[Pacemaker] Migrate/run resource only if m/s resource is master/promoted on target node (follow-up)

[hj lee]

Hi,

Maybe this is related to no-quorum-policy. What is no-quorum-policy? You can
check no-quorum-policy in "crm configure show" command. If you can not see
in "crm configure show" command, then it is "stop" by default. If that is
your case, then please set it to ignore by "crm configure property
no-quorum-policy=ignore". and try it again.

Thanks
hj

On Tue, Dec 29, 2009 at 6:43 AM, Martin Gombač <martin at isg.si> wrote:

> Hi guys once again :-)
>
> my resource Hosting on top of ms_drbd_r0 keeps restarting even if the
> changes aren't local to the node.
> By that, i mean Hosting get's restarted on node1 even if i restart or
> outdate and demote node 2.
>
> My constrains:
> colocation Hosting_on_ms_drbd_r0 inf: Hosting ms_drbd_r0:Master
> order ms_drbd_r0_b4_Hosting inf: ms_drbd_r0:promote Hosting:start
>
> My resources:
> primitive Hosting ocf:heartbeat:Xen \
>   params xmfile="/etc/xen/Hosting.cfg" \
>   meta target-role="Started" allow-migrate="true" is-managed="true" \
>   op monitor interval="120s" timeout="300s"
> primitive drbd_r0 ocf:linbit:drbd \
>   params drbd_resource="r0" \
>   op monitor interval="15s" role="Master" timeout="30s" \
>   op monitor interval="30s" role="Slave" timeout="30"
> ms ms_drbd_r0 drbd_r0 \
>   meta notify="true" master-max="2" inteleave="true"
>
> I use location to:
> location cli-prefer-Hosting Hosting \
>   rule $id="cli-prefer-rule-Hosting" inf: #uname eq ibm1
>
> Example:
> Shuting down heartbeat on node2/ibm2, while resource is running perfectly
> fine on node1, makes resource Hosting restart on node1.
>
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: native_print: Hosting
>  (ocf::heartbeat:Xen):    Started ibm1
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: clone_print:  Master/Slave
> Set: ms_drbd_r0
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: short_print:      Masters: [
> ibm1 ibm2]
> Dec 29 14:21:28 ibm1 pengine: [3716]: WARN: native_color: Resource
> drbd_r0:1 cannot run anywhere
> Dec 29 14:21:28 ibm1 pengine: [3716]: info: master_color: Promoting
> drbd_r0:0 (Master ibm1)
> Dec 29 14:21:28 ibm1 pengine: [3716]: info: master_color: ms_drbd_r0:
> Promoted 1 instances of a possible 2 to master
> Dec 29 14:21:28 ibm1 pengine: [3716]: info: master_color: Promoting
> drbd_r0:0 (Master ibm1
> Dec 29 14:21:28 ibm1 pengine: [3716]: info: master_color: ms_drbd_r0:
> Promoted 1 instances of a possible 2 to master
> Dec 29 14:21:28 ibm1 pengine: [3716]: info: stage6: Scheduling Node ibm2
> for shutdown
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: LogActions: Restart resource
> Hosting    (Started ibm1)
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: LogActions: Leave resource
> drbd_r0:0    (Master ibm1)
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: LogActions: Demote drbd_r0:1
>    (Master -> Stopped ibm2)
> Dec 29 14:21:28 ibm1 pengine: [3716]: notice: LogActions: Stop resource
> drbd_r0:1    (ibm2)
> ...
> Dec 29 14:21:28 ibm1 crmd: [3713]: info: te_rsc_command: Initiating action
> 8: stop Hosting_stop_0 on ibm1(local)
> Dec 29 14:21:28 ibm1 crmd: [3713]: info: do_lrm_rsc_op: Performing
> key=8:4:0:13282265-c62f-4341-9fa5-363cd30ddd3e op=Hosting_stop_0 )
> Dec 29 14:21:28 ibm1 lrmd: [3710]: info: rsc:Hosting:19: stop
> ...
> Dec 29 14:21:34 ibm1 crmd: [3713]: info: match_graph_event: Action
> Hosting_stop_0 (8) confirmed on ibm1..si (rc=0)
> Dec 29 14:21:34 ibm1 crmd: [3713]: info: te_rsc_command: Initiating action
> 9: start Hosting_start_0 on ibm1..si (local)
> Dec 29 14:21:34 ibm1 lrmd: [3710]: info: rsc:Hosting:21: start
> ....
> ...
> .
>
>
> Please advise me on how to configure ordering and collocation constrains,
> to not get Hosting resource restarted each time when something happens to
> backup drbd resource on the second node.
>
>
> Thank you.
> Martin
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>



-- 
Dream with longterm vision!
kerdosa
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091229/31c96101/attachment-0001.htm>