[rabbitmq-discuss] RabbitMQ writes non stop to disk

Matthias Radestock matthias at rabbitmq.com
Thu Oct 27 00:00:23 BST 2011


Raphael,

On 25/10/11 17:22, Raphael Simon wrote:
> Something I didn't mention in my first email is that we have about 8
> brokers running in production and only one is showing these symptoms,
> the throughput is about the same through all brokers.

And there is nothing different to all the other brokers? In particular, 
do the other brokers also have two slow-moving durable queues and other, 
faster-moving queues?

Upgrading to 2.6.1 and getting us the output of 'rabbitmqctl report' 
from all brokers might allow us to spot any abnormalities. Failing that, 
the output of 'rabbitmqctl list_queues name durable arguments pid 
messages_ready messages_unacknowledged consumers memory 
backing_queue_status' would help.

Btw, what is it that you are concerned about? That one of your rabbits 
is doing a lot of disk i/o?

Regards,

Matthias.


More information about the rabbitmq-discuss mailing list