[one-users] migration not working completly

[Jaime Melis]

Hi Ross,

actually in my experience disabling apparmor won't work either. You will
have to modify one of its configuration files in order to make it work.

Add this:
-------8<--------
  /srv/cloud/one/var/** rw,
------->8--------
(If you have a different VMDIR change the above line accordingly).
To the end of /etc/apparmor.d/abstractions/libvirt-qemu
And restart the apparmor service.

Regards,
Jaime




On Mon, Jul 26, 2010 at 5:30 PM, Tino Vazquez <tinova at fdi.ucm.es> wrote:

> Hi Ross,
>
> Are you using Ubuntu per chance? It may be a issue with the apparmor
> service, try disabling it to see if that is the one to blame. In case
> it is, we can provide rules to disable this apparmor behavior.
>
> Regards,
>
> -Tino
>
> --
> Constantino Vázquez Blanco | dsa-research.org/tinova
> Virtualization Technology Engineer / Researcher
> OpenNebula Toolkit | opennebula.org
>
>
>
> On Mon, Jul 26, 2010 at 5:13 PM, Ross Nordeen <rjnordee at mtu.edu> wrote:
> > Tino,
> >
> > I figured out my live migrate problem which turned out to be a bad
> default gw.  As far as the migration and check pointing though I have the
> /srv/cloud/one directory shared out to all nodes via nfs and full
> permissions for oneadmin... I think it is /srv/cloud/one/var/18.  I will
> check the VM_DIR variable in the oned.conf file though and see if it is
> right.  Still if everything else is working it seems like the VM_DIR is
> exported correctly and functioning for the running vm's.
> >
> > -Ross
> >
> > ----- Original Message -----
> > From: "Tino Vazquez" <tinova at fdi.ucm.es>
> > To: "Ross Nordeen" <rjnordee at mtu.edu>
> > Cc: users at lists.opennebula.org
> > Sent: Monday, July 26, 2010 8:41:37 AM GMT -07:00 US/Canada Mountain
> > Subject: Re: [one-users] migration not working completly
> >
> > Hi Ross,
> >
> > There seems to be two issues here:
> >
> > 1) Not live/migrate between cn2 and cn1 --> could it be that the
> > oneadmin user cannot passwordlessly ssh from cn2 to cn1, but it can
> > from cn1 to cn2?
> >
> > 2) The save problem seems to come from the impossibility to save the
> > checkpoint file. This may be due to the fact that /srv/cloud/one
> > directory doesn't exist in the remote nodes, in which case you will
> > need to use the VM_DIR variable in the oned.conf file.
> >
> > Hope it helps,
> >
> > -Tino
> >
> > --
> > Constantino Vázquez Blanco | dsa-research.org/tinova
> > Virtualization Technology Engineer / Researcher
> > OpenNebula Toolkit | opennebula.org
> >
> >
> >
> > On Thu, Jul 22, 2010 at 11:39 PM, Ross Nordeen <rjnordee at mtu.edu> wrote:
> >> I have open nebula deployed with one head node and 2 compute nodes,  I
> have no problems live migrating from cn1 to cn2 but I get failures live/cold
> migrating from cn2 to cn1.  is there any reason I would not able to a) not
> save the state of any of my machines and why live-migration works one way
> but not the other??  Thanks
> >>
> >> -Ross
> >>
> >>
> >> here is my vm.log file after a live-migration, migration, and than
> suspend:
> >>
> >>
> >> Thu Jul 22 11:40:22 2010 [LCM][I]: New VM state is MIGRATE
> >> Thu Jul 22 11:40:22 2010 [VMM][I]: Command execution fail: virsh
> --connect qemu:///system migrate --live one-18 qemu+ssh://cn1/session
> >> Thu Jul 22 11:40:22 2010 [VMM][I]: STDERR follows.
> >> Thu Jul 22 11:40:22 2010 [VMM][I]: Warning: Permanently added
> 'cn2,192.168.1.105' (RSA) to the list of known hosts.
> >> Thu Jul 22 11:40:22 2010 [VMM][I]: error: cannot recv data: Connection
> reset by peer
> >> Thu Jul 22 11:40:22 2010 [VMM][I]: ExitCode: 1
> >> Thu Jul 22 11:40:22 2010 [VMM][E]: Error live-migrating VM, -
> >> Thu Jul 22 11:40:23 2010 [LCM][I]: Fail to life migrate VM. Assuming
