hate bumping, but bump... Also, this might be related: https://issues.apache.org/activemq/browse/AMQ-2016
If anyone has any input, it would be greatly appreciated. Thanks. skomarla wrote: > > http://www.nabble.com/file/p20627785/msg-group-test.zip > msg-group-test.zip > > I have posted my comments in > https://issues.apache.org/activemq/browse/AMQ-1126 since I originally > thought it might be related to the RA, but I'm not sure if it is localized > to the RA. > > Attached is a unit test and some sample code.. It has a junit test with > two tests in it. > Both tests start out with sending some message to a same queue split > across three different group ids (20 to group A, 10 to group B, 5 to group > C). My understanding of message groups leads me to believe that the > messages within a group are delivered sequentially, but across the groups > they are delivered in parallel. > > If you look at the output of the unit tests, it does not seem to be the > case. in many cases, it seems the messages get delivered in FIFO order > ignoring groups entirely... in the listeners, I have setup various sleeps > to simulate slow consumers.. When the message is from group C, the sleep > time is small, and thus all 5 messages from group C should be delivered > before the second message from group is delivered.. it does not seem to be > the case > > 1 test uses the connection factory directly to setup message consumers, > and another uses spring's listener container. Neither seem to do what I > expect. Maybe I just have something setup horribly incorrectly, but I've > seen the exact same bahaviour using the RA. > > Sorry for the double post, but I figured i'd post in the user forum as > well.. soem log output is in the JIRA issue.. > -- View this message in context: http://www.nabble.com/message-grouping-not-working--tp20627785p20803736.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
