Possibly you could just create consumers where effectively (not real code): sendTime < now() - 10min
and just create a new consumer, receive until there's nothing left, then try again a minute later. you'd be able to yank from one queue and re-publish to another. I'd bet you could do this with camel and not even have to write code really. just need a properly formed "expiration selector", and pipe the result to the new queue. On 8/30/07, ttmdev <[EMAIL PROTECTED]> wrote: > > > If all else fails, you can perhaps use a QueueBrowser to monitor the > messages > on the queue. > > ttmdev > > > DominicTulley wrote: > > > > That was my first expectation - that they would go to the DLQ but they > > never appeared to go there. Perhaps the bug you mention is the reason. > > Having found this behaviour I went and did some searching - the outcome > > was that it sounded like brokers were expected to destroy the expired > > messages. > > > > I'll have a look at AMQ-1112. Perhaps it's been fixed and I can get the > > latest snapshot and try again. > > > > Thanks, > > > > -Dominic > > > > Mario Siegenthaler-2 wrote: > >> > >> AFAIK the expired message will be moved into the DeadLetterQueue. > >> Since you're able to configure the DLQ it should be possible to do > >> what you want (using AMQ5). > >> There's an open issue with expired message not being processed until > >> someone tries to consume them (The bug was AMQ-1112 I think). So since > >> your using selectors on the queue I'm not sure whether this applies or > >> not. I'd just give it a try. > >> > >> Mario > >> > >> On 8/30/07, DominicTulley <[EMAIL PROTECTED]> wrote: > >>> > >>> Is there a way, either with filters or timeouts on queues, to get my > >>> hands on > >>> any messages that have been sitting in the queue for "too long"? > >>> > >>> My situtation/plan is to have a queue with multiple consumers. Each > >>> consumer will be using a selector to decide which messages to consume. > >>> Under ideal circumstances, all messages will have a consumer that > wants > >>> them, but situations can arise where a message doesn't match any > >>> selector. > >>> I would like to have some mechanism whereby I can spot that the > message > >>> is > >>> at the "front" of the queue but after n seconds still hasn't been > >>> consumed. > >>> This allows me to arrange for a consumer to start dealing with that > >>> particular message type. > >>> > >>> I looked at the JSMExpiration header which sounded about right, except > >>> that > >>> on expiry the outcome is to destroy the message. I would love to be > >>> able to > >>> ask the expiry processing to simply forward the message to another > >>> queue, > >>> for instance. > >>> > >>> I appreciate any suggestions. > >>> > >>> Thanks, > >>> > >>> -Dominic > >>> -- > >>> View this message in context: > >>> > http://www.nabble.com/How-can-I-find-messages-that-have-been-unconsumed-for-%22too-long%22--tf4353227s2354.html#a12404218 > >>> Sent from the ActiveMQ - User mailing list archive at Nabble.com. > >>> > >>> > >> > >> > > > > > > -- > View this message in context: > http://www.nabble.com/How-can-I-find-messages-that-have-been-unconsumed-for-%22too-long%22--tf4353227s2354.html#a12409934 > Sent from the ActiveMQ - User mailing list archive at Nabble.com. > >
