Clustered queue management#
Get answers to common technical questions about clustered queue management for Rackspace Private Cloud Powered By OpenStack (RPCO) v12.
How does RabbitMQ recover after a single node failure?
If the node goes offline for some time and then returns, it will automatically rejoin the cluster with no user intervention required.
How does a cluster recover from a catastrophic failure of all nodes?
As with an upgrade, the nodes should be brought back in reverse order. However, in a case where nodes go down simultaneously, you must manually start all nodes in quick succession to ensure that they can identify each other, as described in this discussion thread.