[rabbitmq-discuss] HA - queues do not move?

Mon Aug 29 21:31:50 BST 2011

(I'm new to Rabbit, so please forgive me if this is a dumb question)

I followed the instructions at http://www.rabbitmq.com/pacemaker.html
to build a two-node HA environment.  My nodes are RabbitMQ 2.5.1 on
CentOS 6 64-bit, and I'm using a network share for shared storage
instead of DRBD.

The nodes failover just fine, and most things seem to be working as
expected, however, I've noticed that when I take Node 1 down and it
fails over to Node 2, the queues and messages on Node 1 do not move to
Node 2.  Is this expected behavior?  I expected that the point of the
shared storage would be so that messages would not be lost in the
event of a failover, but maybe my expectations are incorrect?

Thanks for any info you can offer.