[one-users] vm migrate stuck in "save" state

[Ruben S. Montero]

Hi Jack,

You may have problems if you are using the last version of Ubuntu
(9.10) . The default configuration of apparmor prevents the libvirt
daemon to write checkpoint files under $ONE_LOCATION...

If you are using Karmic follow slide 47 of [1]. You should apply that
to all the cluster nodes. If this is not the problem you may find a
hint of what is going on in /var/log/kern.log

Cheers

Ruben


REFS
[1] http://www.slideshare.net/rsmontero/building-clouds-one-14

On Wed, Feb 10, 2010 at 6:10 PM, Jack Jill <jacknjill111 at gmail.com> wrote:
> I was attempting to migrate a running vm to a different host via "onevm
> migrate" command. The command kicked off the "Save" phase, but it seems
> stuck there.
>
> oneadmin at uecA:~/scratch$ onevm list
>   ID     USER     NAME STAT CPU     MEM        HOSTNAME        TIME
>   35 oneadmin ubuntuWo runn   0 1048576   192.168.0.100 00 17:56:54
>   37 oneadmin ubuntuWo save   0  262144            uecB 00 01:12:16
>
> oneadmin at uecA:~/scratch$ onevm show 37
> VIRTUAL MACHINE 37
> INFORMATION
> ID             : 37
> NAME           : ubuntuWorkerB
> STATE          : ACTIVE
> LCM_STATE      : SAVE_MIGRATE
> START TIME     : 02/10 09:45:13
> END TIME       : -
> DEPLOY ID:     : one-37
>
> VIRTUAL MACHINE
> TEMPLATE
> CPU=0.25
> DISK=[
>   BUS=virtio,
>   READONLY=no,
>   SOURCE=/opt/nebula/images/ubuntuWorker5.img,
>   TARGET=vda ]
> FEATURES=[
>   ACPI=yes,
>   PAE=yes ]
> MEMORY=256
> NAME=ubuntuWorkerB
> NIC=[
>   BRIDGE=br0,
>   IP=192.168.5.9,
>   MAC=00:03:c0:a8:05:09,
>   MODEL=virtio,
>   NETWORK=Small network,
>   VNID=3 ]
> NIC=[
>   BRIDGE=br0,
>   IP=10.0.1.2,
>   MAC=00:03:0a:00:01:02,
>   MODEL=virtio,
>   NETWORK=RoshPrivateLAN,
>   VNID=4 ]
> RAW=[
>   DATA=,
>   TYPE=kvm ]
> VMID=37
>
>
> 37/vm.log:
> -------------
> Wed Feb 10 09:51:32 2010 [LCM][I]: New VM state is RUNNING
> Wed Feb 10 09:51:47 2010 [VMM][D]: Monitor Information:
>         CPU   : -1
>         Memory: 262144
>         Net_TX: -1
>         Net_RX: -1
> Wed Feb 10 10:02:41 2010 [VMM][D]: Monitor Information:
>         CPU   : -1
>         Memory: 262144
>         Net_TX: -1
>         Net_RX: -1
> Wed Feb 10 10:02:48 2010 [LCM][I]: New VM state is SAVE_MIGRATE
>
> oned.log :
> ------------
> Wed Feb 10 10:02:41 2010 [VMM][D]: Message received: POLL SUCCESS 37
> STATE=a USEDMEMORY=262144
>
> Wed Feb 10 10:02:41 2010 [VMM][D]: Message received: POLL SUCCESS 35
> STATE=a USEDMEMORY=1048576
>
> Wed Feb 10 10:02:48 2010 [ReM][D]: HostPoolInfo method invoked
> Wed Feb 10 10:02:48 2010 [ReM][D]: VirtualMachineMigrate invoked
> Wed Feb 10 10:02:48 2010 [DiM][D]: Migrating VM 37
> Wed Feb 10 10:03:42 2010 [InM][I]: Monitoring host uecB (0)
> Wed Feb 10 10:03:43 2010 [InM][I]: Monitoring host 192.168.0.100 (3)
>
> The VM (id: 37) that I am trying to migrate was originally on 192.168.0.100
> host. "onevm list" indicates it's on uecB host. But I don't see a directory
> "37" (vm directory) on uecB.
>
> Any thoughts on why the VM is stuck in the "Save" state.
> --
> Thanks,
> Rosh
> http://blogs.plexibus.com
>
>
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>
>



-- 
Dr. Ruben Santiago Montero
Associate Professor, Complutense University of Madrid

URL:    http://dsa-research.org/doku.php?id=people:ruben
Weblog: http://blog.dsa-research.org/?author=7