[bug?] Backup from local LVM-Storage

_Lars_

Well-Known Member
Jan 11, 2012
45
2
48
| installed: proxmox-ve_2.0-784729f4-19.iso
| local SAS-Raid (3ware 9690SA) as LVM source
| did 'apt-get-updet && apt-get (dist-)upgrade' right bevor trying to backup

The following happens when I try to backup a (Windows 2003) VM from the local LVM-Storage
(local SAS-RAID) to NFS:

INFO: starting new backup job: vzdump 100 --mode stop --compress 1 --storage backup01 --node node01
INFO: Starting Backup of VM 100 (qemu)
INFO: status = running
INFO: backup mode: stop
INFO: ionice priority: 7
INFO: stopping vm
INFO: creating archive '/mnt/pve/backup01/dump/vzdump-qemu-100-2012_01_11-20_30_38.tgz'
INFO: unable to read '/dev/virt_store_sas/vm-100-disk-1' - No such file or directory
INFO: archive file size: 0KB
INFO: restarting vm
INFO: vm is online again after 31 seconds
INFO: Finished Backup of VM 100 (00:00:31)
INFO: Backup job finished successfully
TASK OK

and if I try it again:

INFO: starting new backup job: vzdump 100 --mode stop --compress 0 --storage backup01 --node node01
INFO: Starting Backup of VM 100 (qemu)
INFO: status = running
INFO: backup mode: stop
INFO: ionice priority: 7
INFO: stopping vm
INFO: creating archive '/mnt/pve/backup01/dump/vzdump-qemu-100-2012_01_11-20_35_43.tar'
INFO: unable to read '/dev/virt_store_sas/vm-100-disk-1' - No such file or directory
INFO: restarting vm
INFO: vm is online again after 29 seconds
ERROR: Backup of VM 100 failed - command ''/usr/lib/qemu-server/vmtar ... failed: exit code 255
INFO: Backup job finished with errors
TASK ERROR: job errors

I highly presume that somewhere between the shutdown of the VM and dumping it's LVM volume
the volume state is set to "not available" and the symlink '/dev/virt_store_sas/vm-100-disk-1'
gets deleted. When I was migrating my VMs to proxmox some time ago I already stumbled over
this problem. I've created all VMs with LVM disks through the web interface and copied their data
using 'dd' from a bash script. While running the script I saw similar errors for missing volume links.
The output of 'lvs' stated them as "not available" and a I had to change the state manually.

Maybe that's the same problem here?

Lars
 
When they removed cLVM from 2.0 I belive they made sure that the volumes are only active on the node that is currently running the vm.

Looks like you may have uncovered a bug.
If the vm is not running then vzdump needs to activate the volume to make the backup and deactivate it after backup.

You could also choose to use snapshot backup instead of stopping the vm.
 
@dietmar:

> Please file a bug: https://bugzilla.proxmox.com

That's what I was looking for! ;)
Done.

@e100: I'm not very familiar with snapshots. Are they "application consistent"?
If the VM is right in the middle of a database operation the snapshot could be
inconsistent, could it not?

 
Last edited:

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!