EXT3 doesn't like the plug being pulled. Not really a CS specific problem.
Destroy the affected system VM's and new ones should spring into life.

- Dave


On 25 April 2013 15:10, Valery Ciareszka <[email protected]> wrote:

> Hi all,
> I'm trying to simulate power failures on hypervisor nodes (reset through
> IPMI)
> After system boots, cloudstack starts system vms, but they keep "starting"
> status in management UI.
> I can see filesystem errors on vnc console of system vms:
> http://thesuki.org/temp/ss/SS-20130425170156.png
> http://thesuki.org/temp/ss/SS-20130425131507.png
> It seems strange that filesystem of VM goes in such corrupted state after
> single unexpected reset.
>
> environment used: CS 4.0.1, centos 6.4, KVM, NFS primary/secondary storage
>
> --
> Regards,
> Valery
>
> http://protocol.by/slayer
>

Reply via email to