On 29 June 2016 at 23:08, Gordon Sim <g...@redhat.com> wrote: > On 29/06/16 21:50, Robbie Gemmell wrote: >> >> On 29 June 2016 at 19:43, Gordon Sim <g...@redhat.com> wrote: >>> >>> On 29/06/16 12:39, Robbie Gemmell wrote: >>>> >>>> >>>> On 29 June 2016 at 11:08, Dale Green <green.dale1...@gmail.com> wrote: >>> >>> >>>>> >>>>> >>>>> Closing the consumer explicitly didn't help (it's closed on session >>>>> close >>>>> anyway). >>>> >>>> >>>> >>>> I thought that could be the case, but just wanted you to check in case >>>> the server did something different with the explicit protocol link >>>> closure (consumers/producers are not explicitly closed at the protocol >>>> level by the client when the session is closed, just implicitly). >>> >>> >>> >>> Does the client not explicitly release any prefetched messages unseen by >>> the >>> application when the consumer is closed? >>> >> >> When the consumer itself is closed yes I believe it does (or rather >> proton, I added a toggle to allow influencing the state of the >> dispositions it was implicitly sending). > > > If the client is sending back a modified or released outcome, and the server > isn't unlocking the message on receiving them, that would to me suggest a > server bug. >
If that is happening then yes I agree it would. However given that the session [or additionally the consumer as above] is being closed explicitly, then I dont see scope for it to keep them locked regardless what the client does. --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@qpid.apache.org For additional commands, e-mail: users-h...@qpid.apache.org
