Hi, let me first describe my use-case as it is probably a little bit different from the usual JMS use case:
We have a single process which fills a JMS queue from a pool of available items (say we have 50 items total). The we have n consumers (lets say 2) which read the items from the queue in a synchronous way (i.e. Spring JmsTemplate#receive). This goes like so: - consumer A fetches 5 items - consumer B fetches 20 items - consumer A fetches 5 items and so on. The problem we have now is that we are not able to fetch all items from the queue - e.g. if there are still 10 items left in the queue and I want to get 10, I only get 5 or so. I've already found the prefetchSize configuration and set it to 0 (also tried 1), but still the same problem. The problem seems to be that each consumer has an internal 'dispatched queue' which already has some entries and consumer A has no possibility to access the dispatched queue of consumer B. For example the web console looks like this: Consumer A; Enqueues 10; Dequeues 5; Dispatched 10; Dispatched Queue 5; Prefetch 0; Max Pending 0 Consumer B; Enqueues 20; Dequeues 10; Dispatched 20; Dispatched Queue 10; Prefetch 0; Max Pending 0 In this case there would be e.g. still 15 items total in the queue, but with consumer A I would onyl be able to get 5 items and with consumer B only 10 items. I've tried various configuration options which I found in the ActiveMQ docu (dispatchAsync, alwaysSessionAsync, sendAsync, ...), but none seemed to help. Can anyone help? Does anyone know how to configure/disable this dispatched queue? thanks Christoph
