[rabbitmq-discuss] erlang-client issue, discarding content?

Thu Jul 30 20:52:18 BST 2009

mplivelog1 ~: sudo /usr/sbin/rabbitmqctl list_queues -p myqueue name
consumers messages messages_ready

Listing queues ...
storage 5 1 0
...done.

I currently have it running again after a restart and that's usually the
response I get, usually only 0-5 items in the queue, since it's being
processed.

When the error was occurring and I checked the queue, there were effectively
*zero* items in the queue according to that command I ran above in the
queue. In python I definitely do acknowledge items via:

self.channel.basic_ack(data.delivery_tag)

where self.channel is amqp.Connection(host=host, **info).channel()

Suhail

On Thu, Jul 30, 2009 at 12:46 PM, Matthias Radestock <matthias at lshift.net>wrote:

> Suhail Doshi wrote:
>
>> Looking in my logs shows a large gap in time, in bold is where it is:
>>
>> =INFO REPORT==== 30-Jul-2009::17:26:56 ===
>>    alarm_handler: {set,{system_memory_high_watermark,[]}}
>>
>
> This indicates that rabbit got close to running out of memory and told all
> connected clients to stop sending any more messages. Hence the "Discarding
> ..." message you were seeing in the erlang client. Your consumers should
> continue to receive messages, albeit slowly if rabbit is swapping.
>
> Perhaps your consumers are not acknowledging received messages, causing
> them to keep piling up at the server?
>
> I recommend checking the queue lengths/sizes with 'rabbitmqctl
> list_queues'.
>
>
> Regards,
>
> Matthias.
>

-- 
http://mixpanel.com
Blog: http://blog.mixpanel.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20090730/72b8d8fa/attachment.htm 