[rabbitmq-discuss] RabbitMQ Roadmap

Simon MacMullen simon at rabbitmq.com
Thu May 24 11:37:38 BST 2012


On 24/05/12 11:33, Brendan Hay wrote:
> Good stuff!
>
> The physical layout comment is regarding having to specify -sname/-name
> in the x-ha-policy .. ie having clients aware of the nodes that compose
> a cluster.
> In the imaginary places in my head it would be nicer to set something
> along the lines of 'one slave', 'two slaves', ..., or '50% of the
> cluster has slaves' etc.

Oh, of course. Yes, I want that to happen too. Basically clients should 
not have to know anything about how the server / cluster is configured.

> Migrating/moving a queue around inside a cluster is another nice to have
> - scenario would be whatever you're using to do load-balancing
> throws a nana and causes too many queues to be declared on a single
> machine vs a more desired even spread. Basically fine tuning your queue
> distribution, on the fly.

For mirrored queues that's conceptually quite easy; add some mirrors, 
wait for them to catch up, remove the old ones and you're done.

For non-mirrored queues this would be fairly horrible... unless you're 
happy for the queue to pause while data is migrated.

Cheers, Simon

-- 
Simon MacMullen
RabbitMQ, VMware


More information about the rabbitmq-discuss mailing list