After updating the kernel on rocky 9 on a new install it fails to boot, system falls in to emergency mode due initrd-switch-root.service entered failed state.
I am able to select the older kernel to get it to boot fine.
I have removed dnf remove kernel-5.14.0-70.17.1.el9_0.x86_64 kernel-core-5.14.0-70.17.1.el9_0.x86_64 kernel-devel-5.14.0-70.17.1.el9_0.x86_64 kernel-modules-5.14.0-70.17.1.el9_0.x86_64
to get rid of the offending kernel
If you are running dnf update during a kickstart in the %post section, this is a known issue. Running a dnf update after a kickstart should work normally.
Yeah this issue affects stream 9 also. I’m actually writing up a bug report as we speak upstream to see if there’s documentation that explains this weird behavior. As a work around, you can probably do this (untested at the moment):
There is a comment on the red hat bugzilla that alludes to this behavior too: 1969362 – grubby: apply current arguments to future kernels - so even with the above, the root=… parameter isn’t being populated in /etc/default/grub, so my script above wouldn’t cut it.
As an additional edit, doing the below seemed to work, but it expects you to know what the volume/partitions/uuid’s are and such. If you’re doing kickstart installs, it’s likely you’ll know this information.