[system tests 4.1] Pacemaker status failed with 'unknown error'

[Andrey Sledzinskiy]

Bug is reproduced on {"build_id": "2014-04-17_04-17-26", "mirantis": "yes", "build_number": "281", "nailgun_sha": "7a05e365240ab27c492b20585ef8ac8557102cc0", "ostf_sha": "3690709d95c1f42506b6d955b44221910cec8527", "fuelmain_sha": "e161450291dced29afda979560a5c8a78de4bb14", "astute_sha": "55df06b2e84fa5d71a1cc0e78dbccab5db29d968", "release": "4.1B", "fuellib_sha": "31673174fbfd4c1f82d9c44eb6eac0d84241be45"}

Steps:
1. Create next cluster - HA, Centos, KVM, Flat Nova network without tagging, Cinder LVM
2. Add 3 controlles, 2 compute nodes
3. Deploy cluster
4. After successful deploy ssh to controller and execute 'crm_mon -1'

Actual result :
3 Nodes configured, 3 expected votes
9 Resources configured

Online: [ node-1.test.domain.local node-3.test.domain.local node-5.test.domain.local ]

 vip__management_old (ocf::mirantis:ns_IPaddr2): Started node-1.test.domain.local
 vip__public_old (ocf::mirantis:ns_IPaddr2): Started node-1.test.domain.local
 Clone Set: clone_p_haproxy [p_haproxy]
     Started: [ node-1.test.domain.local node-3.test.domain.local node-5.test.domain.local ]
 Clone Set: clone_p_mysql [p_mysql]
     Started: [ node-1.test.domain.local node-3.test.domain.local node-5.test.domain.local ]
 openstack-heat-engine (ocf::mirantis:openstack-heat-engine): Started node-3.test.domain.local

Failed actions:
    vip__management_old_monitor_2000 (node=node-1.test.domain.local, call=11, rc=1, status=Timed Out, last-rc-change=Thu Apr 17 16:39:45 2014
, queued=0ms, exec=0ms
): unknown error
    vip__public_old_monitor_2000 (node=node-1.test.domain.local, call=22, rc=1, status=Timed Out, last-rc-change=Thu Apr 17 16:39:45 2014
, queued=0ms, exec=0ms
): unknown error

Logs are attached

[Vladimir Kuklin (vkuklin)]

report shows that some shell commands used for ipaddr resources monitoring, usually these are simple calls of ifconfig or `ip` commands. it is expected behaviour if these commands fail. this should not usually happen, but pacemaker should handle this situation and everything should work ok. so, please provide info, which tests have failed or which functionality is broken.

[Andrey Sledzinskiy (asledzinskiy)]

It turned out to be system test problem. We assert on (ocf::heartbeat:IPaddr2\) but now it's been changed to (ocf::mirantis:ns_IPaddr2) . We need to fix it in test

[Dmitry Borodaenko (angdraug)]

https://review.openstack.org/89981

[Dmitry Borodaenko (angdraug)]

https://review.openstack.org/90490

[Andrey Sledzinskiy (asledzinskiy)]

verified