Fuel for OpenStack

  1. Bug #1394188
  2. Comment #35

Comment 35 for bug 1394188

Revision history for this message
Aleksandr Didenko (adidenko) wrote : Re: Deployment of additional/replace controller fails

Some further details on this issue:

1) When I remove node-4 from the environment, it still stays in corosync cluster and it's marked as OFFLINE.
2) Then, when I add it back, it gets new name (in my case node-6) but it gets exactly the same IP address as before.
3) When I deploy changes with newly added node-6, I get timeouts on this step in puppet on existing secondary controller:

2015-04-20 13:49:09 +0000 Exec[stop-pacemaker](provider=posix) (debug): Executing 'service pacemaker stop || true'
2015-04-20 13:49:09 +0000 Puppet (debug): Executing 'service pacemaker stop || true'

Here's what I see in pacemaker log:
http://paste.openstack.org/show/204883/