CT running nfs server always fails online migration

[mir]

Hi,

I seem to have a problem with a CT running nfs server, online migration always fails!
Is it me who has forgotten some special config option or are this expected behavior since nfs server and CT seems a bit problematic - no support for nfsv4?

 

[dietmar]

Maybe related to this bug: http://bugzilla.openvz.org/show_bug.cgi?id=2242

 

[mir]

Maybe related to this bug: http://bugzilla.openvz.org/show_bug.cgi?id=2242

I don't think this bug is related. My CT is installaled locally it is the CT which runs a NFS server inside. It does not make any difference whether an active mount exists or not at the time of migration,

 

[dietmar]

What error do you get exactly?

 

[mir]

This is the output:

Aug 14 01:37:04 starting migration of CT 102 to node 'esx1' (192.168.2.8)
Aug 14 01:37:04 container is running - using online migration
Aug 14 01:37:05 starting rsync phase 1
Aug 14 01:37:05 # /usr/bin/rsync -aH --delete --numeric-ids --sparse /var/lib/vz/private/102 root@192.168.2.8:/var/lib/vz/private
Aug 14 01:41:09 start live migration - suspending container
Aug 14 01:41:09 dump container state
Aug 14 01:41:09 # vzctl --skiplock chkpnt 102 --dump --dumpfile /var/lib/vz/dump/dump.102
Aug 14 01:41:09 Setting up checkpoint...
Aug 14 01:41:09 join context..
Aug 14 01:41:09 dump...
Aug 14 01:41:09 Can not dump container: No such device
Aug 14 01:41:09 ERROR: Failed to dump container state: Checkpointing failed
Aug 14 01:41:09 aborting phase 1 - cleanup resources
Aug 14 01:41:09 removing copied files on target node
Aug 14 01:41:13 start final cleanup
Aug 14 01:41:13 ERROR: migration aborted (duration 00:04:09): Failed to dump container state: Checkpointing failed
TASK ERROR: migration aborted