Created attachment 741609 [details]
Cluster Configuration

Description of problem:
When a 3 node cluster hosts 2 groups (APACHE & FTP) with constraint not to run
on the same node, "crm resource move group host" moves the desired group to the
desired host, but also affects the second group by stopping it and then
starting it again.

Version-Release number of selected component (if applicable):
corosync-2.3.6-5.1.x86_64
libcorosync4-2.3.6-5.1.x86_64
libpacemaker3-1.1.16-3.6.x86_64
pacemaker-1.1.16-3.6.x86_64
pacemaker-cli-1.1.16-3.6.x86_64
sbd-1.3.0-2.1.x86_64


How reproducible:
Always

Steps to Reproduce:
1.Create 3 node cluster (hasuse1, hasuse2, hasuse3)
2.Create 2 groups (APACHE and FTP)
3.Set collocation constraint with score -1000 in order the 2 groups avoid each
other
4.Run "crm resource move GROUP node_without_resource 30" 

Actual results:
The group not defined in the "crm resource move" command is being stopped and
again started on the same node.

Expected results:
The group not defined in the "crm resource move" command must not be restarted
at all as it is not running on the same node.

Additional info:
Cluster configuration is attached.