Pacemaker kills RabbitMQ

Bug #1512754 reported by Andrey Grebennikov
14
This bug affects 3 people
Affects Status Importance Assigned to Milestone
Mirantis OpenStack
Fix Released
High
Alexey Lebedeff
6.0.x
Won't Fix
High
MOS Maintenance
6.1.x
Fix Released
High
Sergii Rizvan
7.0.x
Fix Released
High
Sergii Rizvan
8.0.x
Fix Released
High
Alexey Lebedeff

Bug Description

Fuel 6.1, Ubuntu,
RabbitMQ OCF script updated to the latest from the MU3.

We noticed the instance of RabbitMQ was re-created some time ago, it turned out Pacemaker restarted it.
This is a part of a log from Pacemaker:

Oct 29 20:45:15 [10696] controller3 pacemaker_remoted: notice: operation_finished: p_rabbitmq-server_monitor_30000:29874:stderr [ ^M ]
Oct 29 20:45:15 [10696] controller3 pacemaker_remoted: notice: operation_finished: p_rabbitmq-server_monitor_30000:29874:stderr [ Crash dump was written to: erl_crash.dump^M ]
Oct 29 20:45:15 [10696] controller3 pacemaker_remoted: notice: operation_finished: p_rabbitmq-server_monitor_30000:29874:stderr [ Kernel pid terminated (application_controller) ({application_start_failure,kernel,{{shutdown,{failed_to_start_child,net_sup,{shutdown,{failed_to_start_child,auth,{"Too short cookie string",[{auth,in^M ]
Oct 29 20:45:15 [10699] controller3 crmd: notice: process_lrm_event: Operation p_rabbitmq-server_monitor_30000: unknown error (node=controller3, call=40805, rc=1, cib-update=275893, confirmed=false)
Oct 29 20:45:15 [10699] controller3 crmd: notice: process_lrm_event: GGUTTPLDI003.ebiz.verizon.com-p_rabbitmq-server_monitor_30000:40805 [ \r\nCrash dump was written to: erl_crash.dump\r\nKernel pid terminated (application_controller) ({application_start_failure,kernel,{{shutdown,{failed_to_start_child,net_sup,{shutdown,{failed_to_start_child,auth,{"Too short cookie string",[{auth,in\r\n ]

Revision history for this message
Andrey Grebennikov (agrebennikov) wrote :

Seems I hit the same bug with erlang cookies once again, but the output is a bit different:

Nov 13 17:10:52 [10699] controller3 crmd: notice: process_lrm_event: controller3-p_rabbitmq-server_monitor_30000:40837 [ \r\nCrash dump was written to: erl_crash.dump\r\nKernel pid terminated (application_controller) ({application_start_failure,kernel,{{shutdown,{failed_to_start_child,net_sup,{shutdown,{failed_to_start_child,auth,{{case_clause,eof},[{auth,read_cooki\r\n ]

Revision history for this message
Andrey Grebennikov (agrebennikov) wrote :
tags: added: customer-found support
Revision history for this message
Polina Petriuk (ppetriuk) wrote :
Revision history for this message
Alexey Lebedeff (alebedev-a) wrote :

Yes, attached crashdump is also about this bug.

Revision history for this message
Polina Petriuk (ppetriuk) wrote :
Revision history for this message
Sergii Rizvan (srizvan) wrote :
Sergii Rizvan (srizvan)
tags: added: on-verification
Sergii Rizvan (srizvan)
tags: removed: on-verification
Revision history for this message
Timur Nurlygayanov (tnurlygayanov) wrote :

Not reproduced on the latest MOS 8.0 ISO on my environment.

Revision history for this message
Sergii Rizvan (srizvan) wrote :

Verified on 6.1 on Ubuntu.
Packages:
fuel-ha-util
Version:
6.1.0-6777.1

Sergii Rizvan (srizvan)
tags: added: on-verification
Revision history for this message
Sergii Rizvan (srizvan) wrote :

Verified on 7.0 on Ubuntu.
Packages:
fuel-ha-utils
Version:
7.0.0-7257.1.gita808cd9

tags: removed: on-verification
tags: added: 7.0-mu-2
Revision history for this message
Roman Rufanov (rrufanov) wrote :

please check on 6.0

Revision history for this message
Alexey Stupnikov (astupnikov) wrote :

We no longer support MOS5.1, MOS6.0, MOS6.1
We deliver only Critical/Security fixes to MOS7.0, MOS8.0.
We deliver only High/Critical/Security fixes to MOS9.2.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.