[Users] host non responsive in a 1 host DC with 1 vm

[Neil]

Hi guys,

I've got a test oVirt setup with only 1 host and 1 test VM but I had a
power failure and the host is now non responsive, despite being able
to log into it via SSH. Manually fencing results in a.... "Manual
fence did not revoke the selected SPM (ovirthosttest) since the master
storage domain was not active or could not use another host for the
fence operation."   When I try put the host into maintenance I get
..."Manual fence did not revoke the selected SPM (ovirthosttest) since
the master storage domain was not active or could not use another host
for the fence operation."  When I try and activate the data domain I
get... "Failed to activate Storage Domain MAIN (Data Center Default)
by admin at internal".  My VM was in an unknown state, but considering
I'd already had a power failure, I logged into my oVirt engine
postgres command line and ran "update vm_dynamic set status = 0 where
vm_guid = (select vm_guid from vm_static where vm_name = 'zimbra');"
to try and force my VM into a down state, but that hasn't helped
either.  I'd imagine that if I had another host to add to the cluster
SPM would hopefully move to the new host and the problem would
disappear, but I don't have one unfortunately.

Any suggestions?

Centos 6.4

vdsm-python-4.10.3-0.31.20.el6.x86_64
vdsm-4.10.3-0.31.20.el6.x86_64
vdsm-cli-4.10.3-0.31.20.el6.noarch
vdsm-xmlrpc-4.10.3-0.31.20.el6.noarch
libvirt-lock-sanlock-0.10.2-18.el6.x86_64
libvirt-client-0.10.2-18.el6.x86_64
libvisual-0.4.0-9.1.el6.x86_64
libvirt-python-0.10.2-18.el6.x86_64
libvirt-0.10.2-18.el6.x86_64

On my engine it's Centos 6.4 with ovirt-engine-3.2.0-1.39.el6.noarch
My engine has an NFS shared RAID array which the hosts the entire data
domain.

I could can the entire install and start again, but I'd prefer to
avoid doing that if possible as I'm still in the process of migrating
email to my zimbra VM.

Any help is appreciated greatly.

Thanks.

Regards.

Neil Wilson.