[Pacemaker] stonith resource issue
Shravan Mishra
shravan.mishra at gmail.com
Fri Oct 8 12:54:28 EDT 2010
Looks like I found the issue but I don't know how to fix it:
I looked into run_stonith_agent() in st_client.c
It is calling fence_legacy instead of external/ipmi.
Since I don't know how to enable tracing for different subsystems so I
put my own message in the code and this is what I got:
========
Oct 8 12:32:25 c687 stonith-ng: [1436]: notice: run_stonith_agent:
action monitor with device fence_legacy
===========
I'm using external/ipmi.
This is what my cib looks like:
=======
node ha1.itactics.com \
attributes standby="off"
node ha2.itactics.com \
attributes standby="off"
primitive ha1.itactics.com-stonith stonith:external/safe/ipmi \
op monitor interval="20" timeout="180" \
params target_role="started" hostname="ha1.itactics.com" ipaddr="192.168.2.3"
primitive ha2.itactics.com-stonith stonith:external/safe/ipmi \
op monitor interval="20" timeout="180" \
params target_role="started" hostname="ha2.itactics.com" ipaddr="192.168.2.7"
location ha1.itactics.com-stonith-placement ha1.itactics.com-stonith \
rule $id="ri-ha1.itactics.com-stonith-placement-1" -inf: #uname eq
ha1.itactics.com
location ha2.itactics.com-stonith-placement ha2.itactics.com-stonith \
rule $id="ri-ha2.itactics.com-stonith-placement-1" -inf: #uname eq
ha2.itactics.com
property $id="cib-bootstrap-options" \
dc-version="1.1.2-e0d731c2b1be446b27a73327a53067bf6230fb6a" \
cluster-infrastructure="openais" \
expected-quorum-votes="2"
======
I'm confused why fence-legacy is getting called.
Another side question is that I have build pacemaker with the following flag
--with-trace-date
how do I enable tracing to see the all the useful information that is
printed out.
Please help.
Thanks
Shravan
On Thu, Oct 7, 2010 at 10:04 AM, Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:
> Hi,
>
> On Wed, Oct 06, 2010 at 01:32:06PM -0400, Shravan Mishra wrote:
>> Please fine hb_report.
>
> hb_report couldn't find the logs, probably because you have both
> syslog and to file logging. Anyway, it could be that stuff such
> as external/safe/ipmi cannot work, i.e. that you can't create
> subdirectories within external. Please try to move your agent
> one directory up.
>
> Thanks,
>
> Dejan
>
>> Thanks
>> Shravan
>>
>> On Wed, Oct 6, 2010 at 1:14 PM, Shravan Mishra <shravan.mishra at gmail.com> wrote:
>> > Hi,
>> >
>> > Please find the stonith and crmd logs attached.
>> >
>> > I have pruned stonith.logs as it contained lots of repeatable messages.
>> >
>> > I 'm in the process of installing Date::parse perl module then I'll
>> > send the info from hb_report.
>> >
>> > Thanks for your help
>> >
>> > Shravan
>> >
>> >
>> > On Wed, Oct 6, 2010 at 11:26 AM, Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:
>> >> Hi,
>> >>
>> >> On Wed, Oct 06, 2010 at 11:04:34AM -0400, Shravan Mishra wrote:
>> >>> Hi guys,
>> >>>
>> >>> I'm having a weird problem with my stonith resources.They are
>> >>> constantly starting and stopping.
>> >>>
>> >>> I'm using:
>> >>>
>> >>> pacemaker=1.1.3
>> >>> corosync=1.2.8
>> >>> glue=glue_1.0-10
>> >>
>> >> Hmm, which version is this really? Can you do hb_report -V? You
>> >> should be running at least 1.0.6.
>> >>
>> >>> 2.6.29.6-0.6.smp.gcc4.1.x86_64
>> >>>
>> >>> My configuration looks like this:
>> >>>
>> >>> =======================
>> >>> node ha1.itactics.com
>> >>> node ha2.itactics.com
>> >>> primitive ha1.itactics.com-stonith stonith:external/safe/ipmi \
>> >>> op monitor interval="20" timeout="180" \
>> >>> params target_role="started" hostname="ha1.itactics.com" ipaddr="192.168.2.3"
>> >>> primitive ha2.itactics.com-stonith stonith:external/safe/ipmi \
>> >>> op monitor interval="20" timeout="180" \
>> >>> params target_role="started" hostname="ha2.itactics.com" ipaddr="192.168.2.7"
>> >>> location ha1.itactics.com-stonith-placement ha1.itactics.com-stonith \
>> >>> rule $id="ri-ha1.itactics.com-stonith-placement-1" -inf: #uname eq
>> >>> ha1.itactics.com
>> >>
>> >> You can reduce such locations:
>> >>
>> >> location ha1.itactics.com-stonith-placement ha1.itactics.com-stonith -inf: ha1.itactics.com
>> >>
>> >>> location ha2.itactics.com-stonith-placement ha2.itactics.com-stonith \
>> >>> rule $id="ri-ha2.itactics.com-stonith-placement-1" -inf: #uname eq
>> >>> ha2.itactics.com
>> >>> property $id="cib-bootstrap-options" \
>> >>> dc-version="1.1.2-e0d731c2b1be446b27a73327a53067bf6230fb6a" \
>> >>> cluster-infrastructure="openais" \
>> >>> expected-quorum-votes="2" \
>> >>> stonith-enabled="true"
>> >>> =========================
>> >>>
>> >>> An excerpt from /var/log/messages
>> >>> ==========================
>> >>> Oct 6 11:00:02 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1150: monitor
>> >>> Oct 6 11:00:03 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1150] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>
>> >> Hmm, there should be more in the logs. Where do crmd and
>> >> stonith-ng log?
