Hi, One of Artemis broker clients is facing a strange issue where consumer application with 5 concurrent threads stop pulling messages from the queue after limited successful message processing. When the consumer application first starts, it receives the messages over 5 threads but after sometime the receiving stops even though the connection/session/consumers are active on the broker and messages are also available on given queue.
They also have a durable topic subscriber. Both queue consumers and topic subscriber are sharing same connection and both of them using message selectors as well with prefetch of 1. Client application restart is also not helping in resuming the queue consumption. In the meanwhile the topic subscriber which also uses same connection as queue consumer works fine. If we restart the broker node, queue consumer resumes again successfully and then stops after some time of successful operations. They are connecting to broker using OpenWire over Apache NMS. What could be possibly wrong and how can we debug further? Any recommendation on connection sharing? Thanks Shiv
