This is an automated email from the ASF dual-hosted git repository.
jbertram pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/activemq-website.git
The following commit(s) were added to refs/heads/main by this push:
new b42378cca Add info about removeAllMessageGroups for manual rebalancing
groups
b42378cca is described below
commit b42378cca8e2ae0dc9d08c62769358997e4a5685
Author: Justin Bertram <jbert...@apache.org>
AuthorDate: Thu Jul 20 10:12:42 2023 -0500
Add info about removeAllMessageGroups for manual rebalancing groups
---
src/message-groups.md | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/message-groups.md b/src/message-groups.md
index 19468c490..e61ba0888 100644
--- a/src/message-groups.md
+++ b/src/message-groups.md
@@ -129,6 +129,8 @@ The following code snippet shows how to wait for two
consumers (or two seconds)
```
As [the appropriate test
case](https://github.com/apache/activemq/blob/master/activemq-unit-tests/src/test/java/org/apache/activemq/usecases/MessageGroupDelayedTest.java)
shows, adding a small time pause before dispatching or setting a minimum
consumer number, ensures equal message group distribution.
+If you need to rebalance the message groups manually for any reason you can do
so by executing the `removeAllMessageGroups` operation on the JMX MBean of the
corresponding queue.
+
### Competing demands of memory consumption, load balancing, complexity, etc.
The default behavior called `CachedMessageGroupMap` is limited to 1024 message
groups in an LRU cache may not match you expectation w.r.t message order.
`CachedMessageGroupMap` has bounded memory use, but only keeps track of up to
1024 (or the maximum configured size) groups, then loses track of any groups
older than the newest 1024. In this way, if there are more groups than the
maximum, **ordering will be lost for the oldest groups**.
