnbzQsfG1g==

End timestamp: 2023-06-05 10:09:57 -0600 (run time: 0m 53s)

The new database will need to be renamed to be read by AIDE:

$ sudo mv /var/lib/aide/aide.db.new.gz /var/lib/aide/aide.db.gz

Perform a manual check:

$ sudo /usr/sbin/aide --check

Example output:

2023-06-05 10:16:08 -0600 (AIDE 0.16)
AIDE found NO differences between database and filesystem. Looks okay!!

...

Unauthorized changes to the baseline configuration could make the system vulnerable to various attacks or allow unauthorized access to the operating system. Changes to operating system configurations can have unintended side effects, some of which may be relevant to security.

Detecting such changes and providing an automated response can help avoid unintended, negative consequences that could ultimately affect the security state of the operating system. The operating system's information management officer (IMO)/information system security officer (ISSO) and system administrators (SAs) must be notified via email and/or monitoring system trap when there is an unauthorized modification of a configuration item.

Notifications provided by information systems include messages to local computer consoles, and/or hardware indications, such as lights.

This capability must take into account operational requirements for availability for selecting an appropriate response. The organization may choose to shut down or restart the information system upon security function anomaly detection.

Satisfies: SRG-OS-000363-GPOS-00150, SRG-OS-000446-GPOS-00200, SRG-OS-000447-GPOS-00201

Verify that RHEL 9 routinely executes a file integrity scan for changes to the system baseline. The command used in the example will use a daily occurrence.

Check the cron directories for scripts controlling the execution and notification of results of the file integrity application. For example, if AIDE is installed on the system, use the following commands:

$ ls -al /etc/cron.* | grep aide

-rwxr-xr-x 1 root root 29 Nov 22 2015 aide

$ grep aide /etc/crontab /var/spool/cron/root

/etc/crontab: 30 04 * * * root usr/sbin/aide
/var/spool/cron/root: 30 04 * * * root usr/sbin/aide

$ sudo more /etc/cron.daily/aide

#!/bin/bash
/usr/sbin/aide --check | /bin/mail -s "$HOSTNAME - Daily aide integrity check run" root@sysname.mil

If the file integrity application does not exist, or a script file controlling the execution of the file integrity application does not exist, or the file integrity application does not notify designated personnel of changes, this is a finding.

Configure the file integrity tool to run automatically on the system at least weekly and to notify designated personnel if baseline configurations are changed in an unauthorized manner. The AIDE tool can be configured to email designated personnel with the use of the cron system.
 
The following example output is generic. It will set cron to run AIDE daily and to send email at the completion of the analysis

$ sudo more /etc/cron.daily/aide
 
#!/bin/bash
/usr/sbin/aide --check | /bin/mail -s "$HOSTNAME - Daily aide integrity check run" root@sysname.mil

RHEL 9 installation media ships with an optional file integrity tool called Advanced Intrusion Detection Environment (AIDE). AIDE is highly configurable at install time. This requirement assumes the "aide.conf" file is under the "/etc" directory.

File integrity tools use cryptographic hashes for verifying file contents and directories have not been altered. These hashes must be FIPS 140-2/140-3-approved cryptographic hashes.

Verify that AIDE is configured to use FIPS 140-2/140-3 file hashing with the following command:

$ grep sha512 /etc/aide.conf 

All=p+i+n+u+g+s+m+S+sha512+acl+xattrs+selinux

If the "sha512" rule is not being used on all uncommented selection lines in the "/etc/aide.conf" file, or another file integrity tool is not using FIPS 140-2/140-3-approved cryptographic hashes for validating file contents and directories, this is a finding.

Configure the file integrity tool to use FIPS 140-2/140-3 cryptographic hashes for validating file and directory contents. 

If AIDE is installed, ensure the "sha512" rule is present on all uncommented file and directory selection lists. Exclude any log files, or files expected to change frequently, to reduce unnecessary notifications.

Protecting the integrity of the tools used for auditing purposes is a critical step toward ensuring the integrity of audit information. Audit information includes all information (e.g., audit records, audit settings, and audit reports) needed to successfully audit information system activity.

Audit tools include, but are not limited to, vendor-provided and open-source audit tools needed to successfully view and manipulate audit information system activity and records. Audit tools include custom queries and report generators.

It is not uncommon for attackers to replace the audit tools or inject code into the existing tools to provide the capability to hide or erase system activity from the audit logs.

To address this risk, audit tools must be cryptographically signed to provide the capability to identify when the audit tools have been modified, manipulated, or replaced. An example is a checksum hash of the file or files.

Satisfies: SRG-OS-000256-GPOS-00097, SRG-OS-000257-GPOS-00098, SRG-OS-000258-GPOS-00099, SRG-OS-000278-GPOS-00108

Check that AIDE is properly configured to protect the integrity of the audit tools with the following command:

$ sudo cat /etc/aide.conf | grep /usr/sbin/au

/usr/sbin/auditctl p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/auditd p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/ausearch p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/aureport p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/autrace p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/augenrules p+i+n+u+g+s+b+acl+xattrs+sha512
 
If AIDE is not installed, ask the system administrator (SA) how file integrity checks are performed on the system.

If any of the audit tools listed above do not have a corresponding line, ask the SA to indicate what cryptographic mechanisms are being used to protect the integrity of the audit tools.  If there is no evidence of integrity protection, this is a finding.

Add or update the following lines to "/etc/aide.conf", to protect the integrity of the audit tools.
 
/usr/sbin/auditctl p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/auditd p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/ausearch p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/aureport p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/autrace p+i+n+u+g+s+b+acl+xattrs+sha512
/usr/sbin/augenrules p+i+n+u+g+s+b+acl+xattrs+sha512

RHEL 9 installation media ships with an optional file integrity tool called Advanced Intrusion Detection Environment (AIDE). AIDE is highly configurable at install time. This requirement assumes the "aide.conf" file is under the "/etc" directory.

ACLs can provide permissions beyond those permitted through the file mode and must be verified by the file integrity tools.

Verify that that AIDE is verifying ACLs with the following command:

$ grep acl /etc/aide.conf 

All= p+i+n+u+g+s+m+S+sha512+acl+xattrs+selinux

If the "acl" rule is not being used on all uncommented selection lines in the "/etc/aide.conf" file, or ACLs are not being checked by another file integrity tool, this is a finding.

Configure the file integrity tool to check file and directory ACLs. 

If AIDE is installed, ensure the "acl" rule is present on all uncommented file and directory selection lists.

RHEL 9 installation media ships with an optional file integrity tool called Advanced Intrusion Detection Environment (AIDE). AIDE is highly configurable at install time. This requirement assumes the "aide.conf" file is under the "/etc" directory.

Extended attributes in file systems are used to contain arbitrary data and file metadata with security implications.

Verify that AIDE is configured to verify extended attributes with the following command:

$ grep xattrs /etc/aide.conf 

All= p+i+n+u+g+s+m+S+sha512+acl+xattrs+selinux

If the "xattrs" rule is not being used on all uncommented selection lines in the "/etc/aide.conf" file, or extended attributes are not being checked by another file integrity tool, this is a finding.

Configure the file integrity tool to check file and directory extended attributes. 

If AIDE is installed, ensure the "xattrs" rule is present on all uncommented file and directory selection lists.

rsyslogd is a system utility providing support for message logging. Support for both internet and Unix domain sockets enables this utility to support both local and remote logging. Couple this utility with "gnutls" (which is a secure communications library implementing the SSL, TLS, and DTLS protocols), to create a method to securely encrypt and offload auditing.

Satisfies: SRG-OS-000479-GPOS-00224, SRG-OS-000051-GPOS-00024, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 has the rsyslogd package installed with the following command:

$ sudo dnf list --installed rsyslog

Example output:

rsyslog.x86_64          8.2102.0-101.el9_0.1

If the "rsyslogd" package is not installed, this is a finding.

The rsyslogd package can be installed with the following command:
 
$ sudo dnf install rsyslogd

The rsyslog-gnutls package provides Transport Layer Security (TLS) support for the rsyslog daemon, which enables secure remote logging.

Satisfies: SRG-OS-000480-GPOS-00227, SRG-OS-000120-GPOS-00061

Verify that RHEL 9 has the rsyslog-gnutls package installed with the following command:

$ sudo dnf list --installed rsyslog-gnutls

Example output:

rsyslog-gnutls.x86_64          8.2102.0-101.el9_0.1

If the "rsyslog-gnutls" package is not installed, this is a finding.

The  rsyslog-gnutls package can be installed with the following command:
 
$ sudo dnf install rsyslog-gnutls

The "rsyslog" service must be running to provide logging services, which are essential to system administration.

Verify that "rsyslog" is active with the following command:

$ systemctl is-active rsyslog 

active

If the rsyslog service is not active, this is a finding.

To enable the rsyslog service, run the following command:

$ sudo systemctl enable --now rsyslog

Unintentionally running a rsyslog server accepting remote messages puts the system at increased risk. Malicious rsyslog messages sent to the server could exploit vulnerabilities in the server software itself, could introduce misleading information into the system's logs, or could fill the system's storage leading to a denial of service.

If the system is intended to be a log aggregation server, its use must be documented with the information system security officer (ISSO).

Verify that RHEL 9 is not configured to receive remote logs using rsyslog with the following commands:

$ grep -i modload /etc/rsyslog.conf /etc/rsyslog.d/*
$ModLoad imtcp
$ModLoad imrelp

$ grep -i serverrun /etc/rsyslog.conf /etc/rsyslog.d/*
$InputTCPServerRun 514
$InputRELPServerRun 514

Note: An error stating no files or directories may be returned. This is not a finding.

If any modules are being loaded in the "/etc/rsyslog.conf" file or in the "/etc/rsyslog.d" subdirectories, ask to see the documentation for the system being used for log aggregation. If the documentation does not exist or does not specify the server as a log aggregation system, this is a finding.

Configure RHEL 9 to not receive remote logs using rsyslog.

Remove the lines in /etc/rsyslog.conf and any files in the /etc/rsyslog.d directory that match any of the following:

$ModLoad imtcp
$ModLoad imudp
$ModLoad imrelp
$InputTCPServerRun [0-9]*
$UDPServerRun [0-9]*
$InputRELPServerRun [0-9]*

The rsyslog daemon must be restarted for the changes to take effect:

$ sudo systemctl restart rsyslog.service

Logging remote access methods can be used to trace the decrease in the risks associated with remote user access management. It can also be used to spot cyberattacks and ensure ongoing compliance with organizational policies surrounding the use of remote access methods.

Verify that RHEL 9 monitors all remote access methods.

Check that remote access methods are being logged by running the following command:

$ grep -rE '(auth.\*|authpriv.\*|daemon.\*)' /etc/rsyslog.conf

/etc/rsyslog.conf:authpriv.*
 
If "auth.*", "authpriv.*" or "daemon.*" are not configured to be logged, this is a finding.

