Ok, I understand. Thank you for your explanation.
Le 20/02/2012 15:05, Andreas Kurz a écrit :
Hello,
On 02/18/2012 03:15 PM, Sébastien ROHAUT wrote:
Hi,
I'm using Pacemaker 1.1.6 and corosync 1.4.2 under Fedora Core 16, using
default rpms from repositories.
I'm trying to get a working Pacemaker cluster and I'm testing it by
creating the "cluster from cratch configuration" from
http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf.
<http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf>
I already worked with pacemaker 1.0.x two years ago, and it worked well
(with the older documentation).
I'm encountering a problem. I added my IPAddr resource and tried
failvover/failback. So I stop my first node, and my resource moves to
node2. After a minute I restart node1, but my resource doesn't come
back, my resource stays on node2.
Without any constraints the cluster is free to move/locate resources
wherever it likes ... base policy is to balance all resources evenly
over available nodes ... you only have one resource -- cluster decides
to leave it as it is.
As explained page 49, my resource should go back to node1. I didn't
apply any stickiness value to my resource or in rsc_default. Did the
behavior changed in this release ?
How to make it working ?
Define a location constraint, preferring node1.
Regards,
Andreas
_______________________________________________
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
Project Home: http://www.clusterlabs.org
Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
Bugs: http://bugs.clusterlabs.org
_______________________________________________
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
Project Home: http://www.clusterlabs.org
Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
Bugs: http://bugs.clusterlabs.org
