Bug 1208273

Summary:	crm cluster stop took too long to complete (~1hr)
Product:	[SUSE Linux Enterprise High Availability Extension] PUBLIC SUSE Linux Enterprise High Availability 15 SP3	Reporter:	Pri Bal <priyanka.14balotra>
Component:	Pacemaker	Assignee:	Yan Gao <ygao>
Status:	RESOLVED DUPLICATE	QA Contact:	SUSE Linux Enterprise High Availability Team <ha-bugs>
Severity:	Normal
Priority:	P5 - None	CC:	priyanka.14balotra
Version:	unspecified
Target Milestone:	unspecified
Hardware:	Other
OS:	Other
Whiteboard:
Found By:	---	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---

Description Pri Bal 2023-02-15 06:32:02 UTC

crm cluster stop (cluster level operation) took too long to complete. The stop operation began at 17:21:04 and completed at 18:21:03 


Oct 01 17:21:04 FILE-2 pacemakerd          [30001] (stop_child)         notice: Stopping pacemaker-based | sent signal 15 to process 30015
Oct 01 17:21:04 FILE-2 pacemaker-based     [30015] (crm_signal_dispatch)        notice: Caught 'Terminated' signal | 15 (invoking handler)
Oct 01 17:21:04 FILE-2 pacemaker-based     [30015] (cib_shutdown)       info: Waiting on 1 clients to disconnect (0)
Oct 01 17:21:04 FILE-2 pacemaker-based     [30015] (cib_shutdown)       info: All clients disconnected (0)
Oct 01 17:21:04 FILE-2 pacemaker-based     [30015] (terminate_cib)      info: initiate_exit: Exiting from mainloop...
Oct 01 17:21:04 FILE-2 pacemaker-based     [30015] (crm_cluster_disconnect)     info: Disconnecting from cluster infrastructure: corosync
Oct 01 17:21:04 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:21:04 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:21:05 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:21:05 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:21:05 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
.
.
.
Oct 01 17:24:03 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:24:03 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (crm_cs_flush)       info: Sent 0 CPG messages  (2 remaining, last=13): Try again (6)
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (escalate_shutdown)  error: Child pacemaker-based not terminating in a timely manner, forcing
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (stop_child)         notice: Stopping pacemaker-based | sent signal 11 to process 30015
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (pcmk_child_exit)    error: pacemaker-based[30015] terminated with signal 11 (core=0)
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (pcmk_shutdown_worker)       notice: Shutdown complete
Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (qb_ipcs_us_withdraw)        info: withdrawing server sockets
Oct 01 18:21:03 FILE-2 pacemakerd          [30001] (crm_xml_cleanup)    info: Cleaning up memory from libxml2
Oct 01 18:21:03 FILE-2 pacemakerd          [30001] (crm_exit)   info: Exiting pacemakerd | with status 0

Comment 1 Yan Gao 2023-02-15 14:47:34 UTC

(In reply to Pri Bal from comment #0)
> Oct 01 17:24:04 FILE-2 pacemakerd          [30001] (qb_ipcs_us_withdraw)    
> info: withdrawing server sockets

Was pacemaker-controld already stopped by then? Was there anything else logged in pacemaker.log or syslog after this and until 18:21:03?

Could it be the system encountering certain difficulties such as high load or somehow freezing/hanging?

Comment 2 Yan Gao 2023-03-16 09:38:24 UTC

This looks like a duplicate of bsc#1205214.

Let's track it there since supportconfig was provided there.

*** This bug has been marked as a duplicate of bug 1205214 ***