Add or update the following lines to the "/etc/rsyslog.conf" file:

auth.*;authpriv.*;daemon.* /var/log/secure

The "rsyslog" service must be restarted for the changes to take effect with the following command:

$ sudo systemctl restart rsyslog.service

The auditd service does not include the ability to send audit records to a centralized server for management directly. However, it can use a plug-in for audit event multiplexor (audispd) to pass audit records to the local syslog server.

Satisfies: SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify RHEL 9 is configured use the audisp-remote syslog service with the following command:

$ sudo grep active /etc/audit/plugins.d/syslog.conf 

active = yes

If the "active" keyword does not have a value of "yes", the line is commented out, or the line is missing, this is a finding.

Edit the /etc/audit/plugins.d/syslog.conf file and add or update the "active" option:

active = yes

The audit daemon must be restarted for changes to take effect.

Information stored in one location is vulnerable to accidental or incidental deletion or alteration.

Offloading is a common process in information systems with limited audit storage capacity.

RHEL 9 installation media provides "rsyslogd", a system utility providing support for message logging. Support for both internet and Unix domain sockets enables this utility to support both local and remote logging. Coupling this utility with "gnutls" (a secure communications library implementing the SSL, TLS and DTLS protocols) creates a method to securely encrypt and offload auditing.

"Rsyslog" supported authentication modes include:
anon - anonymous authentication
x509/fingerprint - certificate fingerprint authentication
x509/certvalid - certificate validation only
x509/name - certificate validation and subject name authentication

Satisfies: SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify RHEL 9 authenticates the remote logging server for offloading audit logs with the following command:

$ sudo grep -i '$ActionSendStreamDriverAuthMode' /etc/rsyslog.conf /etc/rsyslog.d/*.conf 

 /etc/rsyslog.conf:$ActionSendStreamDriverAuthMode x509/name 

If the value of the "$ActionSendStreamDriverAuthMode" option is not set to "x509/name" or the line is commented out, ask the system administrator (SA) to indicate how the audit logs are offloaded to a different system or media. 

If there is no evidence that the transfer of the audit logs being offloaded to another system or media is encrypted, this is a finding.

Configure RHEL 9 to authenticate the remote logging server for offloading audit logs by setting the following option in "/etc/rsyslog.conf" or "/etc/rsyslog.d/[customfile].conf":

$ActionSendStreamDriverAuthMode x509/name

Information stored in one location is vulnerable to accidental or incidental deletion or alteration.

Offloading is a common process in information systems with limited audit storage capacity.

RHEL 9 installation media provides "rsyslogd", a system utility providing support for message logging. Support for both internet and Unix domain sockets enables this utility to support both local and remote logging. Coupling this utility with "gnutls" (a secure communications library implementing the SSL, TLS and DTLS protocols) creates a method to securely encrypt and offload auditing.

"Rsyslog" supported authentication modes include:
anon - anonymous authentication
x509/fingerprint - certificate fingerprint authentication
x509/certvalid - certificate validation only
x509/name - certificate validation and subject name authentication

Satisfies: SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify RHEL 9 encrypts audit records offloaded onto a different system or media from the system being audited via rsyslog with the following command:

$ sudo grep -i '$ActionSendStreamDriverMode' /etc/rsyslog.conf /etc/rsyslog.d/*.conf 

/etc/rsyslog.conf:$ActionSendStreamDriverMode 1 

If the value of the "$ActionSendStreamDriverMode" option is not set to "1" or the line is commented out, this is a finding.

Configure RHEL 9 to encrypt offloaded audit records via rsyslog by setting the following options in "/etc/rsyslog.conf" or "/etc/rsyslog.d/[customfile].conf":

$ActionSendStreamDriverMode 1

Information stored in one location is vulnerable to accidental or incidental deletion or alteration.

Offloading is a common process in information systems with limited audit storage capacity.

RHEL 9 installation media provides "rsyslogd", a system utility providing support for message logging. Support for both internet and Unix domain sockets enables this utility to support both local and remote logging. Coupling this utility with "gnutls" (a secure communications library implementing the SSL, TLS and DTLS protocols) creates a method to securely encrypt and offload auditing.

Satisfies: SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify RHEL 9 uses the gtls driver to encrypt audit records offloaded onto a different system or media from the system being audited with the following command:

$ sudo grep -i '$DefaultNetstreamDriver' /etc/rsyslog.conf /etc/rsyslog.d/*.conf 

/etc/rsyslog.conf:$DefaultNetstreamDriver gtls 

If the value of the "$DefaultNetstreamDriver" option is not set to "gtls" or the line is commented out, this is a finding.

Configure RHEL 9 to use the gtls driver to encrypt offloaded audit records by setting the following options in "/etc/rsyslog.conf" or "/etc/rsyslog.d/[customfile].conf":

$DefaultNetstreamDriver gtls

Information stored in one location is vulnerable to accidental or incidental deletion or alteration.

Offloading is a common process in information systems with limited audit storage capacity.

RHEL 9 installation media provides "rsyslogd", a system utility providing support for message logging. Support for both internet and Unix domain sockets enables this utility to support both local and remote logging. Coupling this utility with "gnutls" (a secure communications library implementing the SSL, TLS and DTLS protocols) creates a method to securely encrypt and offload auditing.

Rsyslog provides three ways to forward message: the traditional UDP transport, which is extremely lossy but standard; the plain TCP based transport, which loses messages only during certain situations but is widely available; and the RELP transport, which does not lose messages but is currently available only as part of the rsyslogd 3.15.0 and above.

Examples of each configuration:
UDP *.* @remotesystemname
TCP *.* @@remotesystemname
RELP *.* :omrelp:remotesystemname:2514
Note that a port number was given as there is no standard port for RELP.

Satisfies: SRG-OS-000479-GPOS-00224, SRG-OS-000480-GPOS-00227, SRG-OS-000342-GPOS-00133

Verify that RHEL 9 audit system offloads audit records onto a different system or media from the system being audited via rsyslog using TCP with the following command:

$ sudo grep @@ /etc/rsyslog.conf /etc/rsyslog.d/*.conf

/etc/rsyslog.conf:*.* @@[remoteloggingserver]:[port]

If a remote server is not configured, or the line is commented out, ask the system administrator (SA) to indicate how the audit logs are offloaded to a different system or media. 

If there is no evidence that the audit logs are being offloaded to another system or media, this is a finding.

Configure RHEL 9 to offload audit records onto a different system or media from the system being audited via TCP using rsyslog by specifying the remote logging server in "/etc/rsyslog.conf"" or "/etc/rsyslog.d/[customfile].conf" with the name or IP address of the log aggregation server.

*.* @@[remoteloggingserver]:[port]"

Cron logging can be used to trace the successful or unsuccessful execution of cron jobs. It can also be used to spot intrusions into the use of the cron facility by unauthorized and malicious users.

Verify that "rsyslog" is configured to log cron events with the following command:

Note: If another logging package is used, substitute the utility configuration file for "/etc/rsyslog.conf" or "/etc/rsyslog.d/*.conf" files.

$ sudo grep -s cron /etc/rsyslog.conf /etc/rsyslog.d/*.conf

/etc/rsyslog.conf:*.info;mail.none;authpriv.none;cron.none                          /var/log/messages
/etc/rsyslog.conf:cron.*                                                           /var/log/cron             

If the command does not return a response, check for cron logging all facilities with the following command:

$ sudo grep -s /var/log/messages /etc/rsyslog.conf /etc/rsyslog.d/*.conf

/etc/rsyslog.conf:*.info;mail.none;authpriv.none;cron.none                          /var/log/messages

If "rsyslog" is not logging messages for the cron facility or all facilities, this is a finding.

Configure "rsyslog" to log all cron messages by adding or updating the following line to "/etc/rsyslog.conf" or a configuration file in the /etc/rsyslog.d/ directory:

cron.* /var/log/cron

The rsyslog daemon must be restarted for the changes to take effect:

$ sudo systemctl restart rsyslog.service

Without establishing what type of events occurred, the source of events, where events occurred, and the outcome of events, it would be difficult to establish, correlate, and investigate the events leading up to an outage or attack.

Audit record content that may be necessary to satisfy this requirement includes, for example, time stamps, source and destination addresses, user/process identifiers, event descriptions, success/fail indications, filenames involved, and access control or flow control rules invoked.

Associating event types with detected events in audit logs provides a means of investigating an attack, recognizing resource utilization or capacity thresholds, or identifying an improperly configured RHEL 9 system.

Satisfies: SRG-OS-000062-GPOS-00031, SRG-OS-000037-GPOS-00015, SRG-OS-000038-GPOS-00016, SRG-OS-000039-GPOS-00017, SRG-OS-000040-GPOS-00018, SRG-OS-000041-GPOS-00019, SRG-OS-000042-GPOS-00021, SRG-OS-000051-GPOS-00024, SRG-OS-000054-GPOS-00025, SRG-OS-000122-GPOS-00063, SRG-OS-000254-GPOS-00095, SRG-OS-000255-GPOS-00096, SRG-OS-000337-GPOS-00129, SRG-OS-000348-GPOS-00136, SRG-OS-000349-GPOS-00137, SRG-OS-000350-GPOS-00138, SRG-OS-000351-GPOS-00139, SRG-OS-000352-GPOS-00140, SRG-OS-000353-GPOS-00141, SRG-OS-000354-GPOS-00142, SRG-OS-000358-GPOS-00145, SRG-OS-000365-GPOS-00152, SRG-OS-000392-GPOS-00172, SRG-OS-000475-GPOS-00220, SRG-OS-000055-GPOS-00026

Verify that the RHEL 9 audit service package is installed.

Check that the audit service package is installed with the following command:

$ sudo dnf list --installed audit

Example output:

audit-3.0.7-101.el9_0.2.x86_64

If the "audit" package is not installed, this is a finding.

Install the audit service package (if the audit service is not already installed) with the following command:

$ sudo dnf install audit

Without establishing what type of events occurred, it would be difficult to establish, correlate, and investigate the events leading up to an outage or attack. Ensuring the "auditd" service is active ensures audit records generated by the kernel are appropriately recorded.

Additionally, a properly configured audit subsystem ensures that actions of individual system users can be uniquely traced to those users so they can be held accountable for their actions.

Satisfies: SRG-OS-000062-GPOS-00031, SRG-OS-000037-GPOS-00015, SRG-OS-000038-GPOS-00016, SRG-OS-000039-GPOS-00017, SRG-OS-000040-GPOS-00018, SRG-OS-000041-GPOS-00019, SRG-OS-000042-GPOS-00021, SRG-OS-000051-GPOS-00024, SRG-OS-000054-GPOS-00025, SRG-OS-000122-GPOS-00063, SRG-OS-000254-GPOS-00095, SRG-OS-000255-GPOS-00096, SRG-OS-000337-GPOS-00129, SRG-OS-000348-GPOS-00136, SRG-OS-000349-GPOS-00137, SRG-OS-000350-GPOS-00138, SRG-OS-000351-GPOS-00139, SRG-OS-000352-GPOS-00140, SRG-OS-000353-GPOS-00141, SRG-OS-000354-GPOS-00142, SRG-OS-000358-GPOS-00145, SRG-OS-000365-GPOS-00152, SRG-OS-000392-GPOS-00172, SRG-OS-000475-GPOS-00220

Verify the audit service is configured to produce audit records with the following command:

$ systemctl status auditd.service

auditd.service - Security Auditing Service
Loaded:loaded (/usr/lib/systemd/system/auditd.service; enabled; vendor preset: enabled)
Active: active (running) since Tues 2022-05-24 12:56:56 EST; 4 weeks 0 days ago

If the audit service is not "active" and "running", this is a finding.

