[Linux-HA] Re: heartbeat 2.0.8: can no longer start it

[greno at verizon.net]

Ok, not so broken.  Just needed to set user:group on new cib.xml.  jeesh...


>From: greno at verizon.net
>Date: 2007/02/05 Mon AM 11:26:19 CST
>To: linux-ha at lists.linux-ha.org
>Subject: heartbeat 2.0.8: can no longer start it

>Ok, I've officially broken it.  Heartbeat no longer runs right.  I've removed all files except cib.xml from the crm directory but still no luck.  Here is the error:
>
>Feb  5 12:13:20 server1 cib: [4115]: info: readCibXmlFile: Reading cluster configuration from: /var/lib/heartbeat/crm/cib.xml
>Feb  5 12:13:20 server1 heartbeat: [4087]: WARN: Exiting /usr/lib/heartbeat/cib process 4115 killed by signal 11.
>Feb  5 12:13:20 server1 heartbeat: [4087]: ERROR: Exiting /usr/lib/heartbeat/cib process 4115 dumped core
>Feb  5 12:13:20 server1 heartbeat: [4087]: ERROR: Respawning client "/usr/lib/heartbeat/cib":
>Feb  5 12:13:20 server1 heartbeat: [4087]: info: Starting child client "/usr/lib/heartbeat/cib" (100,101)
>Feb  5 12:13:21 server1 heartbeat: [4116]: info: Starting "/usr/lib/heartbeat/cib" as uid 100  gid 101 (pid 4116)
>Feb  5 12:13:21 server1 cib: [4116]: ERROR: crm_is_writable: /var/lib/heartbeat/crm/cib.xml must be owned and r/w by user hacluster
>Feb  5 12:13:21 server1 cib: [4116]: WARN: crm_is_writable: /var/lib/heartbeat/crm/cib.xml should be owned and r/w by group haclient
>
>
>cib.xml:  created new file but no help.
>
> <cib have_quorum="true">
>   <configuration>
>     <crm_config/>
>     <nodes>
>     </nodes>
>     <resources>
>     </resources>
>     <constraints>
>     </constraints>
>   </configuration>
> </cib>
>
>ha.cf:
>
>logfacility     daemon
>keepalive 1
>deadtime 10
>warntime 5
>initdead 20   
>udpport 694   
>ping 192.168.1.1 
>bcast eth0 eth1  
>auto_failback off 
>                  
>respawn root /usr/lib/heartbeat/pingd -m 100 -d 5s -a pingd_score
>apiauth cibmon   uid=hacluster
>respawn hacluster /usr/lib/heartbeat/cibmon -d
>
>node    server1  # cluster member
>node    server2  # cluster member
>use_logd yes
>compression     bz2
>compression_threshold 2
>crm yes