>> >>
>> >> Thanks,
>> >>
>> >> Dejan
>> >>
>> >>> Oct 6 11:00:03 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1151: stop
>> >>> Oct 6 11:00:03 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1152: start
>> >>> Oct 6 11:00:03 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1153: monitor
>> >>> Oct 6 11:00:04 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1153] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>> Oct 6 11:00:04 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1154: stop
>> >>> Oct 6 11:00:04 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1155: start
>> >>> Oct 6 11:00:04 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1156: monitor
>> >>> Oct 6 11:00:06 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1156] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>> Oct 6 11:00:06 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1157: stop
>> >>> Oct 6 11:00:06 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1158: start
>> >>> Oct 6 11:00:06 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1159: monitor
>> >>> Oct 6 11:00:07 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1159] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>> Oct 6 11:00:07 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1160: stop
>> >>> Oct 6 11:00:08 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1161: start
>> >>> Oct 6 11:00:08 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1162: monitor
>> >>> Oct 6 11:00:09 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1162] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>> Oct 6 11:00:09 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1163: stop
>> >>> Oct 6 11:00:09 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1164: start
>> >>> Oct 6 11:00:12 ha1 lrmd: [5994]: info: stonithRA plugin: got
>> >>> metadata: <?xml version="1.0"?> <!DOCTYPE resource-agent SYSTEM
>> >>> "ra-api-1.dtd"> <resource-agent name="external/safe/ipmi">
>> >>> <version>1.0</version> <longdesc lang="en"> <!-- no value -->
>> >>> </longdesc> <shortdesc lang="en"><!-- no value
>> >>> --></shortdesc> <!-- no value --> <actions> <action
>> >>> name="start" timeout="15" /> <action name="stop" timeout="15"
>> >>> /> <action name="status" timeout="15" /> <action
>> >>> name="monitor" timeout="15" interval="15" start-delay="15" />
>> >>> <action name="meta-data" timeout="15" /> </actions> <special
>> >>> tag="heartbeat"> <version>2.0</version> </special>
>> >>> </resource-agent>
>> >>> Oct 6 11:00:12 ha1 lrmd: [5994]: info: G_SIG_dispatch: started at
>> >>> 1726679412 should have started at 1726679110
>> >>> Oct 6 11:00:12 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1165: monitor
>> >>> Oct 6 11:00:13 ha1 lrmd: [5994]: info: cancel_op: operation
>> >>> monitor[1165] on stonith::external/safe/ipmi::ha2.itactics.com-stonith
>> >>> for client 5997, its parameters: CRM_meta_interval=[20000]
>> >>> target_role=[started] ipaddr=[192.168.2.7] CRM_meta_timeout=[180000]
>> >>> crm_feature_set=[3.0.2] CRM_meta_name=[monitor]
>> >>> hostname=[ha2.itactics.com] cancelled
>> >>> Oct 6 11:00:13 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1166: stop
>> >>> Oct 6 11:00:13 ha1 lrmd: [5994]: info: rsc:ha2.itactics.com-stonith:1167: start
>> >>> Oct 6 11:00:13 ha1 lrmd: [5994]: info:
>> >>> rsc:ha2.itactics.com-stonith:1168: monitor
>> >>>
>> >>> ===========================
>> >>>
>> >>>
>> >>> I'm on a critical path for this release.
>> >>> I would really appreciate a quick help on this.
>> >>>
>> >>>
>> >>>
>> >>> Thanks a lot.
>> >>>
>> >>>
>> >>> Shravan
>> >>>
>> >>> _______________________________________________
>> >>> Pacemaker mailing list: Pacemaker at 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>> >>
>> >> _______________________________________________
>> >> Pacemaker mailing list: Pacemaker at 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>> >>
>> >
>
>
>> _______________________________________________
>> Pacemaker mailing list: Pacemaker at 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>
More information about the Pacemaker
mailing list