GPT errors

[gospodinov]

Hello. I receive these errors in dmesg after resizing disks of vms.

[4140747.228004] GPT:Primary header thinks Alt. header is not at the end of the disk.
[4140747.228295] GPT:1468006399 != 2097151999
[4140747.228535] GPT:Alternate GPT header not at the end of the disk.
[4140747.228762] GPT:1468006399 != 2097151999
[4140747.228981] GPT: Use GNU Parted to correct GPT errors.
[4140747.229212]  zd1232: p1 p2 p3 p4
[4140747.256411] GPT:Primary header thinks Alt. header is not at the end of the disk.
[4140747.257066] GPT:1468006399 != 2097151999
[4140747.257430] GPT:Alternate GPT header not at the end of the disk.
[4140747.257697] GPT:1468006399 != 2097151999
[4140747.257949] GPT: Use GNU Parted to correct GPT errors.

 

[Kingneutron]

You need to explain more. What exactly did you do? Did you resize the VM disk at the host level, or in-vm?

Are you trying to shrink, or resize-up?

 

[gospodinov]

I resized it at host level. Disk action > + Resize resize-up

 

[leesteken]

I resized it at host level. Disk action > + Resize resize-up

That does not change the partitioning inside the VM or resize the filesystem automatically. GPT keeps a backup of the partition table at the end of the (virtual) drive and will complain that something changed (and the backup is no longer at the end of the (virtual) drive). If you get these messages on the Proxmox host, then that's probably because Proxmox scanned the virtual disks. Simply fix the partitioning (and maybe resize the filesystems to use the additional space) inside the VMs and the messages should go away.

 

[gospodinov]

How to fix it and is it dangerous and one more question ? Should i restart the VM after ?

 

[Kingneutron]

Boot systemrescuecd on the VM and run gparted on in-vm disks, the ones that were resized

https://github.com/nchevsky/systemrescue-zfs/releases

If gparted doesn't fix it you might have to try gdisk to re-write the partition table, or parted

 

[leesteken]

How to fix it

Use the appropriate tools for the operating system inside the VM. You know your VM(s) and what they are running, right? I assumed you had a plan when you decided to grow the virtual disks.

 

[gospodinov]

The error I see is in the proxmox host. Inside the VM I see no errors and the disks are sized correctly.

 

[brainsoft]

I also have this problem on the host, possibly related to a couple rough shutdowns when the ups failed and would love a bit more support to clear it up. The host runs fine for a while but ultimately locks up, and these are the last messages on screen.