Description of problem:
When running the 'cibadmin -u --force' command on a node which is not the domain controller, it will get stuck for 30 seconds and then returns "Call cib_upgrade failed (-62): Timer expired", while in fact the scheme version is upgraded correctly. This does not happen on the DC node.


Version-Release number of selected component (if applicable):
pacemaker-libs-1.1.12-1.el7.x86_64
pacemaker-cli-1.1.12-1.el7.x86_64
pacemaker-1.1.12-1.el7.x86_64
pacemaker-cluster-libs-1.1.12-1.el7.x86_64


How reproducible:
Always


Steps to Reproduce:
1. Check that the CIB version can be upgraded: pcs cluster cib | egrep -o '(validate\S+)'
2. Find out which node is the current DC: pcs status | grep DC
3. Perform the upgrade on node other than the DC:  cibadmin --upgrade --force


Actual results:

[root@virt-042 ~]# pcs cluster cib | egrep -o '(validate\S+)'
validate-with="pacemaker-1.2"
[root@virt-042 ~]# pcs status | grep DC
Current DC: virt-043 (3) - partition with quorum

[root@virt-042 ~]# cibadmin --upgrade --force
Call cib_upgrade failed (-62): Timer expired
[root@virt-042 ~]# pcs cluster cib | egrep -o '(validate\S+)'
validate-with="pacemaker-2.1"


Expected results:

[root@virt-042 ~]# cibadmin --upgrade --force
[root@virt-042 ~]# pcs cluster cib | egrep -o '(validate\S+)'
validate-with="pacemaker-2.1"