To enable the auditd service run the following command:

$ sudo systemctl enable --now auditd

It is critical that when the operating system is at risk of failing to process audit logs as required, it takes action to mitigate the failure. Audit processing failures include software/hardware errors; failures in the audit capturing mechanisms; and audit storage capacity being reached or exceeded. Responses to audit failure depend upon the nature of the failure mode.

Verify RHEL 9 takes the appropriate action when an audit processing failure occurs.

Check that RHEL 9 takes the appropriate action when an audit processing failure occurs with the following command:

$ sudo grep disk_error_action /etc/audit/auditd.conf

disk_error_action = HALT

If the value of the "disk_error_action" option is not "SYSLOG", "SINGLE", or "HALT", or the line is commented out, ask the system administrator (SA) to indicate how the system takes appropriate action when an audit process failure occurs. If there is no evidence of appropriate action, this is a finding.

Configure RHEL 9 to shut down by default upon audit failure (unless availability is an overriding concern).

Add or update the following line (depending on configuration "disk_error_action" can be set to "SYSLOG" or "SINGLE" depending on configuration) in "/etc/audit/auditd.conf" file:

disk_error_action = HALT

If availability has been determined to be more important, and this decision is documented with the information system security officer (ISSO), configure the operating system to notify SA staff and ISSO staff in the event of an audit processing failure by setting the "disk_error_action" to "SYSLOG".

It is critical that when the operating system is at risk of failing to process audit logs as required, it takes action to mitigate the failure. Audit processing failures include software/hardware errors; failures in the audit capturing mechanisms; and audit storage capacity being reached or exceeded. Responses to audit failure depend upon the nature of the failure mode.

Verify RHEL 9 takes the appropriate action when the audit storage volume is full. 

Check that RHEL 9 takes the appropriate action when the audit storage volume is full with the following command:

$ sudo grep disk_full_action /etc/audit/auditd.conf

disk_full_action = HALT

If the value of the "disk_full_action" option is not "SYSLOG", "SINGLE", or "HALT", or the line is commented out, ask the system administrator (SA) to indicate how the system takes appropriate action when an audit storage volume is full. If there is no evidence of appropriate action, this is a finding.

Configure RHEL 9 to shut down by default upon audit failure (unless availability is an overriding concern).

Add or update the following line (depending on configuration "disk_full_action" can be set to "SYSLOG" or "SINGLE" depending on configuration) in "/etc/audit/auditd.conf" file:

disk_full_action = HALT

If availability has been determined to be more important, and this decision is documented with the information system security officer (ISSO), configure the operating system to notify SA staff and ISSO staff in the event of an audit processing failure by setting the "disk_full_action" to "SYSLOG".

To ensure RHEL 9 systems have a sufficient storage capacity in which to write the audit logs, RHEL 9 needs to be able to allocate audit record storage capacity.

The task of allocating audit record storage capacity is usually performed during initial installation of RHEL 9.

Satisfies: SRG-OS-000341-GPOS-00132, SRG-OS-000342-GPOS-00133

Verify RHEL 9 allocates audit record storage capacity to store at least one week of audit records when audit records are not immediately sent to a central audit record storage facility.

Note: The partition size needed to capture a week of audit records is based on the activity level of the system and the total storage capacity available. Typically 10.0 GB of storage space for audit records should be sufficient.

Determine which partition the audit records are being written to with the following command:

$ sudo grep log_file /etc/audit/auditd.conf
log_file = /var/log/audit/audit.log 

Check the size of the partition that audit records are written to with the following command and verify whether it is sufficiently large:

 # df -h /var/log/audit/
/dev/sda2 24G 10.4G 13.6G 43% /var/log/audit 

If the audit record partition is not allocated for sufficient storage capacity, this is a finding.

Allocate enough storage capacity for at least one week of audit records when audit records are not immediately sent to a central audit record storage facility.

If audit records are stored on a partition made specifically for audit records, resize the partition with sufficient space to contain one week of audit records.

If audit records are not stored on a partition made specifically for audit records, a new partition with sufficient space will need be to be created.

If security personnel are not notified immediately when storage volume reaches 75 percent utilization, they are unable to plan for audit record storage capacity expansion.

Verify RHEL 9 takes action when allocated audit record storage volume reaches 75 percent of the repository maximum audit record storage capacity with the following command:

$ sudo grep -w space_left /etc/audit/auditd.conf

space_left = 25%

If the value of the "space_left" keyword is not set to 25 percent of the storage volume allocated to audit logs, or if the line is commented out, ask the system administrator (SA) to indicate how the system is providing real-time alerts to the SA and information system security officer (ISSO). If the "space_left" value is not configured to the correct value, this is a finding.

Configure RHEL 9 to initiate an action to notify the SA and ISSO (at a minimum) when allocated audit record storage volume reaches 75 percent of the repository maximum audit record storage capacity by adding/modifying the following line in the /etc/audit/auditd.conf file.

space_left  = 25%

If security personnel are not notified immediately when storage volume reaches 75 percent utilization, they are unable to plan for audit record storage capacity expansion.

Verify RHEL 9 notifies the SA and ISSO (at a minimum) when allocated audit record storage volume reaches 75 percent of the repository maximum audit record storage capacity with the following command:

$ sudo grep -w space_left_action /etc/audit/auditd.conf

space_left_action = email

If the value of the "space_left_action" is not set to "email", or if the line is commented out, ask the SA to indicate how the system is providing real-time alerts to the SA and ISSO.

If there is no evidence that real-time alerts are configured on the system, this is a finding.

Configure RHEL 9 to initiate an action to notify the SA and ISSO (at a minimum) when allocated audit record storage volume reaches 75 percent of the repository maximum audit record storage capacity by adding/modifying the following line in the /etc/audit/auditd.conf file.

space_left_action = email

If action is not taken when storage volume reaches 95 percent utilization, the auditing system may fail when the storage volume reaches capacity.

Verify RHEL 9 takes action when allocated audit record storage volume reaches 95 percent of the repository maximum audit record storage capacity with the following command:

$ sudo grep -w admin_space_left /etc/audit/auditd.conf

admin_space_left = 5%

If the value of the "admin_space_left" keyword is not set to 5 percent of the storage volume allocated to audit logs, or if the line is commented out, ask the system administrator (SA) to indicate how the system is taking action if the allocated storage is about to reach capacity. If the "space_left" value is not configured to the correct value, this is a finding.

Configure RHEL 9 to initiate an action when allocated audit record storage volume reaches 95 percent of the repository maximum audit record storage capacity by adding/modifying the following line in the /etc/audit/auditd.conf file.

admin_space_left  = 5%

If action is not taken when storage volume reaches 95 percent utilization, the auditing system may fail when the storage volume reaches capacity.

Verify that RHEL 9 is configured to take action in the event of allocated audit record storage volume reaches 95 percent of the repository maximum audit record storage capacity with the following command:

$ sudo grep admin_space_left_action /etc/audit/auditd.conf

admin_space_left_action = single

If the value of the "admin_space_left_action" is not set to "single", or if the line is commented out, ask the system administrator (SA) to indicate how the system is providing real-time alerts to the SA and information system security officer (ISSO).

If there is no evidence that real-time alerts are configured on the system, this is a finding.

Configure "auditd" service  to take action in the event of allocated audit record storage volume reaches 95 percent of the repository maximum audit record storage capacity.

Edit the following line in "/etc/audit/auditd.conf" to ensure that the system is forced into single user mode in the event the audit record storage volume is about to reach maximum capacity:

admin_space_left_action = single 

The audit daemon must be restarted for changes to take effect.

It is critical that when the operating system is at risk of failing to process audit logs as required, it takes action to mitigate the failure. Audit processing failures include software/hardware errors; failures in the audit capturing mechanisms; and audit storage capacity being reached or exceeded. Responses to audit failure depend upon the nature of the failure mode.

Verify that RHEL 9 takes the appropriate action when the audit files have reached maximum size with the following command:

$ sudo grep max_log_file_action /etc/audit/auditd.conf

max_log_file_action = ROTATE

If the value of the "max_log_file_action" option is not "ROTATE", "SINGLE", or the line is commented out, ask the system administrator (SA)to indicate how the system takes appropriate action when an audit storage volume is full. If there is no evidence of appropriate action, this is a finding.

Configure RHEL 9 to rotate the audit log when it reaches maximum size.

Add or update the following line in "/etc/audit/auditd.conf" file:

max_log_file_action = ROTATE

Enriched logging is needed to determine who, what, and when events occur on a system. Without this, determining root cause of an event will be much more difficult.

When audit logs are not labeled before they are sent to a central log server, the audit data will not be able to be analyzed and tied back to the correct system.

Satisfies: SRG-OS-000039-GPOS-00017, SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify that RHEL 9 Audit Daemon is configured to label all offloaded audit logs, with the following command:

$ sudo grep name_format /etc/audit/auditd.conf

name_format = hostname

If the "name_format" option is not "hostname", "fqd", or "numeric", or the line is commented out, this is a finding.

Edit the /etc/audit/auditd.conf file and add or update the "name_format" option:

name_format = hostname

The audit daemon must be restarted for changes to take effect.

The audit system should have an action setup in the event the internal event queue becomes full so that no data is lost.  Information stored in one location is vulnerable to accidental or incidental deletion or alteration.

Offloading is a common process in information systems with limited audit storage capacity.

Satisfies: SRG-OS-000342-GPOS-00133, SRG-OS-000479-GPOS-00224

Verify that RHEL 9 audit system is configured to take an appropriate action when the internal event queue is full:

$ sudo grep -i overflow_action /etc/audit/auditd.conf 

overflow_action = syslog

If the value of the "overflow_action" option is not set to "syslog", "single", "halt" or the line is commented out, ask the system administrator (SA) to indicate how the audit logs are offloaded to a different system or media.

If there is no evidence that the transfer of the audit logs being offloaded to another system or media takes appropriate action if the internal event queue becomes full, this is a finding.

Edit the /etc/audit/auditd.conf file and add or update the "overflow_action" option:

overflow_action = syslog

The audit daemon must be restarted for changes to take effect.

It is critical for the appropriate personnel to be aware if a system is at risk of failing to process audit logs as required. Without this notification, the security personnel may be unaware of an impending failure of the audit capability, and system operation may be adversely affected.

Audit processing failures include software/hardware errors, failures in the audit capturing mechanisms, and audit storage capacity being reached or exceeded.

