When remediating either of $subj profiles on a Secure Boot enabled system, the OS fails to reboot with:
[FAILED] Failed to mount /boot/efi.
See 'systemctl status boot-efi.mount' for details.
...
Starting default.target
You are in emergency mode. After logging in, type "journalctl -xb" to view
system logs, "systemctl reboot" to reboot, "systemctl default" or "exit"
to boot into default mode.
Give root password for maintenance
(or press Control-D to continue):
and the systemctl status says
mount: /boot/efi: unknown filesystem type 'vfat'.
This can also be reproduced manually:
[root@localhost ~]# mount /boot/efi
mount: /boot/efi: unknown filesystem type 'vfat'.
[root@localhost ~]# modprobe vfat
modprobe: ERROR: could not insert 'vfat': Operation not permitted
It seems that the vfat.ko module cannot be loaded at this stage because the kernel is in a Secure Boot Lockdown mode,
[root@localhost ~]# dmesg | grep -i secure
[ 0.000000] secureboot: Secure boot enabled
[ 0.000000] Kernel is locked down from EFI Secure Boot mode; see man kernel_lockdown.7
[ 0.002870] secureboot: Secure boot enabled
Which was remediated / enabled by sysctl_kernel_modules_disabled (or something similar).
but there are more problems caused by the lockdown mode - ie. the boot continues and locks up on /run/systemd/generator/systemd-cryptsetup@*.service.
So we probably should not be automatically remediating something as dangerous, so please disable Ansible remediation for sysctl_kernel_modules_disabled (and we can test if that helped).
Description of problem:
When remediating either of $subj profiles on a Secure Boot enabled system, the OS fails to reboot with:
and the
systemctl status
saysThis can also be reproduced manually:
It seems that the
vfat.ko
module cannot be loaded at this stage because the kernel is in a Secure Boot Lockdown mode,Which was remediated / enabled by
sysctl_kernel_modules_disabled
(or something similar).Turns out there is active Ansible remediation for the rule, but Bash remediation has been disabled by https://github.com/ComplianceAsCode/content/pull/6586. Well, this issue is probably the real cause why.
This particular problem can be fixed by the equivalent of
but there are more problems caused by the lockdown mode - ie. the boot continues and locks up on
/run/systemd/generator/systemd-cryptsetup@*.service
.So we probably should not be automatically remediating something as dangerous, so please disable Ansible remediation for
sysctl_kernel_modules_disabled
(and we can test if that helped).SCAP Security Guide Version:
master @ b79ef8779969e528749c653deb4d50ec5162fdb7
Operating System Version:
RHEL-9, probably RHEL-8, definitely RHEL-10