[rabbitmq-discuss] rabbitmq-server crashes hard while consuming 31GB of RAM
mh at foldr3.com
Wed Dec 21 14:36:19 GMT 2011
On 12/21/2011 02:05 PM, Alexis Richardson wrote:
> adding Ask to cc..
Thank you very much. I have identified the piece of code that was
causing the problem at hand.
This does *not* appear to be a celery issue given what I know at this
> On Wed, Dec 21, 2011 at 12:42 PM, Muharem Hrnjadovic <mh at foldr3.com> wrote:
>> On 12/21/2011 12:49 PM, Alexis Richardson wrote:
>>> On Wed, Dec 21, 2011 at 11:44 AM, Matthias Radestock
>>> <matthias at rabbitmq.com> wrote:
>>>> I was wrong. You have more than 47,000 connections! And the same number of
>>>> channels (unsurprisingly). But you "only" have about 16,000 queues.
>>>> Congratulations. This is by far the fattest rabbit I've ever seen (except
>>>> those stuffed full of messages).
>>> I've seen Rabbits with 100,000s of queues per broker, in social
>>> network / social gaming cases...
>>> There's also the 'cloud management' case but they tend to have more
>>> brokers and fewer queues / broker.
>> I believe we have a fault in the task function that is run by celery
>> (i.e. in our code) with the effect that a *new* rabbitmq connection is
>> opened upon every invocation while the already utilised ones cannot be
>> "garbage collected" for some reason.
>> We should never have that many open connections. Looking into this..
>> Best regards/Mit freundlichen Grüßen
>> Muharem Hrnjadovic <mh at foldr3.com>
>> Public key id : B2BBFCFC
>> Key fingerprint : A5A3 CC67 2B87 D641 103F 5602 219F 6B60 B2BB FCFC
Best regards/Mit freundlichen Grüßen
Muharem Hrnjadovic <mh at foldr3.com>
Public key id : B2BBFCFC
Key fingerprint : A5A3 CC67 2B87 D641 103F 5602 219F 6B60 B2BB FCFC
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 900 bytes
Desc: OpenPGP digital signature
More information about the rabbitmq-discuss