This requirement applies to each audit data storage repository (i.e., distinct information system component where audit records are stored), the centralized audit storage capacity of organizations (i.e., all audit data storage repositories combined), or both.

Satisfies: SRG-OS-000046-GPOS-00022, SRG-OS-000343-GPOS-00134

Verify that RHEL 9 is configured to notify the SA and/or ISSO (at a minimum) in the event of an audit processing failure with the following command:

$ sudo grep action_mail_acct /etc/audit/auditd.conf

action_mail_acct = root

If the value of the "action_mail_acct" keyword is not set to "root" and/or other accounts for security personnel, the "action_mail_acct" keyword is missing, or the retuned line is commented out, ask the SA to indicate how they and the ISSO are notified of an audit process failure. If there is no evidence of the proper personnel being notified of an audit processing failure, this is a finding.

Configure "auditd" service to notify the SA and ISSO in the event of an audit processing failure.

Edit the following line in "/etc/audit/auditd.conf" to ensure that administrators are notified via email for those situations:

action_mail_acct = root

The audit daemon must be restarted for changes to take effect.

Without establishing what type of events occurred, the source of events, where events occurred, and the outcome of events, it would be difficult to establish, correlate, and investigate the events leading up to an outage or attack.

If option "local_events" isn't set to "yes" only events from network will be aggregated.

Satisfies: SRG-OS-000062-GPOS-00031, SRG-OS-000480-GPOS-00227

Verify that the RHEL 9 audit system is configured to audit local events with the following command:

$ sudo grep local_events /etc/audit/auditd.conf 

local_events = yes 

If "local_events" isn't set to "yes", if the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for local events by adding or updating the following line in "/etc/audit/auditd.conf":

local_events = yes 

The audit daemon must be restarted for the changes to take effect.

Unauthorized disclosure of audit records can reveal system and configuration data to attackers, thus compromising its confidentiality.

Satisfies: SRG-OS-000057-GPOS-00027, SRG-OS-000058-GPOS-00028, SRG-OS-000059-GPOS-00029, SRG-OS-000206-GPOS-00084

Verify the audit logs are group-owned by "root" or a restricted logging group. 

First determine if a group other than "root" has been assigned to the audit logs with the following command:

$ sudo grep log_group /etc/audit/auditd.conf

Then determine where the audit logs are stored with the following command:

$ sudo grep -iw log_file /etc/audit/auditd.conf

log_file = /var/log/audit/audit.log

Then using the location of the audit log file, determine if the audit log is group-owned by "root" using the following command:

$ sudo stat -c "%G %n" /var/log/audit/audit.log

root /var/log/audit/audit.log

If the audit log is not group-owned by "root" or the configured alternative logging group, this is a finding.

Change the group of the directory of "/var/log/audit" to be owned by a correct group.

Identify the group that is configured to own audit log:

$ sudo grep -P '^[ ]*log_group[ ]+=.*$' /etc/audit/auditd.conf

Change the ownership to that group:

$ sudo chgrp ${GROUP} /var/log/audit

Unauthorized disclosure of audit records can reveal system and configuration data to attackers, thus compromising its confidentiality.

Satisfies: SRG-OS-000057-GPOS-00027, SRG-OS-000058-GPOS-00028, SRG-OS-000059-GPOS-00029, SRG-OS-000206-GPOS-00084

Verify the audit logs directory is owned by "root". 

First determine where the audit logs are stored with the following command:

$ sudo grep -iw log_file /etc/audit/auditd.conf

log_file = /var/log/audit/audit.log

Then using the location of the audit log file, determine if the audit log directory is owned by "root" using the following command:

$ sudo ls -ld /var/log/audit

drwx------ 2 root root 23 Jun 11 11:56 /var/log/audit

If the audit log directory is not owned by "root", this is a finding.

Configure the audit log to be protected from unauthorized read access by setting the correct owner as "root" with the following command:

$ sudo chown root /var/log/audit

Only authorized personnel should be aware of errors and the details of the errors. Error messages are an indicator of an organization's operational state or can identify the RHEL 9 system or platform. Additionally, Personally Identifiable Information (PII) and operational information must not be revealed through error messages to unauthorized personnel or their designated representatives.

The structure and content of error messages must be carefully considered by the organization and development team. The extent to which the information system is able to identify and handle error conditions is guided by organizational policy and operational requirements.

Satisfies: SRG-OS-000057-GPOS-00027, SRG-OS-000058-GPOS-00028, SRG-OS-000059-GPOS-00029, SRG-OS-000206-GPOS-00084

Verify the audit logs have a mode of "0600". 

First determine where the audit logs are stored with the following command:

$ sudo grep -iw log_file /etc/audit/auditd.conf

log_file = /var/log/audit/audit.log

Then using the location of the audit log file, determine if the audit log files as a mode of "0640" with the following command:

$ sudo ls -la /var/log/audit/*.log

rw-------. 2 root root 237923 Jun 11 11:56 /var/log/audit/audit.log

If the audit logs have a mode more permissive than "0600", this is a finding.

Configure the audit logs to have a mode of "0600" with the following command:

Replace "[audit_log_file]" to the correct audit log path, by default this location is "/var/log/audit/audit.log".

$ sudo chmod 0600 /var/log/audit/[audit_log_file]

Check the group that owns the system audit logs:

$ sudo grep -m 1 -q ^log_group /etc/audit/auditd.conf

If the log_group is not defined or it is set to root, configure the permissions the following way:

$ sudo chmod 0640 $log_file
$ sudo chmod 0440 $log_file.*

Otherwise, configure the permissions the following way:

$ sudo chmod 0600 $log_file
$ sudo chmod 0400 $log_file.*

If option "freq" is not set to a value that requires audit records being written to disk after a threshold number is reached, then audit records may be lost.

Verify that audit system is configured to flush to disk after every 100 records with the following command:

$ sudo grep freq /etc/audit/auditd.conf 

freq = 100 

If "freq" isn't set to a value between "1" and "100", the value is missing, or the line is commented out, this is a finding.

Configure RHEL 9 to flush audit to disk by adding or updating the following rule in "/etc/audit/auditd.conf":

freq = 100

The audit daemon must be restarted for the changes to take effect.

Without establishing what type of events occurred, the source of events, where events occurred, and the outcome of events, it would be difficult to establish, correlate, and investigate the events leading up to an outage or attack.

Audit record content that may be necessary to satisfy this requirement includes, for example, time stamps, source and destination addresses, user/process identifiers, event descriptions, success/fail indications, filenames involved, and access control or flow control rules invoked.

Enriched logging aids in making sense of who, what, and when events occur on a system. Without this, determining root cause of an event will be much more difficult.

Satisfies: SRG-OS-000255-GPOS-00096, SRG-OS-000480-GPOS-00227

Verify that RHEL 9 audit system is configured to resolve audit information before writing to disk, with the following command:

$ sudo grep log_format /etc/audit/auditd.conf

log_format = ENRICHED

If the "log_format" option is not "ENRICHED", or the line is commented out, this is a finding.

Edit the /etc/audit/auditd.conf file and add or update the "log_format" option:

log_format = ENRICHED

The audit daemon must be restarted for changes to take effect.

Audit data should be synchronously written to disk to ensure log integrity. This setting assures that all audit event data is written disk.

Verify that the audit system is configured to write logs to the disk with the following command:

$ sudo grep write_logs /etc/audit/auditd.conf 

write_logs = yes 

If "write_logs" does not have a value of "yes", the line is commented out, or the line is missing, this is a finding.

Configure the audit system to write log files to the disk.

Edit the /etc/audit/auditd.conf file and add or update the "write_logs" option to "yes":

write_logs = yes 

The audit daemon must be restarted for changes to take effect.

Without the capability to restrict the roles and individuals that can select which events are audited, unauthorized personnel may be able to prevent the auditing of critical events. Misconfigured audits may degrade the system's performance by overwhelming the audit log. Misconfigured audits may also make it more difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Verify that the files in directory "/etc/audit/rules.d/" and "/etc/audit/auditd.conf" file have a mode of "0640" or less permissive with the following command:

# stat -c "%a %n"  /etc/audit/rules.d/*.rules

640 /etc/audit/rules.d/audit.rules 

If the files in the "/etc/audit/rules.d/" directory or the "/etc/audit/auditd.conf" file have a mode more permissive than "0640", this is a finding.

Configure the files in directory "/etc/audit/rules.d/" and the "/etc/audit/auditd.conf" file to have a mode of "0640" with the following commands:

$ sudo chmod 0640 /etc/audit/rules.d/audit.rules
$ sudo chmod 0640 /etc/audit/rules.d/[customrulesfile].rules
$ sudo chmod 0640 /etc/audit/auditd.conf

Without the capability to restrict the roles and individuals that can select which events are audited, unauthorized personnel may be able to prevent the auditing of critical events. Misconfigured audits may degrade the system's performance by overwhelming the audit log. Misconfigured audits may also make it more difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Verify the mode of /etc/audit/auditd.conf with the command:

$ sudo stat -c "%a %n" /etc/audit/auditd.conf

640 /etc/audit/auditd.conf

If "/etc/audit/auditd.conf" does not have a mode of "0640", this is a finding.

Set the mode of /etc/audit/auditd.conf file to 0640 with the command:

$ sudo chmod 0640 /etc/audit/auditd.conf

Without the capability to generate audit records, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

If auditing is enabled late in the startup process, the actions of some startup processes may not be audited. Some audit systems also maintain state information only available if auditing is enabled before a given process is created.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

Allocating an audit_backlog_limit of sufficient size is critical in maintaining a stable boot process. With an insufficient limit allocated, the system is susceptible to boot failures and crashes.

Satisfies: SRG-OS-000254-GPOS-00095, SRG-OS-000341-GPOS-00132

Verify RHEL 9 allocates a sufficient audit_backlog_limit to capture processes that start prior to the audit daemon with the following command:

$ sudo grubby --info=ALL | grep args | grep -v 'audit_backlog_limit=8192' 

If the command returns any outputs, and audit_backlog_limit is less than "8192", this is a finding.

Configure RHEL 9 to allocate sufficient audit_backlog_limit to capture processes that start prior to the audit daemon with the following command:

$ sudo grubby --update-kernel=ALL --args=audit_backlog_limit=8192

It is critical for the appropriate personnel to be aware if a system is at risk of failing to process audit logs as required. Without this notification, the security personnel may be unaware of an impending failure of the audit capability, and system operation may be adversely affected.

Audit processing failures include software/hardware errors, failures in the audit capturing mechanisms, and audit storage capacity being reached or exceeded.

This requirement applies to each audit data storage repository (i.e., distinct information system component where audit records are stored), the centralized audit storage capacity of organizations (i.e., all audit data storage repositories combined), or both.

Verify that RHEL 9 is configured to notify the appropriate interactive users in the event of an audit processing failure.

