[rabbitmq-discuss] Confusing disk free space limit warning

[Matthias Radestock]

Mark,

On 18/09/12 03:22, Mark Hingston wrote:
> On 17/09/12 7:41 PM, Matthias Radestock wrote:
>> Ah. That's a bug in the stomp plug-in. It's been around since 2.8.3.
>> Will fix.
>
> Ok, thanks for the explanation. So is my best option to disable the
> stomp plugin until this is fixed?

Yes, if you can live w/o it. Or make sure you don't hit the alarms ;)

>> So your question really comes down to how would you expect a client to
>> detect and deal with a slow server / congested network.
>
> Thanks for the explanation. I guess now I'm trying to figure out what
> the best way is to defend against this situation so that my messages
> don't get lost.

As you noted, publisher confirms would be one way and are definitely the 
way to go when wanting to ensure that no messages are lost, ever, on 
their delivery to rabbit. I don't know about their status in celery.

However, before you go down that route, consider carefully what other 
failure modes there are in your system. It's all very well to ensure 
that the message delivery to rabbit is reliable, but if the connecting 
application can explode and lose messages then that may not actually 
gain you very much.

On the rabbit/connection side, messages will only be lost if the 
connection is dropped or rabbit encounters some catastrophic internal 
error. Alarm conditions are transitory, i.e. you can recover from them 
w/o restarting rabbit, by, say, consuming messages (to free up memory) 
or clearing some disk space.

So depending on how reliable your network and machines are (and leaving 
aside the alarm-handler-killing stomp bug), the most likely place for 
messages to get dropped may not actually be on the rabbit side.

Regards,

Matthias