#[TEMPORARY SOLUTION] Kernel panic when upgrading to newer image
1 messages · Page 1 of 1 (latest)
which image, and on what hardware?
and did you run any post-install steps
No
No post install steps
Kinoite-nvidia-open-hardened
@sterile sonnet
A second
CPU: 13th Gen Intel(R) Core(TM) i7-1355U (12) @ 5.oo GHz
GPU 1: Nvidia GeForce RTX 3050 4GB Laptop GPU [Discrete]
GPU 2: Intel Iris Xe Graphics @ 1.30 GHz [Integrated]
interesting
thanks
unfortunately this doesn't give me any clear ideas
assuming zero changes were made aside from updating
that makes me inclined towards nvidia shipping a bug
I can send the error message from before kernel panic
you could try binary searching the date tags
which look like :20250810
instead of :latest
Excuse me, but how do I do that?
rpm-ostree rebase to your current change but change the tags like above
from latest to a date
like this
error: ../../grub-core/kern/mm.c:552:out of memory
Then it says to press any key to continue
Kernel panic error is: VFS: Unable to mount root fs on unknown block(0,0)
moment 
looks like this issue https://github.com/ublue-os/bazzite/issues/3147
@sterile sonnet do you have secure boot enabled
No
and our key enrolled
oh
well that will prevent nvidia drivers from loading
although it shouldn't cause the error in question
Ah understood
error: while pulling 20250810: Remote "kinoite-nvidia-open-hardened" not found
?
what command did you run
the rebase target should match the string in rpm-ostree status exactly aside from the tag
if you omit any part of it, it will break
Got it
Pulling currently
20250810 is okay
Testing 20250825
20250825 is okay
Testing 20250903
20250903 is not okay
Testing 20250829
20250829 is not okay
Testing 20250827
20250827 is okay
Testing 20250828
20250828 is okay
@junior garnet @regal jay the problem is between 20250828 and 20250829
seems like it must be an upstream issue then, there weren't any secureblue commits on August 28-29 that could cause anything like this
Excuse me, how would one track down the issue causing it and how would one sent the report to upstream?
run rpm-ostree db diff while having those two date tags in staged/current deployments
it will tell you which packages changed
I have found the temporary solution in the provided issue: disable the TPM chip in BIOS settings
I also checked the changed packages between 20250828 and 20250829, it was a kernel update
This issue
[TEMPORARY SOLUTION] Kernel panic when upgrading to newer image
Update: I have found the upstream issue on this problem, https://bugzilla.redhat.com/show_bug.cgi?id=2263643
The root of the error has been found and the patch has been submitted last month, and is planned to be shipped in fedora 44
Which expected to release 2026-04-14
The specific patch is https://src.fedoraproject.org/rpms/grub2/pull-request/198