Find the alias maps that are being used with the following command:

$ postconf alias_maps 

alias_maps = hash:/etc/aliases

Query the Postfix alias maps for an alias for the root user with the following command:

$ postmap -q root hash:/etc/aliases
isso

If an alias is not set, this is a finding.

Edit the aliases map file (by default /etc/aliases) used by Postfix and configure a root alias (using the user ISSO as an example):

root:    ISSO

and then update the aliases database with the command:

$ sudo newaliases

"audispd-plugins" provides plugins for the real-time interface to the audit subsystem, "audispd". These plugins can do things like relay events to remote machines or analyze events for suspicious behavior.

Verify that RHEL 9 has the audispd-plugins package for installed with the following command:

$ sudo dnf list --installed audispd-plugins

Example output:

audispd-plugins.x86_64          3.0.7-101.el9_0.2

If the "audispd-plugins" package is not installed, this is a finding.

The audispd-plugins package can be installed with the following command:
 
$ sudo dnf install audispd-plugins

Misuse of privileged functions, either intentionally or unintentionally by authorized users, or by unauthorized external entities that have compromised information system accounts, is a serious and ongoing concern and can have significant adverse impacts on organizations. Auditing the use of privileged functions is one way to detect such misuse and identify the risk from insider threats and the advanced persistent threat.

Satisfies: SRG-OS-000326-GPOS-00126, SRG-OS-000327-GPOS-00127

Verify that RHEL 9 is configured to audit the execution of the "execve" system call with the following command:

$ sudo auditctl -l | grep execve

-a always,exit -F arch=b32 -S execve -C uid!=euid -F euid=0 -k execpriv 
-a always,exit -F arch=b64 -S execve -C uid!=euid -F euid=0 -k execpriv
-a always,exit -F arch=b32 -S execve -C gid!=egid -F egid=0 -k execpriv 
-a always,exit -F arch=b64 -S execve -C gid!=egid -F egid=0 -k execpriv

If the command does not return all lines, or the lines are commented out, this is a finding.

Configure RHEL 9 to audit the execution of the "execve" system call.

Add or update the following file system rules to "/etc/audit/rules.d/audit.rules":

-a always,exit -F arch=b32 -S execve -C uid!=euid -F euid=0 -k execpriv 
-a always,exit -F arch=b64 -S execve -C uid!=euid -F euid=0 -k execpriv
-a always,exit -F arch=b32 -S execve -C gid!=egid -F egid=0 -k execpriv 
-a always,exit -F arch=b64 -S execve -C gid!=egid -F egid=0 -k execpriv 

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000064-GPOS-00033, SRG-OS-000466-GPOS-00210, SRG-OS-000458-GPOS-00203

Verify that RHEL 9 is configured to audit the execution of the "chmod", "fchmod", and "fchmodat" system calls with the following command:

$ sudo auditctl -l | grep chmod

-a always,exit -F arch=b32 -S chmod,fchmod,fchmodat -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S chmod,fchmod,fchmodat -F auid>=1000 -F auid!=unset -k perm_mod

If both the "b32" and "b64" audit rules are not defined for the "chmod", "fchmod", and "fchmodat" system calls, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chmod", "fchmod", and "fchmodat" syscalls.

Add or update the following rules in "/etc/audit/rules.d/audit.rules":

-a always,exit -F arch=b32 -S chmod,fchmod,fchmodat -F auid>=1000 -F auid!=unset -k perm_mod

-a always,exit -F arch=b64 -S chmod,fchmod,fchmodat -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000064-GPOS-00033, SRG-OS-000466-GPOS-00210, SRG-OS-000458-GPOS-00203, SRG-OS-000474-GPOS-00219

Verify that RHEL 9 is configured to audit the execution of the  "chown", "fchown", "fchownat", and "lchown" system calls with the following command:

$ sudo auditctl -l | grep chown

-a always,exit -F arch=b32 -S chown,fchown,fchownat,lchown -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S chown,fchown,fchownat,lchown -F auid>=1000 -F auid!=unset -k perm_mod

If both the "b32" and "b64" audit rules are not defined for the "chown", "fchown", "fchownat", and "lchown" system calls, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chown", "fchown", "fchownat", and "lchown"" system calls.

Add or update the following rules in "/etc/audit/rules.d/audit.rules":

-a always,exit -F arch=b32 -S chown,fchown,fchownat,lchown -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S chown,fchown,fchownat,lchown -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000458-GPOS-00203, SRG-OS-000462-GPOS-00206, SRG-OS-000463-GPOS-00207, SRG-OS-000471-GPOS-00215, SRG-OS-000474-GPOS-00219, SRG-OS-000466-GPOS-00210, SRG-OS-000064-GPOS-00033

Verify that RHEL 9 is configured to audit the execution of the "setxattr", "fsetxattr", "lsetxattr", "removexattr", "fremovexattr", and "lremovexattr" system calls with the following command:

$ sudo auditctl -l | grep xattr

-a always,exit -F arch=b32 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid>=1000 -F auid!=unset -k perm_mod

-a always,exit -F arch=b32 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid=0 -k perm_mod
-a always,exit -F arch=b64 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid=0 -k perm_mod

If both the "b32" and "b64" audit rules are not defined for the "chmod", "fchmod", and "fchmodat" system calls "setxattr", "fsetxattr", "lsetxattr", "removexattr", "fremovexattr", and "lremovexattr" system calls, or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to audit the execution of the "setxattr", "fsetxattr", "lsetxattr", "removexattr", "fremovexattr", and "lremovexattr" system calls by adding or updating the following lines to "/etc/audit/rules.d/audit.rules":

-a always,exit -F arch=b32 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid>=1000 -F auid!=unset -k perm_mod

-a always,exit -F arch=b32 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid=0 -k perm_mod
-a always,exit -F arch=b64 -S setxattr,fsetxattr,lsetxattr,removexattr,fremovexattr,lremovexattr -F auid=0 -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "umount" command with the following command:

$ sudo auditctl -l | grep umount

-a always,exit -F path=/usr/bin/umount -F perm=x -F auid>=1000 -F auid!=unset -k privileged-mount

If the command does not return an audit rule for "umount" or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "umount" command by adding or updating the following rules in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/umount -F perm=x -F auid>=1000 -F auid!=unset -k privileged-mount

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000466-GPOS-00210

Verify that RHEL 9 is configured to audit the execution of the "chacl" command with the following command:

$ sudo auditctl -l | grep chacl

-a always,exit -F path=/usr/bin/chacl -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chacl" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/chacl -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "setfacl" command with the following command:

$ sudo auditctl -l | grep setfacl

-a always,exit -F path=/usr/bin/setfacl -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "setfacl" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/setfacl -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account that is being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000468-GPOS-00212, SRG-OS-000471-GPOS-00215, SRG-OS-000463-GPOS-00207, SRG-OS-000465-GPOS-00209

Verify that RHEL 9 is configured to audit the execution of the "chcon" command with the following command:

$ sudo auditctl -l | grep chcon

-a always,exit -F path=/usr/bin/chcon -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chcon" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/chcon -F perm=x -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000463-GPOS-00207, SRG-OS-000465-GPOS-00209

Verify that RHEL 9 is configured to audit the execution of the "semanage" command with the following command:

$ sudo auditctl -l | grep semanage

-a always,exit -F path=/usr/sbin/semanage -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "semanage" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/semanage -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000463-GPOS-00207, SRG-OS-000465-GPOS-00209

Verify that RHEL 9 is configured to audit the execution of the "setfiles" command with the following command:

$ sudo auditctl -l | grep setfiles

-a always,exit -F path=/usr/sbin/setfiles -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "setfiles" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/setfiles -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000463-GPOS-00207, SRG-OS-000465-GPOS-00209

Verify that RHEL 9 is configured to audit the execution of the "setsebool" command with the following command:

$ sudo auditctl -l | grep setsebool

 -a always,exit -F path=/usr/sbin/setsebool -F perm=x -F auid>=1000 -F auid!=unset -F key=privileged 

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate an audit event for any successful/unsuccessful use of the "setsebool " command by adding or updating the following rules in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F path=/usr/sbin/setsebool -F perm=x -F auid>=1000 -F auid!=unset -F key=privileged 

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000466-GPOS-00210, SRG-OS-000467-GPOS-00211, SRG-OS-000468-GPOS-00212

Verify that RHEL 9 is configured to audit successful/unsuccessful attempts to use the "rename", "unlink", "rmdir", "renameat", and "unlinkat" system calls with the following command:

$ sudo auditctl -l | grep 'rename\|unlink\|rmdir'

-a always,exit -F arch=b32 -S rename,unlink,rmdir,renameat,unlinkat -F auid>=1000 -F auid!=unset -k delete
-a always,exit -F arch=b64 -S rename,unlink,rmdir,renameat,unlinkat -F auid>=1000 -F auid!=unset -k delete

If the command does not return an audit rule for "rename", "unlink", "rmdir", "renameat", and "unlinkat" or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to generate an audit event for any successful/unsuccessful use of the "rename", "unlink", "rmdir", "renameat", and "unlinkat" system calls by adding or updating the following rules in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F arch=b32 -S rename,unlink,rmdir,renameat,unlinkat -F auid>=1000 -F auid!=unset -k delete
-a always,exit -F arch=b64 -S rename,unlink,rmdir,renameat,unlinkat -F auid>=1000 -F auid!=unset -k delete

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000064-GPOS-00033, SRG-OS-000458-GPOS-00203, SRG-OS-000461-GPOS-00205

Verify that RHEL 9 is configured to audit successful/unsuccessful attempts to use the "truncate", "ftruncate", "creat", "open", "openat", and "open_by_handle_at" system calls with the following command:

$ sudo auditctl -l | grep 'open\|truncate\|creat' 

-a always,exit -F arch=b32 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EPERM -F auid>=1000 -F auid!=unset -k perm_access
-a always,exit -F arch=b64 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EPERM -F auid>=1000 -F auid!=unset -k perm_access

-a always,exit -F arch=b32 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EACCES -F auid>=1000 -F auid!=unset -k perm_access
-a always,exit -F arch=b64 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EACCES -F auid>=1000 -F auid!=unset -k perm_access

If the output does not produce rules containing "-F exit=-EPERM", this is a finding.

If the output does not produce rules containing "-F exit=-EACCES", this is a finding.

