Hi,

We are currently performing a setup using hornetq as a remote message queue hosted in AWS. There is only one queue end point used and the consumers are local to the region on another host.

To achieve HA and DR we are looking to run the message queues in a cluster, and curious if anyone else has performed this type of setup. Reading around I have noticed some setups that include a HA , however most of these appear to reference what is similar to a single region setup

Region A will be the primary site (With 2 queue servers, one in each availability zone)

Region B will be the secondary dr site (With 2 queue servers, one in each availability zone)

From the documentation I have been reading all 4 servers would need to be in the same cluster, some servers are configured to be live servers and other servers being configured to be backup servers.

Some of the challenges in the setup are

• Region B is designed to be a DR site and not process requests unless Region A is unable to serve requests
• Messages should be replicated to Region B from Region A in the event that Region A is lost, minimal message loss will occur
• The consumers at Region A should not connect to Region B and vice versa, The consumers at Region B should not connect to Region A

So far the proposed solution would be both message queue servers in Region A in the cluster in a live role, the message queue servers in Region B configured to be a backup server in the cluster

The question would be has anyone done a setup similar to this or have advice on how to approach the desired outcome ?

Thanks,

Steve