[rabbitmq-discuss] durable queue is down or inaccessible

Tue Nov 12 17:20:20 GMT 2013

On 10/11/13 07:57, carlhoerberg wrote:
> With mirror all queues, but often when nodes crashes we see:
>
> =ERROR REPORT==== 10-Nov-2013::04:40:14 ===
> connection <0.26611.6>, channel 2 - soft error:
> {amqp_error,not_found,
>              "home node 'rabbit at wombat-48510-01' of durable queue
> 'images.processed' in vhost 'nwdwgqag' is down or inaccessible",
>              'queue.declare'}
>
> in the logs. Why doesn't the queues failover properly?

Make sure you are running the latest version of RabbitMQ.

Make 100% sure that the aforementioned queue is in fact mirrored - 
check the policy in management / ctl (when the queue is alive, obviously).

Matthias.