Total size of inbound packages is 148 MiB. Need to download 148 MiB.
After this operation, 0 B extra will be used (install 260 MiB, remove 260 MiB).
[ 1/10] kernel-modules-core-0:6.12.4-200.fc41.x86_64 100% | 114.7 KiB/s | 38.4 MiB | 05m43s
Running pre-uninstall scriptlet: kernel-core-0:6.12.4-200.fc41.x86_64
Finished pre-uninstall scriptlet: kernel-core-0:6.12.4-200.fc41.x86_64
Scriptlet output:
Module digimend/13 is not installed for kernel 6.12.4-200.fc41.x86_64 (x86_64x86_64). Skipping…
Module digimend/13 is not built for kernel 6.12.4-200.fc41.x86_64 (x86_64x86_64). Skipping…
Sign command: /lib/modules/6.12.4-200.fc41.x86_64/build/scripts/sign-file
Signing key: /var/lib/dkms/mok.key
Public certificate (MOK): /var/lib/dkms/mok.pub
Cleaning build area… done.
Building module(s)…(bad exit status: 2)
Failed command:
make -j16 KERNELRELEASE=6.12.4-200.fc41.x86_64 KVERSION=6.12.4-200.fc41.x86_64
Error! Bad return status for module build on kernel: 6.12.4-200.fc41.x86_64 (x86_64)
Consult /var/lib/dkms/digimend/13/build/make.log for more information.
Autoinstall on 6.12.4-200.fc41.x86_64 failed for module(s) digimend(10).
Error! One or more modules failed to install during autoinstall.
Refer to previous errors for more information.
/usr/lib/kernel/install.d/40-dkms.install failed with exit status 11.
[RPM] %posttrans(kernel-core-6.12.4-200.fc41.x86_64) scriptlet failed, exit status 11
Complete!
I saw the same root cause for my fedora VM running under Parallels.
The parallels tools uses dkms to build and that failed.
(Parallels have not update their code to support the 6.12 kernels yet).
I was without a initrd.
To fix I booted to the previous kernal and use dracut to make the missing
initrd with:
Note that many 3rd party apps that use dkms to build modules seem unable to match the pace of fedora kernel upgrades. Apps that are provided through rpmfusion are specifically tested for function before releasing upgrades so this is seldom a problem for those apps.
Both these problems (digimend and parallels) report the same type issue.
Even after I removed the offending module, I still get the same problem.
A new kernel, 6.12.5-200 was introduced and the same issue. interstingly, after I boot from a previous kernel I can run dracut --kver 6.12.5-200.fc41.x86_64
with no problems. Why can’t the “dnf update” do the same?
In my case because the dkms process failed and that prevents the initrd being created. I saw the error in the output of the dnf update when it was running scripts i recall.
Had the same issue coming from 6.11.(something) to 6.12.5-200.x86_64. Had a kernel panic on the 6.12.5-200 after reboot, threw a fit about initramfs.
Rebooted into prior kernel, ran the suggested dracut command: dracut --kver 6.12.5-200.fc41.x86_64
Rebooted and the new kernel did throw a “dkms.service failed to start” but booted successfully after a few moments.
After login, I checked “systemctl status dkms.service”, was in failed state and wouldn’t restart…log output from status showed that “anbox-binder” and “anbox-ashmem” were the cause.
Took the following steps to fully resolve:
Reinstalled the relevant to make sure nothing was missed or corrupt sudo dnf reinstall kernel-devel kernel-headers
(to confirm the packages matched the kernel) uname -r rpm -q kernel-devel kernel-headers
get bad dkms modules all cleaned up, again in this case looks like it was Anbox modules sudo dkms remove anbox-binder/1 --all sudo dkms remove anbox-ashmem/1 --all sudo dkms autoinstall --verbose
make sure dkms was all updated sudo dnf update dkms
restart and verify the dkms unit was successfully started sudo systemctl restart dkms.service sudo systemctl status dkms.service
voila… resolved. After reboot, dkms no longer complains when booting into 6.12.5-200.x86_64.