Failure on HA migrate

[vince_122]

Hi there !

I am trying to enable a 3-nodes cluster with HA.
I enabled a shared disk storage [ NFS ] on my 3 nodes to store Containers's files.
I have my CT powered on, and "Managed by HA". I tried to unplug the node where the CT is launched, it automatically mounts on another node.

My real problem is the way to enable Live Migration between cluster nodes, as I am having this issue :

root@srv19:/etc/cluster# clusvcadm -r pvevm:100 -m srv18
Trying to relocate pvevm:100 to srv18...Success
pvevm:100 is now running on srv18
root@srv19:/etc/cluster# clusvcadm -M pvevm:100 -m srv19
Trying to migrate pvevm:100 to srv19...Failure

On the GUI Syslog I have this error :

ERROR: migration aborted (duration 00:00:00): Failed to dump container state: Checkpointing failed

I am not able to provide any logs, as this error produces no more output.
Here is my Cluster.conf :

<?xml version="1.0"?>
<cluster config_version="20" name="CLUSTER-PROXMOX">
<cman keyfile="/var/lib/pve-cluster/corosync.authkey"/>
<clusternodes>
<clusternode name="srv17" nodeid="1" votes="1">
<fence>
</fence>
</clusternode>
<clusternode name="srv18" nodeid="2" votes="1">
<fence>
</fence>
</clusternode>
<clusternode name="srv19" nodeid="3" votes="1">
<fence>
</fence>
</clusternode>
</clusternodes>
<fencedevices>
</fencedevices>
<rm>
<pvevm autostart="1" vmid="100"/>
</rm>
</cluster>

Any help to enable this online migration would be appreciated !

Thanks,
Vince.

 

[dietmar]

I am not able to provide any logs, as this error produces no more output.Here is my Cluster.conf :

You should have an entry in the task log on the GUI (with more detailed information)?

 

[vince_122]

You should have an entry in the task log on the GUI (with more detailed information)?

You're right, I added this after :

ERROR: migration aborted (duration 00:00:00): Failed to dump container state: Checkpointing failed

 

[dietmar]

Oh, that seems to be an OpenVZ bug.

 

[tom]

See also https://bugzilla.proxmox.com/show_bug.cgi?id=71