Latest R9 kernel build fails

Rocky Linux Help & Support
1

When I try to install the latest kernel the boot fails because initramfs-5.14.0-570.21.1.el9_6.x86_64.img isn’t found.
When I try to create initramfs it fails with this error
dracut --kver 5.14.0-570.21.1.el9_6.x86_64
Error on line 0 of configuration file
40FC7BD3B87F0000:error:80000002:system library:BIO_new_file:No such file or directory:crypto/bio/bss_file.c:67:calling fopen(/etc/ssh/wdlive-openssl.cnf, rb)
40FC7BD3B87F0000:error:10000080:BIO routines:BIO_new_file:no such file:crypto/bio/bss_file.c:75:
40FC7BD3B87F0000:error:07000072:configuration file routines:def_load:no such file:crypto/conf/conf_def.c:179:
dracut: ‘/usr/lib/dracut/ossl-files --config’ does not return a path!!

If it matters this is installed in VirtualBox running in a host that is also running R9, but because of this error I’m reluctant to ungrade the host until I can resolve this problem.
Thanks
Pete
Thanks

2

I’m trying to ask the right question here because this appears to be the kernel that is part of the rebase of Rocky to verion 9.6. Thus it is part of a large package update to all of the critical system packages. If you are trying to upgrade to this kernel in isolation to the rest of the upgraded system files then that may be the reason for your problem.

Can you elaborate on the Rocky vm origins and installed state?
Did you do the default install or custom partitioning?
Are you using drive encription?
This command will give us an idea of the disk layout

sudo lsblk -o name,size,type,fstype,uuid,mountpoints

This command will give us an idea of the system origins:

ls -l /lib/modules

I just upgraded my Rocky vm to 9.6 today and had no issues. It is a custom partition layout consisting of one ext4 partition for root and one vfat partition for /boot/efi.
There could be many causes to the error you are seeing so the more background information provided then the more causes we can eliminate.

3

Thanks for your response, I think that I understand your questions.

I use this vm for just this purpose to see if there are any issues.

This is simply a straight nothing fancy install starting with 9.5 and accepting the standard layout and from there I just performed periodic updates

When I updated from kernel 5.14.0-503.40.1.el9_5.x86_64 to 5.14.0-570.18.1.el9_6.x86_64 I first ran into this problem.

I hoped that the next kernel upgrade would solve the problem, but when I updated to 5.14.0-570.21.1.el9_6.x86_64 I got the same error.

I’m currently running 5.14.0-503.40.1.el9_5.x86_64 as the only kernel that has a initramfs file.

I hope that that answers your question.

lsblk -o name,size,type,fstype,uuid,mountpoint
NAME SIZE TYPE FSTYPE UUID MOUNTPOINT
sda 32G disk
├─sda1 1G part xfs 8c9b3caa-7f74-49af-beef-9f28b6996204 /boot
└─sda2 31G part LVM2_member 1JUSwH-uH2u-Q3RH-47XH-2ob7-g1r4-iseJNe
├─rl_skylarvm-root 27.8G lvm xfs 83592f13-77b4-45be-b8c5-b9fde2e4d9c0 /
└─rl_skylarvm-swap 3.2G lvm swap e3526d6b-b6a3-4701-ad2e-324558917921 [SWAP]

ls -l /lib/modules
total 12
drwxr-xr-x 8 root root 4096 Jun 8 17:03 5.14.0-503.40.1.el9_5.x86_64
drwxr-xr-x 7 root root 4096 Jun 8 17:03 5.14.0-570.18.1.el9_6.x86_64
drwxr-xr-x 7 root root 4096 Jun 13 06:43 5.14.0-570.21.1.el9_6.x86_64
drwxr-xr-x 3 root root 19 Jun 8 16:59 5.14.0-570.el9.x86_64

Pete

4

So it sounds like the real problem was upgrading from 9.5 to 9.6 using dnf upgrade? But the original post doesn’t say that, and “kernel build”, sounds like you are building your own kernel.

5

OOps my, I didn’t realize/notice that thisI was an upgrade from 9.5 to 9.6 and yes I was just doing a dnf upgrade, as for building a kernel I’ve not built one since fedora 6 and for the life of me I surely don’t remember how to do that, if I did I would have taken a whack at trying that.
I hope that someone can point me in the right direction to fix this if not then I guess I’ll be facing a reinstall which I’d rather not do if I can avoid it.

6

OK, so it sounds like you did a “normal” dnf upgrade (9.6 was recently released), and maybe something went wrong during that (big) upgrade.

Are you still able to boot with the older kerne? If so perhaps look at the dnf logs from the date / time of the upgrade, look for any errors or warnings. In addition, do ‘dnf check’ as root.