> that the VM is still RUNNING (will poll VM).
> >> Thu Jul 22 11:40:23 2010 [VMM][D]: Monitor Information:
> >> .
> >> .
> >> .
> >> .
> >> .
> >> Thu Jul 22 15:09:04 2010 [LCM][I]: New VM state is MIGRATE
> >> Thu Jul 22 15:09:04 2010 [VMM][I]: Command execution fail: virsh
> --connect qemu:///system migrate --live one-18 qemu+ssh://cn1/session
> >> Thu Jul 22 15:09:04 2010 [VMM][I]: STDERR follows.
> >> Thu Jul 22 15:09:04 2010 [VMM][I]: Warning: Permanently added
> 'cn2,192.168.1.105' (RSA) to the list of known hosts.
> >> Thu Jul 22 15:09:04 2010 [VMM][I]: error: cannot recv data: Connection
> reset by peer
> >> Thu Jul 22 15:09:04 2010 [VMM][I]: ExitCode: 1
> >> Thu Jul 22 15:09:04 2010 [VMM][E]: Error live-migrating VM, -
> >> Thu Jul 22 15:09:05 2010 [LCM][I]: Fail to life migrate VM. Assuming
> that the VM is still RUNNING (will poll VM).
> >> Thu Jul 22 15:09:05 2010 [VMM][D]: Monitor Information:
> >> .
> >> .
> >> .
> >> .
> >> .
> >> Thu Jul 22 15:11:25 2010 [LCM][I]: New VM state is SAVE_MIGRATE
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: Command execution fail: 'touch
> /srv/cloud/one/var//18/images/checkpoint;virsh --connect qemu:///system save
> one-18 /srv/cloud/one/var//18/images/checkpoint'
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: STDERR follows.
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: Warning: Permanently added
> 'cn2,192.168.1.105' (RSA) to the list of known hosts.
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: error: Failed to save domain one-18
> to /srv/cloud/one/var//18/images/checkpoint
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: error: operation failed: failed to
> create '/srv/cloud/one/var//18/images/checkpoint'
> >> Thu Jul 22 15:11:25 2010 [VMM][I]: ExitCode: 1
> >> Thu Jul 22 15:11:25 2010 [VMM][E]: Error saving VM state, -
> >> Thu Jul 22 15:11:25 2010 [LCM][I]: Fail to save VM state while
> migrating. Assuming that the VM is still RUNNING (will poll VM).
> >> Thu Jul 22 15:11:26 2010 [VMM][I]: VM running but new state from monitor
> is PAUSED.
> >> Thu Jul 22 15:11:26 2010 [LCM][I]: VM is suspended.
> >> Thu Jul 22 15:11:26 2010 [DiM][I]: New VM state is SUSPENDED
> >> Thu Jul 22 15:13:20 2010 [DiM][I]: New VM state is ACTIVE.
> >> Thu Jul 22 15:13:20 2010 [LCM][I]: Restoring VM
> >> Thu Jul 22 15:13:20 2010 [LCM][I]: New state is BOOT
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: Command execution fail: virsh
> --connect qemu:///system restore /srv/cloud/one/var//18/images/checkpoint
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: STDERR follows.
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: Warning: Permanently added
> 'cn2,192.168.1.105' (RSA) to the list of known hosts.
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: error: Failed to restore domain from
> /srv/cloud/one/var//18/images/checkpoint
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: error: operation failed: cannot read
> domain image
> >> Thu Jul 22 15:13:21 2010 [VMM][I]: ExitCode: 1
> >> Thu Jul 22 15:13:21 2010 [VMM][E]: Error restoring VM, -
> >> Thu Jul 22 15:13:21 2010 [DiM][I]: New VM state is FAILED
> >> Thu Jul 22 15:13:21 2010 [TM][W]: Ignored: LOG - 18 tm_delete.sh:
> Deleting /srv/cloud/one/var//18/images
> >>
> >> Thu Jul 22 15:13:21 2010 [TM][W]: Ignored: LOG - 18 tm_delete.sh:
> Executed "rm -rf /srv/cloud/one/var//18/images".
> >>
> >> Thu Jul 22 15:13:21 2010 [TM][W]: Ignored: TRANSFER SUCCESS 18 -
> >>
> >> _______________________________________________
> >> Users mailing list
> >> Users at lists.opennebula.org
> >> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
> >>
> >
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20100726/d027b381/attachment-0002.htm>