[Pacemaker] pingd process dies for no reason

Michael Schwartzkopff misch at clusterbau.com
Fri Jan 7 14:00:47 UTC 2011


On Friday 07 January 2011 14:56:03 Patrik.Rapposch at knapp.com wrote:
> Greetings,
> 
> we have a problem, that the ping daemon dies for no reason and we can't
> find why this happened.
> 
> we use following versions on SLES 11.1:
> 
> libpacemaker3-1.1.2-0.6.1
> pacemaker-mgmt-2.0.0-0.3.10
> pacemaker-mgmt-client-2.0.0-0.3.10
> drbd-pacemaker-8.3.8.1-0.2.9
> libpacemaker-devel-1.1.2-0.6.1
> pacemaker-1.1.2-0.6.1
> pacemaker-mgmt-devel-2.0.0-0.3.10
> libcorosync4-1.2.6-0.2.2
> corosync-1.2.6-0.2.2
> libcorosync-devel-1.2.6-0.2.2
> 
> here is the important part of the log trace:
> "
> Jan  5 08:40:30 node2 lrmd: [5990]: info: rsc:OSR_IP:46535: monitor
> Jan  5 08:40:30 node2 lrmd: [5990]: info: rsc:Cluster_IP:46533: monitor
> Jan  5 08:40:33 node2 lrmd: [5990]: WARN: pingd:0:monitor process (PID
> 23937) timed out (try 1).  Killing with signal SIGTERM (15).
> Jan  5 08:40:33 node2 lrmd: [5990]: WARN: operation monitor[48559] on
> ocf::ping::pingd:0 for client 5993, its parameters: CRM_meta_clone=[0]
> host_list=[xxx.xxx.xxx.xxx] CRM_meta_clone_node_max=[1]
> CRM_meta_clone_max=[2] CRM_meta_notify=[false] dampen=[5s]
> CRM_meta_globally_unique=[false] crm_feature_set=[3.0.2] multiplier=[100]
> CRM_meta_name=[monitor] CRM_meta_interval=[15000] CRM_meta_timeout=[5000]
> 
> : pid [23937] timed out
> 
> Jan  5 08:40:33 node2 crmd: [5993]: ERROR: process_lrm_event: LRM
> operation pingd:0_monitor_15000 (48559) Timed Out (timeout=5000ms)
> Jan  5 08:40:33 node2 crmd: [5993]: WARN: update_failcount: Updating
> failcount for pingd:0 on node2 after failed monitor: rc=-2
> (update=value++, time=1294213233)
> Jan  5 08:40:35 node2 pengine: [5992]: notice: unpack_config: On loss of
> CCM Quorum: Ignore
> Jan  5 08:40:35 node2 pengine: [5992]: WARN: unpack_rsc_op: Processing
> failed op drbd_r0:1_promote_0 on node1: unknown exec error (-2)
> Jan  5 08:40:35 node2 pengine: [5992]: WARN: unpack_rsc_op: Processing
> failed op pingd:0_monitor_15000 on node2: unknown exec error (-2)
> Jan  5 08:40:35 node2 pengine: [5992]: notice: clone_print:  Clone Set:
> pingdclone [pingd]
> Jan  5 08:40:35 node2 pengine: [5992]: notice: native_print:      pingd:0
> (ocf::pacemaker:ping):  Started node2 FAILED
> Jan  5 08:40:35 node2 pengine: [5992]: notice: short_print:      Started:
> [ node1 ]"
> 
> the ressource is configured in following way:
> <clone id="pingdclone">
>         <meta_attributes id="pingdclone-meta_attributes">
>           <nvpair id="pingdclone-meta_attributes-globally-unique"
> name="globally-unique" value="false"/>
>         </meta_attributes>
>         <primitive class="ocf" id="pingd" provider="pacemaker"
> type="ping">
>           <instance_attributes id="pingd-instance_attributes">
>             <nvpair id="pingd-instance_attributes-host_list"
> name="host_list" value="xxx.xxx.xxx.xxx"/>
>             <nvpair id="pingd-instance_attributes-multiplier"
> name="multiplier" value="100"/>
>             <nvpair id="nvpair-96877c9e-2825-4d7d-997b-944652f89584"
> name="dampen" value="5s"/>
>           </instance_attributes>
>           <operations>
>             <op id="pingd-monitor-15s" interval="15s" name="monitor"
> timeout="5s"/>
>           </operations>
>         </primitive>
>       </clone>
> 
> thx for your help in advance.
> 
> Mit freundlichen Grüßen / Best Regards
> 
> Patrik Rapposch, BSc

Please use the "ping" resource agent instead of the "pingd"

Greetings,

-- 
Dr. Michael Schwartzkopff
Guardinistr. 63
81375 München

Tel: (0163) 172 50 98
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20110107/f12a8538/attachment-0004.sig>


More information about the Pacemaker mailing list