Do you just see some out-of-order due to redelivery, or do you see
continued out-of-order messages? You can detect redelivered messages by
the JMSRedelivered message property (need to check the exact
naming/spelling).
Art
On Tue, Mar 4, 2025 at 9:51 AM Justin Bertram wrote:
> First off, which
false, droppable = false, jmsXGroupFirstForConsumer = false, text =
>
> 2024-01-25 12:31:16,094 | DEBUG | queue://T24.WAREHOUSEINVENTORY.IN,
> subscriptions=1, memory=0%, size=1, pending=0 toPageIn: 0, force:false,
> Inflight: 1, pagedInMessages.size 1, pagedInPendingDispatch.size 0,
&g
Hello Steve. Your report is a little surprising because a 1-3 seconds
pause is huge. I'm curious under what conditions you see that latency - is
it only when running at full load?
Things that can be done to diagnose this down further:
1. View the broker's average enqueue time for messages in
Perhaps it helps to understand that Topics never store messages themselves
- topic subscriptions store messages. This includes non-persistent
messages stored in memory.
So if there is a send to a topic that has no subscriptions, whether durable
or non-durable, the message ends up discarded.
Art
It's likely the cause of the problem is timing of the matching on the
subscription pattern - does the second topic exist when the subscription is
created?
I know how this operates for JMS, and composites work that way - when the
subscription to a composite is created, then the list of actual
desti
Creating a consumer only to consume 1 message is not ideal - there's a lot
of overhead and work on the broker side when consumers are created.
With that said, since the consumer should be getting closed properly, that
should not cause a leak.
So first, I would prioritize the version update. Seco
Reading the description of the problem, it sounds a lot like a consumer
leak to me. Specifically this part:
Consumers are polling queues on regular basis, some once a second
Is the client using JMS?
I recommend looking for signs of consumer leaks.
Art
On Wed, Aug 9, 2023 at 11:10 AM Clebert
Is the consumer connection factory configured to retry connections after a
failure? If so, in the test, is there some additional step that prevents
the client from reconnecting?
For most applications, I recommend that consumers try indefinitely without
interrupting the application. Any logic pla
Can you upload the full stack dump from the client application when this
happens as a GIST on github.com and send the link? It's OK to filter out
any proprietary bits as long as all of the threads that have ActiveMQ in
the call history are listed.
BTW, the "oneWay" method sends a message over the
-release.html
--
Arthur Naseef
10 matches
Mail list logo
