Hi,
I am working on the mysql RA and I am hitting a strange problem
again. I need to store a cluster wide attribute like this:
crm_attribute --type crm_config --name replication_info -s
mysql_replication -v "10.2.2.160|mysql-bin.000140|18276"
Would this cause a pengine run. Since I added this piece of code I am
having this strange behavior after a master switch, here testvirtbox3
was the previous master.
Node testvirtbox3: standby
Online: [ testvirtbox1 testvirtbox2 ]
reader_vip_2 (ocf::heartbeat:IPaddr2): Started testvirtbox2
reader_vip_3 (ocf::heartbeat:IPaddr2): Started testvirtbox2
writer_vip (ocf::heartbeat:IPaddr2): Started testvirtbox1
reader_vip_1 (ocf::heartbeat:IPaddr2): Started testvirtbox2
Master/Slave Set: ms_MySQL
Masters: [ testvirtbox2 ]
Slaves: [ testvirtbox1 ]
Stopped: [ p_mysql:0 ]
testvirtbox2 is the master but testvirtbox1 has the writer_vip. My
config explicitely has those rules concerning writer_vip. I have to use
cluster attributes since the old implementation using node attributes
easily fails with non-trivial fail over case.
colocation writer_vip_coloc_master inf: ms_MySQL:Master writer_vip
order order_writer_vip_master inf: ms_MySQL:promote writer_vip:start
Regards,
Yves
_______________________________________________
Pacemaker mailing list: Pacemaker@oss.clusterlabs.org
http://oss.clusterlabs.org/mailman/listinfo/pacemaker
Project Home: http://www.clusterlabs.org
Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
Bugs: http://bugs.clusterlabs.org