If the command does not return an audit rule for "truncate", "ftruncate", "creat", "open", "openat", and "open_by_handle_at" or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to generate an audit event for any successful/unsuccessful use of the "truncate", "ftruncate", "creat", "open", "openat", and "open_by_handle_at" system calls by adding or updating the following rules in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F arch=b32 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EPERM -F auid>=1000 -F auid!=unset -k perm_access
-a always,exit -F arch=b64 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EPERM -F auid>=1000 -F auid!=unset -k perm_access

-a always,exit -F arch=b32 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EACCES -F auid>=1000 -F auid!=unset -k perm_access
-a always,exit -F arch=b64 -S truncate,ftruncate,creat,open,openat,open_by_handle_at -F exit=-EACCES -F auid>=1000 -F auid!=unset -k perm_access

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000471-GPOS-00216, SRG-OS-000477-GPOS-00222

Verify that RHEL 9 is configured to audit the execution of the "delete_module" system call with the following command:

$ sudo auditctl -l | grep delete_module

-a always,exit -F arch=b32 -S delete_module -F auid>=1000 -F auid!=unset -k module_chng
-a always,exit -F arch=b64 -S delete_module -F auid>=1000 -F auid!=unset -k module_chng

If both the "b32" and "b64" audit rules are not defined for the "delete_module" system call, or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to generate an audit event for any successful/unsuccessful use of the "delete_module" system call by adding or updating the following rules in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F arch=b32 -S delete_module -F auid>=1000 -F auid!=unset -k module_chng
-a always,exit -F arch=b64 -S delete_module -F auid>=1000 -F auid!=unset -k module_chng

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000471-GPOS-00216, SRG-OS-000477-GPOS-00222

Verify that RHEL 9 is configured to audit the execution of the "init_module" and "finit_module" system calls with the following command:

$ sudo auditctl -l | grep init_module

-a always,exit -F arch=b32 -S init_module,finit_module -F auid>=1000 -F auid!=unset -k module_chng
-a always,exit -F arch=b64 -S init_module,finit_module -F auid>=1000 -F auid!=unset -k module_chng

If both the "b32" and "b64" audit rules are not defined for the "delete_module" system call, or any of the lines returned are commented out, this is a finding.

Configure RHEL 9 to generate an audit event for any successful/unsuccessful use of the "init_module" and "finit_module" system calls by adding or updating the following rules in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F arch=b32 -S init_module,finit_module -F auid>=1000 -F auid!=unset -k module_chng
-a always,exit -F arch=b64 -S init_module,finit_module -F auid>=1000 -F auid!=unset -k module_chng

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000468-GPOS-00212, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "chage" command with the following command:

$ sudo auditctl -l | grep chage

-a always,exit -F path=/usr/bin/chage -F perm=x -F auid>=1000 -F auid!=unset -k privileged-chage

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chage" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/chage -F perm=x -F auid>=1000 -F auid!=unset -k privileged-chage

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "chsh" command with the following command:

$ sudo auditctl -l | grep chsh

-a always,exit -F path=/usr/bin/chsh -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "chsh" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/chsh -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "crontab" command with the following command:

$ sudo auditctl -l | grep crontab

-a always,exit -F path=/usr/bin/crontab -F perm=x -F auid>=1000 -F auid!=unset -k privileged-crontab

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "crontab" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/crontab -F perm=x -F auid>=1000 -F auid!=unset -k privileged-crontab

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "gpasswd" command with the following command:

$ sudo auditctl -l | grep gpasswd

-a always,exit -F path=/usr/bin/gpasswd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-gpasswd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "gpasswd" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/gpasswd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-gpasswd

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000471-GPOS-00216, SRG-OS-000477-GPOS-00222

Verify that RHEL 9 is configured to audit the execution of the "kmod" command with the following command:

$ sudo auditctl -l | grep kmod

-a always,exit -F path=/usr/bin/kmod -F perm=x -F auid>=1000 -F auid!=unset -k modules

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "kmod" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/kmod -F perm=x -F auid>=1000 -F auid!=unset -k modules

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account that is being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "newgrp" command with the following command:

$ sudo auditctl -l | grep newgrp

-a always,exit -F path=/usr/bin/newgrp -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "newgrp" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/newgrp -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "pam_timestamp_check" command with the following command:

$ sudo auditctl -l | grep timestamp

-a always,exit -F path=/usr/sbin/pam_timestamp_check -F perm=x -F auid>=1000 -F auid!=unset -k privileged-pam_timestamp_check

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "pam_timestamp_check" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/pam_timestamp_check -F perm=x -F auid>=1000 -F auid!=unset -k privileged-pam_timestamp_check

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/gshadow" with the following command:

$ sudo auditctl -l | egrep '(/usr/bin/passwd)' 

-a always,exit -F path=/usr/bin/passwd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-passwd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "passwd" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/passwd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-passwd

The audit daemon must be restarted for the changes to take effect.

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "postdrop" command with the following command:

$ sudo auditctl -l | grep postdrop

-a always,exit -F path=/usr/sbin/postdrop -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "postdrop" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/postdrop -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "postqueue" command with the following command:

$ sudo auditctl -l | grep postqueue

-a always,exit -F path=/usr/sbin/postqueue -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "postqueue" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/postqueue -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "ssh-agent" command with the following command:

$ sudo auditctl -l | grep ssh-agent

-a always,exit -F path=/usr/bin/ssh-agent -F perm=x -F auid>=1000 -F auid!=unset -k privileged-ssh

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "ssh-agent" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/ssh-agent -F perm=x -F auid>=1000 -F auid!=unset -k privileged-ssh

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "ssh-keysign" command with the following command:

$ sudo auditctl -l | grep ssh-keysign

-a always,exit -F path=/usr/libexec/openssh/ssh-keysign -F perm=x -F auid>=1000 -F auid!=unset -k privileged-ssh

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "ssh-keysign" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/libexec/openssh/ssh-keysign -F perm=x -F auid>=1000 -F auid!=unset -k privileged-ssh

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000064-GPOS-00033, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000466-GPOS-00210

Verify that RHEL 9 is configured to audit the execution of the "su" command with the following command:

$ sudo auditctl -l | grep /usr/bin/su

-a always,exit -F path=/usr/bin/su -F perm=x -F auid>=1000 -F auid!=unset -k privileged-priv_change

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "su" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/su -F perm=x -F auid>=1000 -F auid!=unset -k privileged-priv_change

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000466-GPOS-00210

Verify that RHEL 9 is configured to audit the execution of the "sudo" command with the following command:

$ sudo auditctl -l | grep /usr/bin/sudo

-a always,exit -F path=/usr/bin/sudo -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "sudo" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/sudo -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "sudoedit" command with the following command:

$ sudo auditctl -l | grep /usr/bin/sudoedit

-a always,exit -F path=/usr/bin/sudoedit -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "sudoedit" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/sudoedit -F perm=x -F auid>=1000 -F auid!=unset -k priv_cmd

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "unix_chkpwd" command with the following command:

$ sudo auditctl -l | grep unix_chkpwd

-a always,exit -F path=/usr/sbin/unix_chkpwd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "unix_chkpwd" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/unix_chkpwd -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000064-GPOS-00033, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "unix_update" command with the following command:

$ sudo auditctl -l | grep unix_update

-a always,exit -F path=/usr/sbin/unix_update -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "unix_update" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/unix_update -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "userhelper" command with the following command:

$ sudo auditctl -l | grep userhelper

-a always,exit -F path=/usr/sbin/userhelper -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "userhelper" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/userhelper -F perm=x -F auid>=1000 -F auid!=unset -k privileged-unix-update

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit record specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000466-GPOS-00210

Verify that RHEL 9 is configured to audit the execution of the "usermod" command with the following command:

$ sudo auditctl -l | grep usermod

-a always,exit -F path=/usr/sbin/usermod -F perm=x -F auid>=1000 -F auid!=unset -k privileged-usermod

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "usermod " command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/sbin/usermod -F perm=x -F auid>=1000 -F auid!=unset -k privileged-usermod

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records that are specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Audit records can be generated from various components within the information system (e.g., module or policy filter).

When a user logs on, the auid is set to the uid of the account that is being authenticated. Daemons are not user sessions and have the loginuid set to -1. The auid representation is an unsigned 32-bit integer, which equals 4294967295. The audit system interprets -1, 4294967295, and "unset" in the same way.

The system call rules are loaded into a matching engine that intercepts each system call made by all programs on the system. Therefore, it is very important to use system call rules only when absolutely necessary since these affect performance. The more rules, the bigger the performance hit. The performance can be helped, however, by combining system calls into one rule whenever possible.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 is configured to audit the execution of the "mount" command with the following command:

$ sudo auditctl -l | grep /usr/bin/mount

-a always,exit -F path=/usr/bin/mount -F perm=x -F auid>=1000 -F auid!=unset -k privileged-mount

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records upon successful/unsuccessful attempts to use the "mount" command by adding or updating the following rule in "/etc/audit/rules.d/audit.rules":

-a always,exit -F path=/usr/bin/mount -F perm=x -F auid>=1000 -F auid!=unset -k privileged-mount

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Misuse of the init command may cause availability issues for the system.

Verify that RHEL 9 is configured to audit the execution of the "init" command with the following command:

$ sudo auditctl -l | grep init

-a always,exit -F path=/usr/sbin/init -F perm=x -F auid>=1000 -F auid!=unset -k privileged-init

If the command does not return a line, or the line is commented out, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful uses of the "init" command by adding or updating the following rule in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F path=/usr/sbin/init -F perm=x -F auid>=1000 -F auid!=unset -k privileged-init

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Misuse of the poweroff command may cause availability issues for the system.

Verify that RHEL 9 is configured to audit the execution of the "poweroff" command with the following command:

$ sudo auditctl -l | grep poweroff

-a always,exit -F path=/usr/sbin/poweroff -F perm=x -F auid>=1000 -F auid!=unset -k privileged-poweroff

If the command does not return a line, or the line is commented out, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful uses of the "poweroff" command by adding or updating the following rule in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F path=/usr/sbin/poweroff -F perm=x -F auid>=1000 -F auid!=unset -k privileged-poweroff

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Misuse of the reboot command may cause availability issues for the system.

Verify that RHEL 9 is configured to audit the execution of the "reboot" command with the following command:

$ sudo auditctl -l | grep reboot

-a always,exit -F path=/usr/sbin/reboot -F perm=x -F auid>=1000 -F auid!=unset -k privileged-reboot

If the command does not return a line, or the line is commented out, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful uses of the "reboot" command by adding or updating the following rule in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F path=/usr/sbin/reboot -F perm=x -F auid>=1000 -F auid!=unset -k privileged-reboot

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Misuse of the shutdown command may cause availability issues for the system.

Verify that RHEL 9 is configured to audit the execution of the "shutdown" command with the following command:

$ sudo cat /etc/audit/rules.d/* | grep shutdown

-a always,exit -F path=/usr/sbin/shutdown -F perm=x -F auid>=1000 -F auid!=unset -k privileged-shutdown

If the command does not return a line, or the line is commented out, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful uses of the "shutdown" command by adding or updating the following rule in the "/etc/audit/rules.d/audit.rules" file:

-a always,exit -F path=/usr/sbin/shutdown -F perm=x -F auid>=1000 -F auid!=unset -k privileged-shutdown

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

The changing of file permissions could indicate that a user is attempting to gain access to information that would otherwise be disallowed. Auditing DAC modifications can facilitate the identification of patterns of abuse among both authorized and unauthorized users.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

Verify that RHEL 9 generates an audit record for all uses of the "umount" and system call with the following command:

$ sudo grep "umount" /etc/audit/audit.* 

If the system is configured to audit this activity, it will return a line like the following:

-a always,exit -F arch=b32 -S umount -F auid>=1000 -F auid!=unset -k privileged-umount

If the command does not return a line, or the line is commented out, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful use of the "umount" system call by adding or updating the following rules in "/etc/audit/audit.rules" and adding the following rules to "/etc/audit/rules.d/perm_mod.rules" or updating the existing rules in files in the "/etc/audit/rules.d/" directory:

-a always,exit -F arch=b32 -S umount -F auid>=1000 -F auid!=unset -k privileged-umount

The audit daemon must be restarted for the changes to take effect.

