[rabbitmq-discuss] Monitoring Message Throughput

[Gavin M. Roy]

Hey Matthias,

On Sat, Feb 6, 2010 at 4:17 PM, Matthias Radestock <matthias at lshift.net>wrote:

> By "throughput" do you mean:
> - rate at which messages are enqueued?
>

For me, yes


> - rate at which messages are dequeued?
>

For me, yes

If it's "messages enqueued", what happens to messages that are part of a
> transaction? Are they included in the figure immediately or only when
> the tx has committed?
>

tx committed


>
> If it's "messages dequeued", what happens to message redeliveries (e.g.
> when a message has been delivered but then the client disconnected w/o
> acking them and the message is subsequently redelivered to another
> client)? Are they included in the figure?
>

Only acked messages, though redeliveries would be a good counter to keep as
well.

>
> Also,
> - are you interested in message counts, i.e. messages per second, or
> data volumes, i.e. octets per second?
>

messages per second, though I could see benefit in data volume as well.


> - over what period of time should the figures be averaged?
>

I'd not have Rabbit provide the averaging and aggregation, just straight out
counters.

- do the stats of a durable queue need to survive a broker restart?
>

Yes


> - do the stats still need to be accessible after a queue has been deleted?
>

No


> - what stats other than per-queue throughputs would you like to see?
>

Perhaps per exchange?

Thanks,

Gavin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20100206/5f651873/attachment.htm