[Linux-HA] heartbeat 2.0.2 problem on two nodes

Guochun Shi gshi at ncsa.uiuc.edu
Wed Oct 12 10:21:42 MDT 2005

My guess is there is some communication problem that two nodes cannot see each other,
but wee need log here to see what the problem is.


At 02:56 PM 10/11/2005 +0200, you wrote:
>I am running heartbeat 2.0.2 with this config on two nodes. I start heartbeat on node1 and it starts IPaddr resource, but then I start heartbeat on node2 and stop node1 and node2 doesnt start resources, it doesnt discover failure and cibdamin -Q show a different cib.xml than on node1. Why I dont get two nodes synced? If I started node2 with cib.xml empty and it generates  an empty one. If I copy node1 cib.xml to node2.xml then I have the IPaddr resource up on both nodes!
>Another question is good to have suppress_cib_writes to true, is this just for updating and addind uncomplete id fields to xml? The problem is that the cib.xml get to confused with uids, states, etc. how can this be solved?
>*** <http://ha.cf>ha.cf
>debugfile /var/log/ha/ha.debug
>logfile /var/log/ha/ha.log
>node node1
>node node2
>keepalive 2
>deadtime 10
>ucast eth0 <>MailScanner warning: numerical links are often malicious:<>
>ping <>MailScanner warning: numerical links are often malicious:<>
>auto_failback no
>crm yes
>use_logd yes
>debug 1
>*** cib.xml:
> <cib generated="true" cib_feature_revision="1" admin_epoch="0" epoch="0" num_updates="0" have_quorum="false" last_written="Mon Oct 10 20:04:01 2005">
>   <configuration>
>     <crm_config>
>       <nvpair id="transition_idle_timeout" name="transition_idle_timeout" value="120s"/>
>       <nvpair id="symmetric_cluster" name="symmetric_cluster" value="true"/>
>       <nvpair id="no_quorum_policy" name="no_quorum_policy" value="stop"/>
>       <nvpair id="suppress_cib_writes" name="suppress_cib_writes" value="true"/>
>       <nvpair id="stonith_enabled" name="stonith_enabled" value="false"/>
>       <nvpair id="default_resource_stickiness" name="default_resource_stickiness" value="INFINITY"/>
>     </crm_config>
>     <nodes>
>       <node uname="node1" type="member"/>
>       <node uname="node2" type="member"/>
>     </nodes>
>     <resources>
>       <group id="IEL">
>         <primitive class="ocf" id="IEL_IPaddr_70" provider="heartbeat" type="IPaddr">
>           <operations>
>             <op id="1" interval="5s" name="monitor" timeout="5s"/>
>           </operations>
>           <instance_attributes>
>             <attributes>
>               <nvpair name="ip" value="<>MailScanner warning: numerical links are often malicious:<>"/>
>               <nvpair name="netmask" value="24"/>
>               <nvpair name="nic" value="eth0"/>
>             </attributes>
>           </instance_attributes>
>         </primitive>
>       </group>
>     </resources>
>     <constraints>
>        <rsc_location id="rsc_location_IEL" rsc="IEL">
>                <rule id="prefered_location_IEL" score="100">
>                        <expression attribute="#uname" operation="eq" value="node1"/>
>                </rule>
>        </rsc_location>
>     </constraints>
>   </configuration>
>   <status/>
> </cib>
>Linux-HA mailing list
>Linux-HA at lists.linux-ha.org
>See also: http://linux-ha.org/ReportingProblems 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux-ha.org/pipermail/linux-ha/attachments/20051012/e475509d/attachment.html>

More information about the Linux-HA mailing list