On 07/04/2012 09:16 PM, Irfan Ali wrote: > Hi all, > > We are trying to set-up an HA pair on RHEL 6.2 using DRBD (v > 8.4.1-2), Pacemaker (v 1.1.6-3) and Corosync (v 1.4.1-4). We could > make DRBD work independently syncing the two machines of the pair. But > our problem begins when we try to connect DRBD with Pacemaker. Even > though Pacemaker was able to detect the resources corresponding to > DRBD running on both the machines, it does not promoted any one of > them as master. I have pasted the output from 'crm status' below, it > shows that the resource 'ms-drbd' is running as slaves on both the > machines of our HA pair.
Please, don't paste snippets ... only full configurations are useful ...
more inline
>
> We fiddled a lot with the configuration for both Pacemaker and DRBD,
> but couldn't find anything to fix this problem. Any help / suggestions
> from you guys will be highly appreciated.
>
> ===========================================================================================================
>
> [root@c713 linbit]# crm status
>
> ============
>
> Last updated: Wed Jul 4 08:42:37 2012
>
> Last change: Wed Jul 4 07:34:53 2012 via crm_resource on c710.siemlab.com
>
> Stack: openais
>
> Current DC: c710.siemlab.com - partition with quorum
>
> Version: 1.1.6-3.el6-a02c0f19a00c1eb2527ad38f146ebc0834814558
>
> 2 Nodes configured, 2 expected votes
>
> 14 Resources configured.
>
> ============
>
>
>
> Online: [ c713.siemlab.com c710.siemlab.com ]
>
>
>
> Clone Set: connectivity [ping]
>
> Started: [ c710.siemlab.com c713.siemlab.com ]
What is the drbd state in /proc/drbd? everything ok?
>
> Clone Set: powerstatus [cpps0]
>
> Started: [ c710.siemlab.com c713.siemlab.com ]
>
> Master/Slave Set: ms-drbd [drbd0]
>
> Slaves: [ c710.siemlab.com c713.siemlab.com ]
>
> c713.siemlab.com-stonith (stonith:fence_ipmilan): Started
> c710.siemlab.com
>
> c710.siemlab.com-stonith (stonith:fence_ipmilan): Started
> c713.siemlab.com
>
>
>
> ===========================================================================================================
>
> We are using the following CIB configuration related to DRBD:
>
> <master id="ms-drbd">
>
> <meta_attributes id="ma-ms-drbd">
>
> <nvpair id="ma-ms-drbd-1" name="clone-max" value="2"/>
>
> <nvpair id="ma-ms-drbd-2" name="clone-node-max" value="1"/>
>
> <nvpair id="ma-ms-drbd-3" name="notify" value="yes"/>
>
> <nvpair id="ma-ms-drbd-4" name="globally-unique" value="false"/>
>
> <nvpair id="ma-ms-drbd-5" name="master-max" value="1"/>
>
> <nvpair id="ma-ms-drbd-6" name="master-node-max" value="1"/>
>
> <nvpair id="ma-ms-drbd-7" name="target-role" value="Master"/>
>
> <nvpair id="ma-ms-drbd-8" name="ordered" value="true"/>
remove that "ordered" attribute
>
> <nvpair id="ma-ms-drbd-crm-feature-set"
> name="crm-feature-set" value="3.0.5"/>
>
> <nvpair id="ma-ms-drbd-is-managed" name="is-managed" value="true"/>
>
> </meta_attributes>
>
> <primitive class="ocf" id="drbd0" provider="linbit" type="drbd">
>
> <instance_attributes id="ia-drbd">
>
> <nvpair id="ia-drbd-1" name="drbd_resource" value="var_nsm"/>
>
> </instance_attributes>
>
> <operations>
>
> <op id="op-drbd-1" interval="2min" name="monitor"
> role="Master" timeout="5min"/>
>
> <op id="op-drbd-2" interval="0" name="start" role="Master"
> timeout="4min"/>
>
> <op id="op-drbd-3" interval="0" name="stop" role="Master"
> timeout="4min"/>
you don't need roles for star/stop operations
>
> <op id="op-drbd-4" interval="3min" name="monitor"
> role="Slave" timeout="6min"/>
>
> <op id="op-drbd-5" interval="0" name="start" role="Slave"
> timeout="4min"/>
>
> <op id="op-drbd-6" interval="0" name="stop" role="Slave"
> timeout="4min"/>
>
> </operations>
>
> </primitive>
>
> </master>
>
> ===========================================================================================================
>
> The following is the content of drbd.conf:
>
> global {
>
> usage-count no;
>
> }
>
> common {
>
> net {
>
> protocol C;
>
> }
>
> startup {
>
> wfc-timeout 120;
>
> degr-wfc-timeout 120;
>
> }
>
> }
>
> resource var_nsm {
>
> disk {
>
> on-io-error detach;
>
> no-disk-barrier;
>
> no-disk-flushes;
>
> fencing resource-only;
>
> }
>
> handlers {
>
> fence-peer
> "/usr/lib/drbd/crm-fence-peer.sh";
>
> after-resync-target
> "/usr/lib/drbd/crm-unfence-peer.sh";
>
> }
>
> net {
>
> #rate 333M;
>
> after-sb-1pri
> discard-secondary;
... you know that can discard valid data?
Regards,
Andreas
--
Need help with Pacemaker?
http://www.hastexo.com/now
>
> max-buffers 8000;
>
> max-epoch-size 8000;
>
> sndbuf-size 0;
>
> }
>
> on c713.siemlab.com {
>
> device /dev/drbd1;
>
> disk /dev/sdb3;
>
> address 192.168.2.2:7791;
>
> meta-disk internal;
>
> }
>
> on c710.siemlab.com {
>
> device /dev/drbd1;
>
> disk /dev/sdb3;
>
> address 192.168.2.4:7791;
>
> meta-disk internal;
>
> }
>
> }
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker@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://bugs.clusterlabs.org
>
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Pacemaker mailing list: Pacemaker@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://bugs.clusterlabs.org
