[SOLVED] VM Backup issues with proxmox 6.0

[sebmel]

Hi,

i am on an enterprise subscription of proxmox 6.0 and have the latest packages installed. Since a couple of days some of the backups will not be performed correctly:

101: 2019-09-24 01:43:07 INFO: Starting Backup of VM 101 (qemu)
101: 2019-09-24 01:43:07 INFO: status = running
101: 2019-09-24 01:43:08 INFO: update VM 101: -lock backup
101: 2019-09-24 01:43:08 INFO: VM Name: XXX
101: 2019-09-24 01:43:08 INFO: include disk 'ide0' 'local:101/vm-101-disk-1.qcow2' 50G
101: 2019-09-24 01:43:09 INFO: backup mode: snapshot
101: 2019-09-24 01:43:09 INFO: ionice priority: 7
101: 2019-09-24 01:43:09 INFO: creating archive '/opt/backup//dump/vzdump-qemu-101-2019_09_24-01_43_07.vma.lzo'
101: 2019-09-24 01:43:12 ERROR: got timeout
101: 2019-09-24 01:43:12 INFO: aborting backup job
101: 2019-09-24 01:53:12 ERROR: VM 101 qmp command 'backup-cancel' failed - got timeout
101: 2019-09-24 01:53:14 ERROR: Backup of VM 101 failed - got timeout

104: 2019-09-24 01:53:14 INFO: Starting Backup of VM 104 (qemu)
104: 2019-09-24 01:53:14 INFO: status = running
104: 2019-09-24 01:53:15 INFO: update VM 104: -lock backup
104: 2019-09-24 01:53:15 INFO: VM Name: XXX
104: 2019-09-24 01:53:15 INFO: include disk 'virtio0' 'vm01-sata:vm-104-disk-1' 50G
104: 2019-09-24 01:53:15 INFO: include disk 'virtio1' 'vm01-ssd:vm-104-disk-1' 210G
104: 2019-09-24 01:53:16 INFO: backup mode: snapshot
104: 2019-09-24 01:53:16 INFO: ionice priority: 7
104: 2019-09-24 01:53:16 INFO: creating archive '/opt/backup//dump/vzdump-qemu-104-2019_09_24-01_53_14.vma.lzo'
104: 2019-09-24 01:53:19 ERROR: got timeout
104: 2019-09-24 01:53:19 INFO: aborting backup job
104: 2019-09-24 02:03:19 ERROR: VM 104 qmp command 'backup-cancel' failed - got timeout
104: 2019-09-24 02:03:20 ERROR: Backup of VM 104 failed - got timeout

This does not happen for all VMs, but some of them. While trying to reboot the VMs, i noticed that they also cannot be gracefully shutdown. This will result in:

TASK ERROR: VM quit/powerdown failed

Any idea what might cause this?

The Setup of the VMs is quite standard (some on lvm, some as qcow as seen above) and there is only 1 VM instance running in this case.

Regards,
Sebastian

 

[aaron]

Can you connect to the VNC console of the VM? If that also fails with a timeout it sounds like it could be this problem: https://forum.proxmox.com/threads/certain-vms-from-a-cluster-cannot-be-backed-up-and-managed.57016/

Can you power down the VM nicely after a fresh start?

 

[sebmel]

No VNC does not work either and the shutdown works after a hard reset, the issue description matches with the issue you mentioned, so i will try to gather some debug logging for you