OK that makes sense, and it's good to know that "If the level dips back below the threshold the timer is reset".
I guess that the scenario I was testing is when a new producer comes online and there's a full queue. I can clearly intercept the clientConnect then do some querying of QMF objects, but I haven't a clue which objects to look for to figure out that any given queue has exceeded a threshold. As we've previously discussed I don't even believe that I can actually find the queue size at runtime if it was using the broker defaults (though I guess that as I control the broker I can pass that to my QMF client as a parameter). Presumably I can find the number of bytes currently on a queue? Is that the "byteDepth" object on the queue object? Gordon Sim wrote: > > > When the threshold is first exceeded and event is always generated. > However subsequent messages enqueued while above the limit will not all > generate events or this could exacerbate an unexpected situation by > loading the broker with all the event traffic. The event will not repeat > for a period controlled by qpid.alert_repeat_gap / > x-qpid-minimum-alert-repeat-gap (by default I think its one minute). > > If the level dips back below the threshold the timer is reset so that on > again reaching the limit an event is always generated. > > -- View this message in context: http://apache-qpid-users.2158936.n2.nabble.com/QMF-Questions-tp6533747p6537098.html Sent from the Apache Qpid users mailing list archive at Nabble.com. --------------------------------------------------------------------- Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:[email protected]
