[rabbitmq-discuss] Can someone please confirm the clustered durable queue behavior?

[Matthias Radestock]

Eugene,

On 29/12/12 03:21, Eugene Vasilchenko wrote:
> So if I understand it correctly, once the original master is back
> online as a slave, its old messages will get consumed, meaning it's
> possible that we will get some old, previously consumed messages
> again. (Since we already consumed them earlier from the new master?)

No, as I said
<quote>
The old, rejoined master becomes a slave, initially *empty* and
unsynchronised.
</quote>

Matthias.