Powercut changes something on one VM!

[SpongeRob]

Okay, so a little background information:

PVE 6.4-6
Running up to 8 VMs
HP ML330 G6 - X5650
110GB RAM

I now nearly have all my 'everyday' VM's up and running again, however my Storj VM refuses to run.. complaining "attempt to read or write outside of disk 'hd0'"

When I go into Console, and watch this VM start up, all I see is:

Booting ‘Debian GNU/Linux’

error: attempt to read or write outside of disk ‘hd0’.
error: attempt to read or write outside of disk ‘hd0’.
Loading Linux 4.19.0-21-amd64 …
error: attempt to read or write outside of disk ‘hd0’.
Loading initial ramdisk …
error: attempt to read or write outside of disk ‘hd0’.

Press any key to continue…

Failed to boot both default and fall back entries.

Press any key to continue…

I've posted a thread on the Storj forum, and they have kindly pointed out, that maybe I'd be better off posting this here on the ProxMox forum.

Hopefully you guys can (urgently) help me out please, so I can get my Storj node back online and working again please!

Thanks in advance.

 

[Maximiliano]

Hello

Could you please post your VM config (at `/etc/pve/qemu-server/<VMID>.conf`)? Do you have any backups of that VM at hand?

 

[SpongeRob]

Hi Maximiliano, thanks for your reply.



EDIT: Currently booting from .ISO but normally it's set straight too SCSI0

Yes, I have backups (somewhere lol)

 

[Maximiliano]

If you are booting into the iso rather than the disk, just remove the ISO from the VM hardware config.

Otherwise, from your description, it might be the case the data was corrupted during the powercut. I would advise to do a backup of the current state just to be sure and then restore the VM from the previous backup.

Do you have disks with power loss protection?

 

[fabian]

this is an issue with
- grub
- legacy bios
- large disks

see multiple other forum posts and google results. the power cut likely just exposed the problem. the solution is to put /boot on a smaller disk or switch to UEFI.

 

[SpongeRob]

If you are booting into the iso rather than the disk, just remove the ISO from the VM hardware config.

Otherwise, from your description, it might be the case the data was corrupted during the powercut. I would advise to do a backup of the current state just to be sure and then restore the VM from the previous backup.

Do you have disks with power loss protection?

If I let it boot as normal (scsi0) if fails to boot with the messages about reading and writing outside of hd0

The last backup I took was a while ago, not recent. Yes I could try doing this but alas I think either way it wouldn't help me.

In terms of UPS's I have three, however one battery has gone bad, one is for my router, whilst the larger one is for my server. However, these are not automated, they run on demand. If my power goes off unexpectedly it's a dilemma whether to power down the system of not. I usually leave it on for 30 minutes, then shut it down manually if it hasn't come back online in that time period.

 

[SpongeRob]

this is an issue with
- grub
- legacy bios
- large disks

see multiple other forum posts and google results. the power cut likely just exposed the problem. the solution is to put /boot on a smaller disk or switch to UEFI.

Yes, probably all correct Fabian, is there anything I can do to restore or save my current installation?

 

[fabian]

like I said:
- boot using a livecd iso, add a second small disk, move /boot there, reinstall grub on small disk
- or switch to UEFI (also requires a livecd boot and reinstall of bootloader)