Tell me more ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.
up vote 2 down vote favorite

I've got a cluster with 84 Openstack Essex compute nodes, one of which decided to pine for the fjords about 2 hours ago, with 8 running instances on it.

The instances are all running off qcow2 disks on an NFS share, so the disk images are still available.

nova live-migration only seems capable of migrating a running instance from one functioning node to another.

What I want is to 'migrate' the dead instances from the dead compute node to any of the live compute nodes. The fact that the instances will be rebooted (and will have to fsck on boot) doesn't really matter - it's better than losing them completely.

Before I start acting on thoughts about messing with virsh and hacking nova's mysql database directly, does anyone know of any tool or documented procedure for this kind of 'dead-migration'?

share|improve this question
Craig this is very interesting topic, I'm looking forward to the answer. I suggest you to forward it to Openstack dev mailing list. – janisg Aug 8 '12 at 19:46

1 Answer

up vote 1 down vote

I never done this. I hope this reference help you to recover the instances.

http://docs.openstack.org/essex/openstack-compute/admin/content/nova-disaster-recovery-process.html

share|improve this answer
Thanks, but i've seen that. It's not really related to what I needed to do. One of my colleagues ended up editing the nova mysql instances table directly and some manual stuffing around with virsh define, virsh nwdefine, and virst start on the target node. The sort of thing that really should be automated. – Craig Sanders Aug 5 '12 at 9:57

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.