Discussions
Categories
- 17.9K All Categories
- 3.4K Industry Applications
- 3.4K Intelligent Advisor
- 75 Insurance
- 537.7K On-Premises Infrastructure
- 138.7K Analytics Software
- 38.6K Application Development Software
- 6.1K Cloud Platform
- 109.6K Database Software
- 17.6K Enterprise Manager
- 8.8K Hardware
- 71.3K Infrastructure Software
- 105.4K Integration
- 41.6K Security Software
Can't boot UEK after update: initramfs unpacking, invalid magic - UEK compression libs chaged?
After the latest dnf update, which included new dracut (049-202.git20220511.0.1.el8_6) and UEK versions (5.4.17-2136.309.4.el8uek), all UEK kernels will not boot, resulting in a kernel panic caused by the inability to unpack initramfs:
Initfamfs unpacking failed: invalid magic at start of compressed archive
Rebuilding initramfs with different dracut compression options doesn't seem to work (tried gzip, bzip2, xz, lz4, lzo, zstd)
Linux kernels seem to boot fine, so I'd think dracut is not at issue, however UEK kernels that used to boot no longer do, which does seem to point back to dracut although all it does is call an external compression command.
I prefer not to get mired in dnf downgrade problems (previous version was I think 049-191.git20210920.0.2.el8). Any ideas how to get UEK kernel able to unpack the initramfs image files?
(PS - Sorry, mods - not sure if this will be duplicate post; previous one disappeared when trying to edit it so I re-posted)
Best Answer
-
Seems to be fixed as of at least Grub version 2.02-123.0.10.el8_6.8 (grub2-install --version reports 2.03)
Lack of authoritative help here makes one reconsider O/S choice
Answers
-
I have a same problem.
I tried rollback to last backuped system image (release version 8.5).
before 'yum update', the system worked fine.
but after 'yum update', "Initfamfs unpacking failed" problem happened.
(when I boot using RHEL Kernel, It works fine.)
(update)
----------------------------------------------------------------------------
I got a fix this problem.
Solution : downgrade grub2 (2.02-123.0.4 -> 2.02-123.0.3)
dnf downgrade 'grub2*'
-
Had this issue today after a dnf update. Switched to the RedHat kernel as default which worked (https://community.oracle.com/tech/apps-infra/discussion/4467791/oracle-linux-how-to-change-default-kernel)
Will probably stay with the RedHat kernel for this client.
-
Just had the same bug. 2 months still no fix?
-
Still works for 8.6
-
Thanks Tualua for the update.
Upgrading multiple systems remotely next month, was hoping this issue was resolved.
Have to decide whether or not to go down the grub downgrade path (& I suppose add exclude=grub* to /etc/dnf/dnf.conf to prevent future dnf updates from breaking boot & remembering, in future, once this is fixed, to remove the exclusion) or switch the servers to the RedHat kernel.
-
Does Oracle pay attention to posts here? Is there no way grub will be fixed so we can boot UEK or do we have to give up on it?
-
Seems to be fixed as of at least Grub version 2.02-123.0.10.el8_6.8 (grub2-install --version reports 2.03)
Lack of authoritative help here makes one reconsider O/S choice
-
How does 1 upgrade to Grub version 2.02-123.0 when the error happens on boot up?
-
Boot into a non-UEK kernel. Your machines should have stock RedHat kernels as well as UEK ones.