[one-users] Migrated VMs cannot be shutdown

[Zaina AFOULKI]

 Hi Javier,

 As you mentioned I checked the VMs on the host and it seems
 that all migrated VMs freeze.

 However when I migrate a VM using virt-manager it doesn't freeze
 on the destination host.

 So I guess it is more likely something to do with libvirt/qemu/kvm ?

 Thanks again for any ideas,

 --
 Zaina


 On Sat Mar 26 03:07:19 PDT 2011, Javier Fontan  wrote:

 Can you check that the migrated machines are really running? In some
 configurations VMs freeze after migration so when the shutdown message
 is sent to the VM it does nothing. If that is the case the problem
 could be hypervisor related or something that goes wrong with the
 kernel running in the VM. Also give a minute or so after the migration
 is done, maybe the shutdown command was sent before the VM could catch
 it.


 On Thu, 24 Mar 2011 12:15:02 +0100, Zaina AFOULKI wrote:
> Hi,
>
> I'm experiencing a problem when migrating VMs with OpenNebula.
> I'm doing the migration with the "onevm migrate <vm_id> <host_id>"
> command, and it works just fine.
>
> However, once a VM has been migrated, It cannot be shutdown anymore: 
> its
> state becomes "shutdown", then goes back to "running".
>
> Any VM that hasn't been migrated can be shutdown without errors.
>
> Here are the VM logs:
> Thu Mar 24 11:20:22 2011 [DiM][I]: New VM state is ACTIVE.
> Thu Mar 24 11:20:22 2011 [LCM][I]: New VM state is PROLOG.
> Thu Mar 24 11:20:22 2011 [VM][I]: Virtual Machine has no context
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh:
> synema-desktop:/home/oneadmin/vmu.img
> node1:/home/oneadmin/one/var//2788/images/disk.0
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: DST:
> /home/oneadmin/one/var//2788/images/disk.0
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Creating directory
> /home/oneadmin/one/var//2788/images
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Executed "mkdir -p
> /home/oneadmin/one/var//2788/images".
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Executed "chmod a+w
> /home/oneadmin/one/var//2788/images".
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Cloning
> /home/oneadmin/vmu.img
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Executed "cp -r
> /home/oneadmin/vmu.img /home/oneadmin/one/var//2788/images/disk.0".
> Thu Mar 24 11:20:27 2011 [TM][I]: tm_clone.sh: Executed "chmod a+w
> /home/oneadmin/one/var//2788/images/disk.0".
> Thu Mar 24 11:20:42 2011 [LCM][I]: New VM state is BOOT
> Thu Mar 24 11:20:42 2011 [VMM][I]: Generating deployment file:
> /home/oneadmin/one/var/2788/deployment.0
> Thu Mar 24 11:20:44 2011 [LCM][I]: New VM state is RUNNING
> Thu Mar 24 11:20:45 2011 [HKM][I]: Hook ebtables-start successfully
> executed.
> Thu Mar 24 11:21:50 2011 [LCM][I]: New VM state is SAVE_MIGRATE
> Thu Mar 24 11:22:38 2011 [LCM][I]: New VM state is PROLOG_MIGRATE
> Thu Mar 24 11:22:38 2011 [TM][I]: tm_mv.sh: Will not move, source and
> destination are equal
> Thu Mar 24 11:22:38 2011 [LCM][I]: New VM state is BOOT
> Thu Mar 24 11:22:39 2011 [LCM][I]: New VM state is RUNNING
> Thu Mar 24 11:22:46 2011 [HKM][I]: Hook ebtables-start successfully
> executed.
> Thu Mar 24 11:23:22 2011 [LCM][I]: New VM state is SHUTDOWN
> Thu Mar 24 11:23:54 2011 [VMM][I]: Command execution fail:
> /var/tmp/one/vmm/kvm/shutdown one-2788
> Thu Mar 24 11:23:54 2011 [VMM][I]: STDERR follows.
> Thu Mar 24 11:23:54 2011 [VMM][I]: Timeout reached
> Thu Mar 24 11:23:54 2011 [VMM][I]: ExitCode: 1
> Thu Mar 24 11:23:54 2011 [VMM][E]: Error shuting down VM, -
> Thu Mar 24 11:23:54 2011 [LCM][I]: Fail to shutdown VM. Assuming that 
> the
> VM is still RUNNING (will poll VM).
> Thu Mar 24 11:23:55 2011 [VMM][D]: Monitor Information:
>         CPU   : -1
>         Memory: 262144
>         Net_TX: -1
>         Net_RX: -1
>
> Thanks for any ideas.
>
> --
> Zaina