[PVE-User] Backups suddenly slow

Gerald Brandt gbr at majentis.com
Sat Oct 27 10:29:43 CEST 2012


Hi,

I made two changes to my server since the last backup:

1. switched from 2.1 to 2.2
2. converted vmdk storage to qcow2 (for snapshots)

Here's a sample from last weeks backup:

Oct 20 01:01:49 INFO: Starting Backup of VM 101 (qemu)
Oct 20 01:01:49 INFO: status = running
Oct 20 01:01:49 INFO: mode failure - unable to detect lvm volume group
Oct 20 01:01:49 INFO: trying 'suspend' mode instead
Oct 20 01:01:49 INFO: backup mode: suspend
Oct 20 01:01:49 INFO: ionice priority: 7
Oct 20 01:01:49 INFO: suspend vm
Oct 20 01:01:49 INFO: creating archive '/mnt/pve/Backups/dump/vzdump-qemu-101-2012_10_20-01_01_49.tar.lzo'
Oct 20 01:01:49 INFO: adding '/mnt/pve/Backups/dump/vzdump-qemu-101-2012_10_20-01_01_49.tmp/qemu-server.conf' to archive ('qemu-server.conf')
Oct 20 01:01:49 INFO: adding '/mnt/pve/Images/images/101/vm-101-disk-2.vmdk' to archive ('vm-disk-virtio1.vmdk')
Oct 20 01:04:24 INFO: adding '/mnt/pve/Images/images/101/vm-101-disk-1.vmdk' to archive ('vm-disk-virtio0.vmdk')
Oct 20 01:10:11 INFO: Total bytes written: 9381612544 (17.82 MiB/s)
Oct 20 01:10:35 INFO: archive file size: 5.02GB
Oct 20 01:10:36 INFO: resume vm
Oct 20 01:10:37 INFO: vm is online again after 528 seconds
Oct 20 01:10:37 INFO: Finished Backup of VM 101 (00:08:48)

And here is this weeks:

Oct 27 01:24:09 INFO: Starting Backup of VM 101 (qemu)
Oct 27 01:24:09 INFO: status = running
Oct 27 01:24:09 INFO: mode failure - unable to detect lvm volume group
Oct 27 01:24:09 INFO: trying 'suspend' mode instead
Oct 27 01:24:09 INFO: backup mode: suspend
Oct 27 01:24:09 INFO: ionice priority: 7
Oct 27 01:24:09 INFO: suspend vm
Oct 27 01:24:09 INFO: creating archive '/mnt/pve/Backups/dump/vzdump-qemu-101-2012_10_27-01_24_09.tar.lzo'
Oct 27 01:24:09 INFO: adding '/mnt/pve/Backups/dump/vzdump-qemu-101-2012_10_27-01_24_09.tmp/qemu-server.conf' to archive ('qemu-server.conf')
Oct 27 01:24:09 INFO: adding '/mnt/pve/Images/images/101/vm-101-disk-2.qcow2' to archive ('vm-disk-virtio1.qcow2')
Oct 27 01:59:01 INFO: adding '/mnt/pve/Images/images/101/vm-101-disk-1.qcow2' to archive ('vm-disk-virtio0.qcow2')
Oct 27 02:29:40 INFO: Total bytes written: 10161819648 (2.47 MiB/s)
Oct 27 02:29:40 INFO: archive file size: 5.20GB
Oct 27 02:29:41 INFO: resume vm
Oct 27 02:29:41 INFO: vm is online again after 3932 seconds
Oct 27 02:29:41 INFO: Finished Backup of VM 101 (01:05:32)


Is anyone else noticing a change?  Will going back to vmdk help me?

Gerald

ps: are there plans to create full backups using the qcow2 snapshots?



More information about the pve-user mailing list