[Pacemaker] Colocation constraint getting removed

Andrew Beekhof

2015-02-23 23:02:25 UTC

Post by Arjun Pandey
Any pointers on this would be helpful.

Constraints don't get removed automatically unless someone asked for a resource that it references to be deleted.
Other possibilities include, someone asked to delete the constraint and someone uploaded an old config that didn't have it yet.

Post by Arjun Pandey
Regards
Arjun
I am running a 2 node cluster(active-standby mode) with this config
CentOS-6.5
pacemaker 1.1.10-14.el6_5.3
pcs 0.9.90-2.el6.centos.3 - (none) noarch
corosync 1.4.1-17.el6_5.1
Online: [ lakshman shatrughan ]
Master/Slave Set: foo-master [foo]
Masters: [ shatrughan ]
Slaves: [ lakshman ]
eth1-CP (ocf::pw:IPaddr): Started shatrughan
eth2-UP (ocf::pw:IPaddr): Started shatrughan
eth3-UPCP (ocf::pw:IPaddr): Started shatrughan
where IPaddr RA is just modified IPAddr2 RA and foo is a multi-state resource. Additionally i have a
collocation constraint for the IP addr to be collocated with the master.
I have set the migration-threshold as 2 for the VIP. I also have set the failure-timeout to 15s.
All virtual IP's have following properties
Meta Attrs: failure-timeout=3s migration-threshold=2
Operations: monitor interval=1s (RAN-monitor-interval-1s)
However after some time i notice that the colocation constraint i had applied has got removed and the resource get's started where foo standby is located.
From the logs
Dec 17 16:15:32 [27112] lakshman pengine: notice: process_pe_message: Calculated Transition 3: /var/lib/pacemaker/pengine/pe-input-1297.bz2
Dec 17 16:15:32 [27113] lakshman crmd: info: do_state_transition: State transition S_POLICY_ENGINE -> S_TRANSITION_ENGINE [ input=I_PE_SFOOCESS cause=C_IPC_MESSAGE origin=handle_response ]
Dec 17 16:15:32 [27113] lakshman crmd: info: do_te_invoke: Processing graph 3 (ref=pe_calc-dc-1418813132-33) derived from /var/lib/pacemaker/pengine/pe-input-1297.bz2
Dec 17 16:15:32 [27113] lakshman crmd: notice: te_rsc_command: Initiating action 12: monitor foo_monitor_10000 on lakshman (local)
Dec 17 16:15:32 [27113] lakshman crmd: info: do_lrm_rsc_op: Performing key=12:3:8:62cbe40f-174d-458b-8e52-8581e57245a0 op=foo_monitor_10000
Dec 17 16:15:32 [27113] lakshman crmd: notice: process_lrm_event: LRM operation foo_monitor_10000 (call=77, rc=8, cib-update=52, confirmed=false) master
Dec 17 16:15:32 [27108] lakshman cib: info: cib_process_request: Completed cib_modify operation for section status: OK (rc=0, origin=local/crmd/52, version=0.652.73)
Dec 17 16:15:32 [27113] lakshman crmd: info: match_graph_event: Action foo_monitor_10000 (12) confirmed on lakshman (rc=0)
Dec 17 16:15:32 [27113] lakshman crmd: notice: run_graph: Transition 3 (Complete=1, Pending=0, Fired=0, Skipped=0, Incomplete=0, Source=/var/lib/pacemaker/pengine/pe-input-1297.bz2): Complete
Dec 17 16:15:32 [27113] lakshman crmd: info: do_log: FSA: Input I_TE_SFOOCESS from notify_crmd() received in state S_TRANSITION_ENGINE
Dec 17 16:15:32 [27113] lakshman crmd: notice: do_state_transition: State transition S_TRANSITION_ENGINE -> S_IDLE [ input=I_TE_SFOOCESS cause=C_FSA_INTERNAL origin=notify_crmd ]
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28098 id=4d1ef70d-3f75-4308-9c1b-cc410a6e6940
Dec 17 16:15:32 [27108] lakshman cib: info: cib_process_request: Completed cib_query operation for section 'all': OK (rc=0, origin=local/crm_resource/2, version=0.652.73)
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28102 id=14b4a917-4df1-46ed-82e3-0a023b92ed11
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28103 id=4ab59dd4-7f13-4ae6-84cf-ae9c2de5e246
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28104 id=7d24f5e0-2ae2-47be-a88e-1635e1268d3c
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28105 id=c583545c-e590-4f5a-a865-e063d21f35cd
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28106 id=d7dc6094-4c48-4246-94b3-dc2b0a12866f
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28107 id=757a2e15-a1d4-4f3b-b93a-0051b5a9be15
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28108 id=1c9bfb99-bd55-42dc-b1bd-df32d560af90
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28109 id=1b7ede71-66ec-43ad-9754-b204f367ccc3
Dec 17 16:15:32 [27108] lakshman cib: info: cib_process_request: Completed cib_query operation for section 'all': OK (rc=0, origin=local/cibadmin/2, version=0.652.73)
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28110 id=ebd97f0f-81fc-4fd3-adae-30fb9784b563
Dec 17 16:15:32 [27108] lakshman cib: info: cib_process_request: Completed cib_query operation for section //constraints: OK (rc=0, origin=local/cibadmin/2, version=0.652.73)
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_destroy: Destroying 0 events
Dec 17 16:15:32 [27108] lakshman cib: info: crm_client_new: Connecting 0x20bb1d0 for uid=0 gid=0 pid=28111 id=e4891335-7a9b-4189-ba3d-03008b905664
Dec 17 16:15:32 [27108] lakshman cib: info: cib_replace_notify: Replaced: 0.652.73 -> 0.653.1 from lakshman
Dec 17 16:15:32 [27108] lakshman cib: notice: cib:diff: Diff: --- 0.652.73
Dec 17 16:15:32 [27108] lakshman cib: notice: cib:diff: Diff: +++ 0.653.1 dfc649135fe2d861910a19c9f983cc4a
Dec 17 16:15:32 [27108] lakshman cib: notice: cib:diff: -- <rsc_colocation id="colocation-Mgmt_Float-foo-master-INFINITY" rsc="Mgmt_Float" rsc-role="Started" score="INFINITY" with-rsc="foo-master" with-rsc-role="Master"/>
PFA the cluster-log file
I can't really make out why the constraint got removed in the first place.
Thanks for help in advance.
Regards
Arjun
_______________________________________________
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: ***@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