On 11/22/2013 11:00 AM, Paul Colby wrote:
I wonder: would it be possible / practical to monitor the journal for the oldest message? If we can detect the problem occurring, we can respond to it before the cluster is broken. Something to think about anyway.
One relevant change that is coming up is the so called 'linearstore', which is a replacement for the current linux based store that automatically grows the journal for queues as needed. Not sure what the current status on that is (Kim?), but it shouldn't be too long before its ready to try out (I hope!).
That said it does seem like a generally useful thing to be able to detect an old message getting stuck on the queue. I'll have a think about that. And of course if anyone else has ideas about how to do that by all means share them. The java broker I believe has a x-qpid-maximum-message-age property that when reached triggers an alert which could perhaps be copied.
--------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
