CentOS - Boot problem after recent kernel update (DO NOT INSTALL GRUB, SHIM AND KERNEL UPDATES FOR NOW)

today after updating my Virtualmin as usual, after rebooting the google cloud instance it doesn’t comes back
the filesystem was corrupted
it test this several times now and still the same issue

Centos 7 and 8

Google cloud instance

That seems unlikely (maybe impossible?) to be caused by one of our packages. Which specific package(s) did you update?

Also, you really need to include your distro and version. There is no way to make any sort of sense out of it without that info.

i just run yum update on a fresh centos 7 and centos 8 had the same issue

whenever i hit reboot, everything will be gone
even rebooting the instance from the google cloud panel

I got the same issue, the system can’t boot after automatic update and reboot.

So, the kernel package isn’t ours. That’s coming from CentOS and we have no control over it. Can you try upgrading just the kernel without changing any of our packages (just use yum or dnf directly to satisfy yourself that we aren’t breaking it).

Okay, understood

I did the kernel update this morning. It took a long time to reboot (maybe five minutes compared to the usual 20 seconds or so), but it did eventually reboot. I just figured it was fscking. CentOS 7.8.

I usually just let KernelCare handle the kernel unless there’s a significant security problem with the current one. But because there were other updates this morning, I figured I’d do them all.

Is there any “official” KernelCare integration? It seems to work just fine without one, but automatically deferring all non-security kernel updates for a few days to give Igor a chance to update his kernel would be convenient.

Richard

Just set yum to ignore kernel updates, if you’re handling them some other way.

I probably should. Igor is pretty quick about patching.

Richard

This looks like the patch for the BootHole bug.

@ramin Wouldn’t that be the GRUB patch?

Richard

Yes, or at least GRUB is being blamed. If it’s what I saw in the news the other day this hole allows exploits more wicked than usual.

1 Like

Same here my server is down after upgrade

i did excluded the kernel from the update, the same issue happens, once i install virtualmin and reboot it will never comes back
and if i don’t install Virtualmin and update everything to the latest version including the kernel everything will works perfectly until you install virtualmin

Well, I don’t see how it can be related. I could reproduce it but I don’t think it’s Virtualmin related.

Anyone seeing this problem, force shutdown your instance and re-start over again. The “issue” is getting auto-resolved.

it doesn’t, it’s been 20 hours trying everything, everything is fine until you install virtualmin and reboot

If you update Virtualmin via command line, does it change anything? Is Webmin updated with Virtualmin along side or it’s just single Virtualmin package?

I guess it’s related to BootHole: https://www.zdnet.com/google-amp/article/red-hat-enterprise-linux-runs-into-boothole-patch-trouble/

1 Like