[rabbitmq-discuss] Statistics gathering

Wed Jul 24 18:17:32 BST 2013

Thanks for the reply Simon! What I actually need is to collect samples of
statistics, or maybe just set a huge statistics interval...have you ever
tried to set statistics interval to something like a week? Does it affect
performance?
otherwise I saw some connector for collectd and graphite but they seems to
be very experimental and not so used..
Any suggestion..?
On Wed, Jul 24, 2013 at 7:09 PM, Simon MacMullen <simon at rabbitmq.com> wrote:

> On 24/07/13 15:14, Flavio Pompermaier wrote:
>
>> I saw that at
>> http://hg.rabbitmq.com/**rabbitmq-management/raw-file/**
>> rabbitmq_v3_1_3/priv/www/api/**index.html<http://hg.rabbitmq.com/rabbitmq-management/raw-file/rabbitmq_v3_1_3/priv/www/api/index.html>
>> you can access those statistcs via json, the problem is that, from waht
>> I saw,
>> those statistics are run-time statistics,
>>
>
> Not sure what you mean by "run-time statistics". I would expect all
> statistics to be based on things that happened at run-time :-)
>
>  I cannot get info like max queue size
>>
>
> Not directly. You can retrieve a set of samples of queue size over time,
> then scan through them looking for the maximum.
>
> Obviously the samples are... samples, so the queue may have been longer at
> some point between samples.
>
>  or average rate of message processing (from a certain point
>> in time on), am I wrong?
>>
>
> The average rate over the time period you have requested is delivered as
> avg_rate.
>
> Cheers, Simon
>
> --
> Simon MacMullen
> RabbitMQ, Pivotal
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130724/173019cb/attachment.htm>