[ovirt-users] Re: How to discover why a VM is getting suspended without recovery possibility?

Hi Gianluca. On 22 Sep 2020, at 04:24, Gianluca Cecchi <gianluca.cecchi@gmail.com<mailto:gianluca.cecchi@gmail.com>> wrote: On Tue, Sep 22, 2020 at 9:12 AM Vinícius Ferrão via Users <users@ovirt.org<mailto:users@ovirt.org>> wrote: Hi Strahil, yes I can’t find anything recently either. You digged way further then me, I found some regressions on the kernel but I don’t know if it’s related or not: https://patchwork.kernel.org/patch/5526561/ https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1045027 Regarding the OS, nothing new was installed, just regular Windows Updates. And finally about nested virtualisation, it’s disabled on hypervisor. In your original post you wrote about the VM going suspended. So I think there could be something useful in engine.log on the engine and/or vdsm.log on the hypervisor. Could you check those? Yes I goes to suspend. I think this is just the engine don’t knowing what really happened and guessing it was suspended. On engine.log I only have this two lines: # grep "2020-09-22 01:51" /var/log/ovirt-engine/engine.log 2020-09-22 01:51:52,604-03 INFO [org.ovirt.engine.core.vdsbroker.monitoring.VmAnalyzer] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-57) [] VM '351db98a-5f74-439f-99a4-31f611b2d250'(cerulean) moved from 'Up' --> 'Paused' 2020-09-22 01:51:52,699-03 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-57) [] EVENT_ID: VM_PAUSED(1,025), VM cerulean has been paused. Note that I’ve “grepped” with time. There’s only this two lines when it crashed like 2h30m ago. On vdsm.log on the near time with the name of the VM I only found an huge JSON, with the characteristics of the VM. If there something that I should check specifically? Tried some combinations of “grep” but nothing really useful. Also, do you see anything in event viewer of the WIndows VM and/or in Freenas logs? FreeNAS is just cool, nothing wrong there. No errors on dmesg, nor resource starvation on ZFS. No overload on the disks, nothing… the storage is running easy. About Windows Event Viewer it’s my Achilles’ heel; nothing relevant either as far as I’m concerned. There’s of course some mentions of improperly shutdown due to the crash, but nothing else. I’m looking further here, will report back if I found something useful. Thanks, Gianluca