[Pacemaker] Resource stickness not honoured

[Саша Александров]

Hi, all!

While building a 2-node cluster on recent CentOS, I ran into a problem:
when configured cman/corosync (just copied from first node) and started
pacemaker - resources started to move to the new node.

However:

# pcs constraint
Location Constraints:
Ordering Constraints:
  promote ms_u01 then start u01_fs
  promote ms_oracle then start oracle_fs
  promote ms_shared then start shared_fs
Colocation Constraints:
  u01_fs with ms_u01 (with-rsc-role:Master)
  oracle_fs with ms_oracle (with-rsc-role:Master)
  shared_fs with ms_shared (with-rsc-role:Master)

# pcs resource defaults
migration-threshold: 3
resource-stickness: 100

All resources were up and running, and where cleaned up before starting
pacemaker on second node. Why could this be?

# cat /etc/issue
Oracle Linux Server release 6.5
Kernel \r on an \m

# rpm -qa | grep pacemaker
pacemaker-1.1.10-14.el6.x86_64


-- 
Best regards,
Alexandr A. Alexandrov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20140604/001c162e/attachment.html>