[Pacemaker] Can't get all my resources online at once ?

[Karl Katzke]

Joe -  

First, if you're not using it, the CRM shell is -way- easier to do your configuration with.  

Second, I don't know if 100 is a high enough score to get them to colocate. You might need to set them to infinity. Or, I think you can add them to a group to get them to do the same thing. 

-K 


---
Karl Katzke
Systems Analyst II
TAMU - DRGS






>>> Joe Armstrong <jarmstrong at postpath.com> 5/13/2009 6:02 PM >>>
Hi,

I have a really simple config - 1 IP and an httpd but I can't get them both online on the same node.
I have had the httpd try to start on the node without the IP, I've had the IP get configured correctly but no hint of httpd & I've had the IP get moved away from a node while httpd was starting up so I had httpd on one node online and its IP online on the other!  Each restart of heartbeat seems to give a different result.  It seems like the order & co-location constraints are being ignored or something.

I am new to pacemaker so please forgive what must be an obvious error in my config.

Thanks for any help.

Joe

Installed Versions:

/RPMs/heartbeat-2.99.2-8.1.x86_64.rpm            /RPMs/openais-0.80.5-13.1.x86_64.rpm
/RPMs/heartbeat-common-2.99.2-8.1.x86_64.rpm     /RPMs/pacemaker-1.0.3-2.2.x86_64.rpm
/RPMs/heartbeat-resources-2.99.2-8.1.x86_64.rpm  /RPMs/pacemaker-mgmt-1.99.1-2.1.x86_64.rpm
/RPMs/libheartbeat2-2.99.2-8.1.x86_64.rpm        /RPMs/pacemaker-mgmt-client-1.99.1-2.1.x86_64.rpm
/RPMs/libopenais2-0.80.5-13.1.x86_64.rpm         /RPMs/python-lxml-1.3.4-1.el5.rf.x86_64.rpm
/RPMs/libpacemaker3-1.0.3-2.2.x86_64.rpm

Config file:

<cib validate-with="pacemaker-1.0" crm_feature_set="3.0.1" have-quorum="1" admin_epoch="0" epoch="42" num_updates="8" cib-last-written="Wed May 13 15:53:18 2009" dc-uuid="ff29bf8a-9ab6-423f-8002-b74a03b1f1d2">
  <configuration>
    <crm_config>
      <cluster_property_set id="cib-bootstrap-options">
        <nvpair id="cib-bootstrap-options-dc-version" name="dc-version" value="1.0.3-b133b3f19797c00f9189f4b66b513963f9d25db9"/>
      </cluster_property_set>
    </crm_config>
    <nodes>
      <node id="ff29bf8a-9ab6-423f-8002-b74a03b1f1d2" uname="vm2.jarmsdt.postpath.com" type="normal"/>
      <node id="f2dd7f3e-026c-4ba8-8694-786360cfd602" uname="vm1.jarmsdt.postpath.com" type="normal"/>
    </nodes>
    <resources>
      <primitive id="http-1" class="lsb" type="httpd">
    </primitive>
      <primitive id="http-ip-1" class="ocf" provider="heartbeat" type="IPaddr">
        <instance_attributes id="ip-459558">
          <nvpair id="nv013" name="ip" value="10.200.76.188"/>
          <nvpair id="nv014" name="device" value="auto"/>
          <nvpair id="nv015" name="netmask" value="255.255.255.0"/>
          <nvpair id="nv016" name="broadcast" value="10.200.76.255"/>
        </instance_attributes>
      </primitive>
    </resources>
    <constraints>
      <rsc_order first="http-ip-1" first-action="promote" id="http-order" score="INFINITY" then="http-1" then-action="start"/>
      <rsc_colocation id="http-colo" rsc="http-1" score="100" with-rsc="http-ip-1"/>
    </constraints>
    <rsc_defaults/>
    <op_defaults/>
  </configuration>
</cib>

_______________________________________________
Pacemaker mailing list
Pacemaker at oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20090513/3dc9a52a/attachment.htm>