The changing of file permissions could indicate that a user is attempting to gain access to information that would otherwise be disallowed. Auditing DAC modifications can facilitate the identification of patterns of abuse among both authorized and unauthorized users.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215

To determine if the system is configured to audit calls to the  umount2 system call, run the following command:

$ sudo grep "umount2" /etc/audit/audit.* 

If the system is configured to audit this activity, it will return a line.

If no line is returned, this is a finding.

Configure the audit system to generate an audit event for any successful/unsuccessful use of the "umount2" system call by adding or updating the following rules in "/etc/audit/audit.rules" and adding the following rules to "/etc/audit/rules.d/perm_mod.rules" or updating the existing rules in files in the "/etc/audit/rules.d/" directory:

-a always,exit -F arch=b32 -S umount2 -F auid>=1000 -F auid!=unset -k perm_mod
-a always,exit -F arch=b64 -S umount2 -F auid>=1000 -F auid!=unset -k perm_mod

The audit daemon must be restarted for the changes to take effect.

The actions taken by system administrators must be audited to keep a record of what was executed on the system, as well as for accountability purposes. Editing the sudoers file may be sign of an attacker trying to establish persistent methods to a system, auditing the editing of the sudoers files mitigates this risk.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/sudoers" with the following command:

$ sudo auditctl -l | grep /etc/sudoers

-w /etc/sudoers -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/sudoers".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/sudoers -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

The actions taken by system administrators must be audited to keep a record of what was executed on the system, as well as for accountability purposes. Editing the sudoers file may be sign of an attacker trying to establish persistent methods to a system, auditing the editing of the sudoers files mitigates this risk.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/sudoers.d/" with the following command:

$ sudo auditctl -l | grep /etc/sudoers.d

-w /etc/sudoers.d/ -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/sudoers.d/".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/sudoers.d/ -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

In addition to auditing new user and group accounts, these watches will alert the system administrator(s) to any modifications. Any unexpected users, groups, or modifications must be investigated for legitimacy.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/group" with the following command:

$ sudo auditctl -l | egrep '(/etc/group)'  

-w /etc/group -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/group".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/group -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

In addition to auditing new user and group accounts, these watches will alert the system administrator(s) to any modifications. Any unexpected users, groups, or modifications should be investigated for legitimacy.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/gshadow" with the following command:

$ sudo auditctl -l | egrep '(/etc/gshadow)' 

-w /etc/gshadow -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/gshadow".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/gshadow -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

In addition to auditing new user and group accounts, these watches will alert the system administrator(s) to any modifications. Any unexpected users, groups, or modifications should be investigated for legitimacy.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/security/opasswd" with the following command:

$ sudo auditctl -l | egrep '(/etc/security/opasswd)' 

-w /etc/security/opasswd -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/security/opasswd".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/security/opasswd -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

In addition to auditing new user and group accounts, these watches will alert the system administrator(s) to any modifications. Any unexpected users, groups, or modifications should be investigated for legitimacy.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221, SRG-OS-000274-GPOS-00104, SRG-OS-000275-GPOS-00105, SRG-OS-000276-GPOS-00106, SRG-OS-000277-GPOS-00107

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/passwd" with the following command:

$ sudo auditctl -l | egrep '(/etc/passwd)' 

-w /etc/passwd -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/passwd".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/passwd -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

In addition to auditing new user and group accounts, these watches will alert the system administrator(s) to any modifications. Any unexpected users, groups, or modifications should be investigated for legitimacy.

Satisfies: SRG-OS-000004-GPOS-00004, SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000304-GPOS-00121, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000470-GPOS-00214, SRG-OS-000471-GPOS-00215, SRG-OS-000239-GPOS-00089, SRG-OS-000240-GPOS-00090, SRG-OS-000241-GPOS-00091, SRG-OS-000303-GPOS-00120, SRG-OS-000466-GPOS-00210, SRG-OS-000476-GPOS-00221

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/etc/passwd with the following command:

$ sudo auditctl -l | egrep '(/etc/shadow)' 

-w /etc/shadow -p wa -k identity

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/etc/shadow".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /etc/shadow -p wa -k identity

The audit daemon must be restarted for the changes to take effect.

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Satisfies: SRG-OS-000392-GPOS-00172, SRG-OS-000470-GPOS-00214, SRG-OS-000473-GPOS-00218

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/faillock" with the following command:

$ sudo auditctl -l | grep /var/log/faillock

-w /var/log/faillock -p wa -k logins

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/faillock".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /var/log/faillock -p wa -k logins

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Satisfies: SRG-OS-000037-GPOS-00015, SRG-OS-000042-GPOS-00020, SRG-OS-000062-GPOS-00031, SRG-OS-000392-GPOS-00172, SRG-OS-000462-GPOS-00206, SRG-OS-000471-GPOS-00215, SRG-OS-000473-GPOS-00218, SRG-OS-000470-GPOS-00214

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/lastlog" with the following command:

$ sudo auditctl -l | grep /var/log/lastlog
 
-w /var/log/lastlog -p wa -k logins

If the command does not return a line, or the line is commented out, this is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/lastlog".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /var/log/lastlog -p wa -k logins

The audit daemon must be restarted for the changes to take effect.

$ sudo service auditd restart

Without generating audit records specific to the security and mission needs of the organization, it would be difficult to establish, correlate, and investigate the events relating to an incident or identify those responsible for one.

Satisfies: SRG-OS-000392-GPOS-00172, SRG-OS-000470-GPOS-00214, SRG-OS-000473-GPOS-00218

Verify RHEL 9 generates audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/tallylog" with the following command:

$ sudo auditctl -l | grep /var/log/tallylog

-w /var/log/tallylog -p wa -k logins

If the command does not return a line, or the line is commented out, is a finding.

Configure RHEL 9 to generate audit records for all account creations, modifications, disabling, and termination events that affect "/var/log/tallylog".

Add or update the following file system rule to "/etc/audit/rules.d/audit.rules":

-w /var/log/tallylog -p wa -k logins

The audit daemon must be restarted for the changes to take effect.

It is critical for the appropriate personnel to be aware if a system is at risk of failing to process audit logs as required. Without this notification, the security personnel may be unaware of an impending failure of the audit capability, and system operation may be adversely affected.

Audit processing failures include software/hardware errors, failures in the audit capturing mechanisms, and audit storage capacity being reached or exceeded.

Satisfies: SRG-OS-000046-GPOS-00022, SRG-OS-000047-GPOS-00023

Verify the audit service is configured to panic on a critical error with the following command:

$ sudo grep "\-f" /etc/audit/audit.rules 

-f 2

If the value for "-f" is not "2", and availability is not documented as an overriding concern, this is a finding.

Configure RHEL 9 to shut down when auditing failures occur.

Add the following line to the bottom of the /etc/audit/rules.d/audit.rules file:

-f 2

If modification of login user identifiers (UIDs) is not prevented, they can be changed by nonprivileged users and make auditing complicated or impossible.

Satisfies: SRG-OS-000462-GPOS-00206, SRG-OS-000475-GPOS-00220, SRG-OS-000057-GPOS-00027, SRG-OS-000058-GPOS-00028, SRG-OS-000059-GPOS-00029

Verify the audit system prevents unauthorized changes to logon UIDs with the following command:

$ sudo grep -i immutable /etc/audit/audit.rules

--loginuid-immutable

If the "--loginuid-immutable" option is not returned in the "/etc/audit/audit.rules", or the line is commented out, this is a finding.

Configure RHEL 9 auditing to prevent modification of login UIDs once they are set by adding the following line to /etc/audit/rules.d/audit.rules:

--loginuid-immutable

The audit daemon must be restarted for the changes to take effect.

Unauthorized disclosure of audit records can reveal system and configuration data to attackers, thus compromising its confidentiality.

Audit information includes all information (e.g., audit records, audit settings, audit reports) needed to successfully audit RHEL 9 system activity.

In immutable mode, unauthorized users cannot execute changes to the audit system to potentially hide malicious activity and then put the audit rules back.  A system reboot would be noticeable, and a system administrator could then investigate the unauthorized changes.

Satisfies: SRG-OS-000057-GPOS-00027, SRG-OS-000058-GPOS-00028, SRG-OS-000059-GPOS-00029

Verify the audit system prevents unauthorized changes with the following command:

$ sudo grep "^\s*[^#]" /etc/audit/audit.rules | tail -1

-e 2

If the audit system is not set to be immutable by adding the "-e 2" option to the end of "/etc/audit/audit.rules", this is a finding.

Configure the audit system to set the audit rules to be immutable by adding the following line to end of "/etc/audit/rules.d/audit.rules"

-e 2

The audit daemon must be restarted for the changes to take effect.

Use of weak or untested encryption algorithms undermines the purposes of utilizing encryption to protect data. The operating system must implement cryptographic modules adhering to the higher standards approved by the federal government since this provides assurance they have been tested and validated. This includes NIST FIPS-validated cryptography for the following: Provisioning digital signatures, generating cryptographic hashes, and to protect data requiring data-at-rest protections in accordance with applicable federal laws, Executive Orders, directives, policies, regulations, and standards.

Satisfies: SRG-OS-000033-GPOS-00014, SRG-OS-000125-GPOS-00065, SRG-OS-000396-GPOS-00176, SRG-OS-000423-GPOS-00187, SRG-OS-000478-GPOS-00223

Verify that RHEL 9 is in FIPS mode with the following command:

$ sudo fips-mode-setup --check

FIPS mode is enabled.

If FIPS mode is not enabled, this is a finding.

Configure the operating system to implement FIPS mode with the following command

$ sudo fips-mode-setup --enable

Reboot the system for the changes to take effect.

The system must use a strong hashing algorithm to store the password.

Passwords need to be protected at all times, and encryption is the standard method for protecting passwords. If passwords are not encrypted, they can be plainly read (i.e., clear text) and easily compromised.

Satisfies: SRG-OS-000073-GPOS-00041, SRG-OS-000120-GPOS-00061

Verify that the interactive user account passwords are using a strong password hash with the following command:

$ sudo cut -d: -f2 /etc/shadow

$6$kcOnRq/5$NUEYPuyL.wghQwWssXRcLRFiiru7f5JPV6GaJhNC2aK5F3PZpE/BCCtwrxRc/AInKMNX3CdMw11m9STiql12f/ 

Password hashes "!" or "*" indicate inactive accounts not available for logon and are not evaluated.

If any interactive user password hash does not begin with "$6", this is a finding.

