I think I must be missing something simple.  My test cluster is 3
nodes (test01, test02, hyperaxe).  For the moment test01 is doing
nothing.  I'm trying to get drbd/nfs working on the other two.  The
DRBD bit was working great until I rolled in the services that depend
on it (filesystem, nfs, ip address).

Since I added the group containing the additional services, along with
the constraints that tie them to the DRBD resource, all of the
resources fail to run.  Note the output of crm_verify -L -V:

crm_verify[1392]: 2008/08/22_18:11:13 ERROR: unpack_rsc_op: Hard
error: drbd-www:0_monitor_0 failed with rc=5.
crm_verify[1392]: 2008/08/22_18:11:13 ERROR: unpack_rsc_op:
Preventing ms-drbd-www from re-starting on test01
crm_verify[1392]: 2008/08/22_18:11:13 WARN: unpack_rsc_op:
nfs_server_monitor_0 found active nfs_server on test01
crm_verify[1392]: 2008/08/22_18:11:13 WARN: unpack_rsc_op:
drbd-www:0_monitor_0 found active drbd-www:0 on test02
crm_verify[1392]: 2008/08/22_18:11:13 WARN: unpack_rsc_op:
drbd-www:0_monitor_0 found active drbd-www:0 on hyperaxe
crm_verify[1392]: 2008/08/22_18:11:13 WARN: native_color: Resource
drbd-www:0 cannot run anywhere
crm_verify[1392]: 2008/08/22_18:11:13 WARN: native_color: Resource
drbd-www:1 cannot run anywhere
crm_verify[1392]: 2008/08/22_18:11:13 WARN: native_color: Resource
fs_www cannot run anywhere
crm_verify[1392]: 2008/08/22_18:11:13 WARN: native_color: Resource
nfs_ip cannot run anywhere
crm_verify[1392]: 2008/08/22_18:11:13 WARN: native_color: Resource
nfs_server cannot run anywhere

The first error makes sense, as drbd is neither installed nor
configured on test01.  I don't know why it says it found an active
"nfs_server" on test01, as it's definitely not running.  I've since
removed the init script as well.  I also have no idea why drbd-www:0
is flagged as active on test02 and hyperaxe.  It appears the module is
loaded, apparently by the ocf drbd script as I've removed the system
drbd init script, but drbd is "Unconfigured".  No resources visible in
/proc/drbd.

And I'm assuming fs_www, nfs_ip, and nfs_server cannot run anywhere
because they have order/colocation constraints tying them to drbd-www.

The <resources> and <constraints> configs are pasted below.  The full
CIB can be viewed here:  http://pastebin.com/m50fc6123

I have symmetric-cluster = false.

What am I doing wrong?

Thanks!

-Ben

 <resources>
   <master_slave id="ms-drbd-www">
     <meta_attributes id="71bf85c6-176f-4d08-bfc3-22f62ead87eb">
       <attributes>
         <nvpair name="clone_max" value="2"
id="519fb092-7c05-4f19-abba-0a55773d6348"/>
         <nvpair name="clone_node_max" value="1"
id="4d480e7c-361f-4e85-8ff7-b36fa7228925"/>
         <nvpair name="master_max" value="1"
id="4a9b6670-42d7-4133-8fc5-4b8804ea49dd"/>
         <nvpair name="master_node_max" value="1"
id="73262db6-8c33-4e5a-adf6-e34fc8fd08ba"/>
         <nvpair name="notify" value="yes"
id="956bbe97-fc48-4f23-8b7e-7c048369e8e9"/>
         <nvpair name="globally_unique" value="false"
id="4e7daeda-4cf0-4c33-83c1-2de4a032476c"/>
         <nvpair name="target_role" value="stopped"
id="ee6d447a-c844-4bbb-91fe-9ce03536a2db"/>
       </attributes>
     </meta_attributes>
     <primitive id="drbd-www" class="ocf" provider="heartbeat" type="drbd">
       <instance_attributes id="34e97979-3d63-4033-9079-cc5b07ded44c">
         <attributes>
           <nvpair name="drbd_resource" value="www"
id="9e161e49-17a3-438a-979b-14aeb71a7416"/>
         </attributes>
       </instance_attributes>
       <operations>
         <op name="monitor" interval="59s" timeout="10s" role="Master"
id="1b2f0be4-b348-45e2-b2fe-dee233689880"/>
         <op name="monitor" interval="60s" timeout="10s" role="Slave"
id="5f1f5d3e-ec93-4358-b809-8b1a0c9a4436"/>
       </operations>
     </primitive>
   </master_slave>
   <group id="nfs">
     <meta_attributes id="32882692-c39a-4a23-bf44-9b08537383b0">
       <attributes>
         <nvpair name="target_role" value="#default"
id="33ae083e-0665-4f94-b6a2-0e92e4e10be1"/>
       </attributes>
     </meta_attributes>
     <primitive id="fs_www" class="ocf" type="Filesystem" provider="heartbeat">
       <instance_attributes id="404a647f-ca3b-46bd-9fe6-4f95a7099d77">
         <attributes>
           <nvpair name="device" value="/dev/drbd0"
id="111eb283-ebc8-4702-8852-ef17bc57a4f7"/>
           <nvpair name="directory" value="/www"
id="56bf20a3-37e8-400b-a92a-ef98e86dafa7"/>
           <nvpair name="fstype" value="xfs"
id="4220aaed-3300-425c-9c2b-9eee35376338"/>
         </attributes>
       </instance_attributes>
     </primitive>
     <primitive id="nfs_ip" class="ocf" type="IPaddr" provider="heartbeat">
       <instance_attributes id="nfs_ip">
         <attributes>
           <nvpair id="nfs_ip-ip" name="ip" value="192.168.22.213"/>
         </attributes>
       </instance_attributes>
     </primitive>
     <primitive id="nfs_server" class="lsb" type="nfs-kernel-server"/>
   </group>
 </resources>

 <constraints>
   <rsc_location id="drbd-www-loc-1" rsc="ms-drbd-www" node="hyperaxe"
score="100"/>
   <rsc_location id="drbd-www-loc-2" rsc="ms-drbd-www" node="test02"
score="100"/>
   <rsc_location id="drbd-www-loc-3" rsc="nfs" node="hyperaxe" score="100"/>
   <rsc_location id="drbd-www-loc-4" rsc="nfs" node="test02" score="100"/>
   <rsc_location id="drbd-www-master-loc-1" rsc="ms-drbd-www">
     <rule role="master" score="200" id="9c815988-d347-46a3-8556-382e6b07275f">
       <expression attribute="#uname" operation="eq" value="test02"
id="7bb8206f-d5f5-419d-99a1-f0bf6ef93c3b"/>
     </rule>
   </rsc_location>
   <rsc_order from="nfs" action="start" to="ms-drbd-www"
to_action="promote" id="3332c7f7-41ba-426a-9ebb-56793d990549"/>
   <rsc_colocation to="ms-drbd-www" to_role="master" from="nfs"
score="infinity" id="3f453dc6-0db2-43db-9701-73a52d99145a"/>
 </constraints>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to