Is it always the same 36 messages that aren't consumed? Also, do your 60 consumers use any selectors? On Apr 17, 2015 9:26 AM, "mattmadhavan" <mattmadha...@yahoo.com> wrote:
> Hello, > I have been stuck in the problem for a while. I have large volumes of > request messages that are produced by a 'server' app. The requests are > nothing but start Oracle DB rownum and end rownum depending on the > 'request > page size' can be 500 or 1000. So a massage is an ImmutablePair<1, 499> or > ImmutablePair<1, 999> depending on the 'request page size'. > > If the 'request page size' is 500 I have 65182 request messages to be > consumed and if my 'request page size' is 1000 then I have 32592 'request' > messages to be consumed. > > It does not matter how many large volume of messages I have. The broker > stops sending messages with exactly 36 messages to go > (failover:(tcp://127.0.0.1:60000)?*jms.prefetchPolicy.all=1*) or the > consumers stop polling with 36 messages more to poll for > failover:(tcp://127.0.0.1:60000)?*jms.prefetchPolicy.all=0* > > I have been stuck on this issue for three weeks now. Any ideas please? > please refer to the attached screen shots. > > I appreciate in advance! > > Thanks > Matt'M queues-1.png > <http://activemq.2283324.n4.nabble.com/file/n4695147/queues-1.png> > > queues2.png > <http://activemq.2283324.n4.nabble.com/file/n4695147/queues2.png> > > > > > > -- > View this message in context: > http://activemq.2283324.n4.nabble.com/AMQ-stops-sending-messages-with-36-messages-to-go-consistently-on-large-volume-of-messages-tp4695147.html > Sent from the ActiveMQ - User mailing list archive at Nabble.com. >