Lock all interactive user accounts not using SHA-512 hashing until the passwords can be regenerated with SHA-512.

Overriding the system crypto policy makes the behavior of the Libreswan service violate expectations, and makes system configuration more fragmented.

Verify that the IPsec service uses the system crypto policy with the following command:

Note: If the ipsec service is not installed, this requirement is Not Applicable.

$ sudo grep include /etc/ipsec.conf /etc/ipsec.d/*.conf

/etc/ipsec.conf:include /etc/crypto-policies/back-ends/libreswan.config

If the ipsec configuration file does not contain "include /etc/crypto-policies/back-ends/libreswan.config", this is a finding.

Configure Libreswan to use the system cryptographic policy.

Add the following line to "/etc/ipsec.conf":

include /etc/crypto-policies/back-ends/libreswan.config

Unapproved mechanisms that are used for authentication to the cryptographic module are not verified and; therefore, cannot be relied upon to provide confidentiality or integrity, and DOD data may be compromised.

RHEL 9 systems utilizing encryption are required to use FIPS-compliant mechanisms for authenticating to cryptographic modules. 

FIPS 140-3 is the current standard for validating that mechanisms used to access cryptographic modules utilize authentication that meets DOD requirements. This allows for Security Levels 1, 2, 3, or 4 for use on a general-purpose computing system.

Verify that the pam_unix.so module is configured to use sha512 in /etc/pam.d/password-auth with the following command:

$ grep "^password.*pam_unix.so.*sha512" /etc/pam.d/password-auth

password sufficient pam_unix.so sha512

If "sha512" is missing, or the line is commented out, this is a finding.

If the system administrator (SA) can demonstrate that the required configuration is contained in a PAM configuration file included or substacked from the system-auth file, this is not a finding.

Configure RHEL 9 to use a FIPS 140-3 approved cryptographic hashing algorithm for system authentication.

Edit/modify the following line in the "/etc/pam.d/password-auth" file to include the sha512 option for pam_unix.so:

password sufficient pam_unix.so sha512

Centralized cryptographic policies simplify applying secure ciphers across an operating system and the applications that run on that operating system. Use of weak or untested encryption algorithms undermines the purposes of utilizing encryption to protect data.

Satisfies: SRG-OS-000396-GPOS-00176, SRG-OS-000393-GPOS-00173, SRG-OS-000394-GPOS-00174

Verify that RHEL 9 crypto-policies package is installed with the following command:

$ sudo dnf list --installed crypto-policies

Example output:

crypto-policies.noarch          20220223-1.git5203b41.el9_0.1

If the "crypto-policies" package is not installed, this is a finding.

Install the crypto-policies package (if the package is not already installed) with the following command:

$ sudo dnf install crypto-policies

Centralized cryptographic policies simplify applying secure ciphers across an operating system and the applications that run on that operating system. Use of weak or untested encryption algorithms undermines the purposes of utilizing encryption to protect data.

Satisfies: SRG-OS-000396-GPOS-00176, SRG-OS-000393-GPOS-00173, SRG-OS-000394-GPOS-00174

Verify that RHEL 9 custom crypto policies are loaded correctly.

List all of the crypto backends configured on the system.

$ ls -l /etc/crypto-policies/back-ends/ 
lrwxrwxrwx. 1 root root 40 Oct  7 08:44 bind.config -> /usr/share/crypto-policies/FIPS/bind.txt
lrwxrwxrwx. 1 root root 42 Oct  7 08:44 gnutls.config -> /usr/share/crypto-policies/FIPS/gnutls.txt
lrwxrwxrwx. 1 root root 40 Oct  7 08:44 java.config -> /usr/share/crypto-policies/FIPS/java.txt
lrwxrwxrwx. 1 root root 46 Oct  7 08:44 javasystem.config -> /usr/share/crypto-policies/FIPS/javasystem.txt
lrwxrwxrwx. 1 root root 40 Oct  7 08:44 krb5.config -> /usr/share/crypto-policies/FIPS/krb5.txt
lrwxrwxrwx. 1 root root 45 Oct  7 08:44 libreswan.config -> /usr/share/crypto-policies/FIPS/libreswan.txt
lrwxrwxrwx. 1 root root 42 Oct  7 08:44 libssh.config -> /usr/share/crypto-policies/FIPS/libssh.txt
lrwxrwxrwx. 1 root root 39 Oct  7 08:44 nss.config -> /usr/share/crypto-policies/FIPS/nss.txt
lrwxrwxrwx. 1 root root 43 Oct  7 08:44 openssh.config -> /usr/share/crypto-policies/FIPS/openssh.txt
lrwxrwxrwx. 1 root root 49 Oct  7 08:44 opensshserver.config -> /usr/share/crypto-policies/FIPS/opensshserver.txt
lrwxrwxrwx. 1 root root 46 Oct  7 08:44 opensslcnf.config -> /usr/share/crypto-policies/FIPS/opensslcnf.txt
lrwxrwxrwx. 1 root root 43 Oct  7 08:44 openssl.config -> /usr/share/crypto-policies/FIPS/openssl.txt

If the paths do not point the respective files under /usr/share/crypto-policies/FIPS path, this is a finding.

Configure RHEL 9 to FIPS crypto policy.

$ sudo ln -s /usr/share/crypto-policies/FIPS/.txt /etc/crypto-policies/back-ends/.conf

Replace  with every service that is not set to FIPS.

The system must be rebooted to make the changes to take effect.

Overriding the system crypto policy makes the behavior of Kerberos violate expectations, and makes system configuration more fragmented.

Verify that the symlink exists and targets the correct Kerberos crypto policy, with the following command:

file /etc/crypto-policies/back-ends/krb5.config

If command output shows the following line, Kerberos is configured to use the system-wide crypto policy:

/etc/crypto-policies/back-ends/krb5.config: symbolic link to /usr/share/crypto-policies/FIPS/krb5.txt

If the symlink does not exist or points to a different target, this is a finding.

Configure Kerberos to use system crypto policy.

Create a symlink pointing to system crypto policy in the Kerberos configuration using the following command:

$ sudo ln -s /etc/crypto-policies/back-ends/krb5.config /usr/share/crypto-policies/FIPS/krb5.txt

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Transport Layer Security (TLS) encryption is a required security setting as a number of known vulnerabilities have been reported against Secure Sockets Layer (SSL) and earlier versions of TLS. Encryption of private information is essential to ensuring data confidentiality. If private information is not encrypted, it can be intercepted and easily read by an unauthorized party. SQL Server must use a minimum of FIPS 140-3 approved TLS version 1.2, and all non-FIPS-approved SSL and TLS versions must be disabled. NIST 800-53 specifies the preferred configurations for government systems.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

Satisfies: SRG-OS-000250-GPOS-00093, SRG-OS-000423-GPOS-00187

Verify if GnuTLS uses defined DOD-approved TLS Crypto Policy with the following command:

 $ update-crypto-policies --show
FIPS

If the system wide crypto policy is not set to "FIPS", this is a finding.

Configure the RHEL 9 GnuTLS library to use only NIST-approved encryption with the following steps to enable FIPS mode:

$ sudo fips-mode-setup --enable

A reboot is required for the changes to take effect.

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/openssl.config file.

Verify that RHEL 9 OpenSSL library is configured to use only ciphers employing FIPS 140-3 approved algorithms with the following command:

$ sudo grep -i opensslcnf.config /etc/pki/tls/openssl.cnf

.include = /etc/crypto-policies/back-ends/opensslcnf.config

If the "opensslcnf.config" is not defined in the "/etc/pki/tls/openssl.cnf" file, this is a finding.

Configure the RHEL 9 OpenSSL library to use the system cryptographic policy.

Edit the "/etc/pki/tls/openssl.cnf" and add or modify the following line:

.include = /etc/crypto-policies/back-ends/opensslcnf.config

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Remote access (e.g., RDP) is access to DOD nonpublic information systems by an authorized user (or an information system) communicating through an external, nonorganization-controlled network. Remote access methods include, for example, dial-up, broadband, and wireless.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/openssl.config file.

Verify that RHEL 9 OpenSSL library is configured to use TLS 1.2 encryption or stronger with following command:

$ grep -i  minprotocol /etc/crypto-policies/back-ends/opensslcnf.config

TLS.MinProtocol = TLSv1.2
DTLS.MinProtocol = DTLSv1.2

If the "TLS.MinProtocol" is set to anything older than "TLSv1.2" or the "DTLS.MinProtocol" is set to anything older than "DTLSv1.2", this is a finding.

Configure the RHEL 9 OpenSSL library to use only DOD-approved TLS encryption by editing the following line in the "/etc/crypto-policies/back-ends/opensslcnf.config" file:

TLS.MinProtocol = TLSv1.2
DTLS.MinProtocol = DTLSv1.2

A reboot is required for the changes to take effect.

Centralized cryptographic policies simplify applying secure ciphers across an operating system and the applications that run on that operating system. Use of weak or untested encryption algorithms undermines the purposes of utilizing encryption to protect data.

Satisfies: SRG-OS-000396-GPOS-00176, SRG-OS-000393-GPOS-00173, SRG-OS-000394-GPOS-00174

Show the configured systemwide cryptographic policy by running the following command:

$ sudo update-crypto-policies --show
FIPS

If the main policy name is not "FIPS", this is a finding.

If the AD-SUPPORT subpolicy module is included (e.g., "FIPS:AD-SUPPORT"), and Active Directory support is not documented as an operational requirement with the information system security officer (ISSO), this is a finding.

If the NO-ENFORCE-EMS subpolicy module is included (e.g., "FIPS:NO-ENFORCE-EMS"), and not enforcing EMS is not documented as an operational requirement with the ISSO, this is a finding.

If any other subpolicy module is included, this is a finding.

Configure the FIPS policy as the systemwide cryptographic policy by running the following command:

$ sudo update-crypto-policies --set FIPS

Reboot the system to ensure that all aspects of the change take effect.

Without cryptographic integrity protections, information can be altered by unauthorized users without detection.

Cryptographic mechanisms used for protecting the integrity of information include, for example, signed hash functions using asymmetric cryptography enabling distribution of the public key to verify the hash information while maintaining the confidentiality of the secret key used to generate the hash.

RHEL 9 incorporates system-wide crypto policies by default. The employed algorithms can be viewed in the /etc/crypto-policies/back-ends/ directory.

Satisfies: SRG-OS-000423-GPOS-00187, SRG-OS-000426-GPOS-00190

Verify that BIND uses the system crypto policy with the following command:

Note: If the "bind" package is not installed, this requirement is Not Applicable.

$ sudo grep include /etc/named.conf 

include "/etc/crypto-policies/back-ends/bind.config";' 

If BIND is installed and the BIND config file doesn't contain the  include "/etc/crypto-policies/back-ends/bind.config" directive, or the line is commented out, this is a finding.

Configure BIND to use the system crypto policy.

Add the following line to the "options" section in "/etc/named.conf":

include "/etc/crypto-policies/back-ends/bind.config";