VM paused unexpectedly
Hi, today I've noticed that one of my VMs was paused. I took a look at the engine.log and found only this: 2014-09-09 04:00:09,017 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-46) VM XXX 8ad326be-5103-498a-a11d-0ea1eee4bbf4 moved from Up --> Paused The vdsm log provides a better error message: libvirtEventLoop::INFO::2014-09-09 04:00:07,279::vm::4574::vm.Vm::(_onIOError) vmId=`8ad326be-5103-498a-a11d-0ea1eee4bbf4`::abnormal vm stop device ide0-0-1 error libvirtEventLoop::DEBUG::2014-09-09 04:00:07,279::vm::5176::vm.Vm::(_onLibvirtLifecycleEvent) vmId=`8ad326be-5103-498a-a11d-0ea1eee4bbf4`::event Suspended detail 2 opaque None Still, I don't know what exactly caused this. Only *one* VM was affected, so I doubt it was a general error on this host. Does anyone have an explanation for this? ovirt-engine-3.4.3-1.fc19.noarch vdsm-4.14.11.2-0.fc19.x86_64 Regards - Frank
qemu would pause a vm when doing extend on the vm disk and this would result in INFO messages on vm's pause. looks like this is what you are seeing. Dafna On 09/09/2014 02:45 PM, Frank Wall wrote:
Hi,
today I've noticed that one of my VMs was paused. I took a look at the engine.log and found only this:
2014-09-09 04:00:09,017 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-46) VM XXX 8ad326be-5103-498a-a11d-0ea1eee4bbf4 moved from Up --> Paused
The vdsm log provides a better error message:
libvirtEventLoop::INFO::2014-09-09 04:00:07,279::vm::4574::vm.Vm::(_onIOError) vmId=`8ad326be-5103-498a-a11d-0ea1eee4bbf4`::abnormal vm stop device ide0-0-1 error libvirtEventLoop::DEBUG::2014-09-09 04:00:07,279::vm::5176::vm.Vm::(_onLibvirtLifecycleEvent) vmId=`8ad326be-5103-498a-a11d-0ea1eee4bbf4`::event Suspended detail 2 opaque None
Still, I don't know what exactly caused this. Only *one* VM was affected, so I doubt it was a general error on this host. Does anyone have an explanation for this?
ovirt-engine-3.4.3-1.fc19.noarch vdsm-4.14.11.2-0.fc19.x86_64
Regards - Frank _______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
-- Dafna Ron
On Tue, Sep 09, 2014 at 03:09:02PM +0100, Dafna Ron wrote:
qemu would pause a vm when doing extend on the vm disk and this would result in INFO messages on vm's pause.
looks like this is what you are seeing.
For the records, I'm using thin provisioned disks here. Do you mean an internal qemu task which is triggered to extend a thin provisioned disk to the required size? This process shouldn't permanently pause the VM, right? Or do you mean something else? Regards - Frank
On 09/09/2014 03:21 PM, Frank Wall wrote:
On Tue, Sep 09, 2014 at 03:09:02PM +0100, Dafna Ron wrote:
qemu would pause a vm when doing extend on the vm disk and this would result in INFO messages on vm's pause.
looks like this is what you are seeing. For the records, I'm using thin provisioned disks here. Do you mean an internal qemu task which is triggered to extend a thin provisioned disk to the required size? yes
This process shouldn't permanently pause the VM, right? no
Or do you mean something else?
Regards - Frank
-- Dafna Ron
Am 09.09.2014 um 16:38 schrieb Dafna Ron:
On 09/09/2014 03:21 PM, Frank Wall wrote:
On Tue, Sep 09, 2014 at 03:09:02PM +0100, Dafna Ron wrote:
qemu would pause a vm when doing extend on the vm disk and this would result in INFO messages on vm's pause.
looks like this is what you are seeing. For the records, I'm using thin provisioned disks here. Do you mean an internal qemu task which is triggered to extend a thin provisioned disk to the required size? yes
To mitigate (or even get completly rid of it) you can try adding something like this to vdsm.conf in [irs] section volume_utilization_percent = 25 volume_utilization_chunk_mb = 4096 vol_size_sample_interval = 20
This process shouldn't permanently pause the VM, right? no
Or do you mean something else?
Regards - Frank
participants (3)
-
Dafna Ron -
Frank Wall -
InterNetX - Juergen Gotteswinter