Hi Florian
>
> "corosync-cfgtool -s" is identical on all nodes?
Yes, of course node ID are different and the id correspond to the IP of
the local NIC.
> "corosync-objctl | grep member" produces 5 members on all nodes?
Same everywhere, node ID are consistent.
>
> Which rrp_mode are you running on?
none, just one ring because of network issues.
That's the crm status on one node:
[root@db-stat ~]# crm status
============
Last updated: Wed Jun 6 18:48:06 2012
Last change: Wed Jun 6 13:58:44 2012
Current DC: NONE
0 Nodes configured, unknown expected votes
0 Resources configured.
============
and crm status one another node
Node arbiter: standby
Online: [ db-stat-slave db-stat db db-slave ]
prod_mysql_writer (ocf::heartbeat:IPaddr2): Started db-slave
prod_mysql_reader (ocf::heartbeat:IPaddr2): Started db
Master/Slave Set: prod_mysql [p_prod_mysql]
Masters: [ db-slave ]
Stopped: [ p_prod_mysql:1 ]
Master/Slave Set: stat_mysql [p_stat_mysql]
Masters: [ db-stat-slave ]
Slaves: [ db-stat ]
stat_mysql_reader (ocf::heartbeat:IPaddr2): Started db-stat
stat_mysql_writer (ocf::heartbeat:IPaddr2): Started
db-stat-slave
p_pt_heartbeat_stat (ocf::heartbeat:anything): Started
db-stat-slave
p_pt_heartbeat (ocf::heartbeat:anything): Started db-slave
Note that db-stat, although with no crm status output, is active in the
cluster and reacts correctly to node standby/online.
Regards,
Yves
>
> Cheers,
> Florian
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
