snapshot LZO backups Bug..?

nate

Member
Mar 15, 2011
63
0
6
Successful backup of vm to NFS on a stopped vm using snapshot LZO in GUI:

INFO: starting new backup job: vzdump 101 --remove 0 --mode snapshot --compress lzo --storage rn3100prd --node cloud1
INFO: Starting Backup of VM 101 (qemu)
INFO: status = stopped
INFO: backup mode: stop
INFO: ionice priority: 7
INFO: creating archive '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_31_06.tar.lzo'
INFO: adding '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_31_06.tmp/qemu-server.conf' to archive ('qemu-server.conf')
INFO: adding '/var/lib/vz/images/101/vm-101-disk-1.raw' to archive ('vm-disk-ide0.raw')
INFO: Total bytes written: 53687093760 (160.00 MiB/s)
INFO: archive file size: 1.06GB
INFO: Finished Backup of VM 101 (00:05:26)
INFO: Backup job finished successfully
TASK OK

When backing up a running/started vm, it almost finishes and just hangs for like an hour or more.

INFO: starting new backup job: vzdump 101 --remove 0 --mode snapshot --compress lzo --storage rn3100prd --node cloud1
INFO: Starting Backup of VM 101 (qemu)
INFO: status = running
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: Logical volume "vzsnap-cloud1-0" created
INFO: creating archive '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_47_17.tar.lzo'
INFO: adding '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_47_17.tmp/qemu-server.conf' to archive ('qemu-server.conf')
INFO: adding '/mnt/vzsnap0/images/101/vm-101-disk-1.raw' to archive ('vm-disk-ide0.raw')
INFO: Total bytes written: 53687093760 (164.63 MiB/s)

hangs here forever....

Any ideas?
 
Job finally finished:

INFO: starting new backup job: vzdump 101 --remove 0 --mode snapshot --compress lzo --storage rn3100prd --node cloud1
INFO: Starting Backup of VM 101 (qemu)
INFO: status = running
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: Logical volume "vzsnap-cloud1-0" created
INFO: creating archive '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_47_17.tar.lzo'
INFO: adding '/mnt/pve/rn3100prd/dump/vzdump-qemu-101-2012_07_18-10_47_17.tmp/qemu-server.conf' to archive ('qemu-server.conf')
INFO: adding '/mnt/vzsnap0/images/101/vm-101-disk-1.raw' to archive ('vm-disk-ide0.raw')
INFO: Total bytes written: 53687093760 (164.63 MiB/s)
INFO: archive file size: 1.06GB
INFO: Finished Backup of VM 101 (01:29:03)
INFO: Backup job finished successfully
TASK OK

The archive is only 1.06 gb.

Why does it take 5 min to backup when the VM is stopped vs an hour and a half with the VM is running and the backup is snapshot? Seems like this is broken or a bug? I'm using latest version of pve via aptitude update && aptitude full-upgrade.
 
First, the job runs with ionice 7, so that highly depend on other things running on that node which delay the backup task.
Second,LVM snapshots are not really fast.
 
  • Like
Reactions: mjw
ok. that is good info.

Why does it hang between "total bytes written" in the log output and "archive file size" (see below)? Isn't the job already done at that point, I mean the hard work, compressing and moving...? this is only task I'm running on the server so thre are no competing resources and no other vms running.

It just doesnt seem like it should take over an hour for a 2 gb archive file on a gb switch with gb nics, and comparing; the backup takes five min if the vm is stopped.

Any Ideas on how to speed this up?







INFO: adding '/mnt/vzsnap0/images/101/vm-101-disk-1.raw' to archive ('vm-disk-ide0.raw')
INFO: Total bytes written: 53687093760 (164.63 MiB/s)

Hangs here for long time....

INFO: archive file size: 1.06GB
INFO: Finished Backup of VM 101 (01:29:03)
INFO: Backup job finished successfully
 
Why does it hang between "total bytes written" in the log output and "archive file size" (see below)?

The backup job is already done, and the vzdump code only do a file rename there. So the only explanation is that the storage server hangs?
 
Storage server seems to be ok. What logs should I look at? I can use the storage appliance for other activities, move files around etc.. Using readynas 3100.

Every time I go through the scenario same behavior.
 
Can you test it vith cifs share instead of nsf? For me nfs is buggy. Backup jobs takes forever. But with cfs mount to the same storage all backup jobs finish with success and very fast. I backup about 30 VM vith lvm snaphot to nexentastor storage via cifs mount.
 
thank you rahman! What a world of difference that made using cif. snapshot in nfs over an hour, now it takes 5 min using cif. NFS is defiantly buggy.

Thank you both for you comments and suggestions. I do very much appreciate the help.
 
I used debian as well as nexentastor as nfs server without any success. Yesterday I upgraded my nexentastor server and tested nfs again to see if it is nexentastor. 300 GB raw VM (16 GB archived) snapshot backup took 16 hours to finish. So something really wrong with proxmox nfs. The problem is it creates the archive on nfs server in about 30 mins but then it gets really slow to finish the job. With cifs, it is problem free.
 

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!