[
https://issues.apache.org/jira/browse/ARTEMIS-1925?focusedWorklogId=345590&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-345590
]
ASF GitHub Bot logged work on ARTEMIS-1925:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 18/Nov/19 21:23
Start Date: 18/Nov/19 21:23
Worklog Time Spent: 10m
Work Description: jbertram commented on issue #2893: ARTEMIS-1925 combine
STRICT and OFF with redistribution
URL: https://github.com/apache/activemq-artemis/pull/2893#issuecomment-555215393
@AntonRoskvist so the "odd" behavior your seeing isn't specifically related
to this PR. It is simply the way the broker deals with auto-created queues in a
cluster at the moment. I'll try to explain...
Each broker has its own set of addresses and queues - even in a cluster.
When a queue is created on one broker in a cluster the broker on which the
queue is created creates what we call a "local" binding. It then informs all
the other nodes in the cluster about this binding at which point the other
nodes create a "remote" binding. Therefore, at this point all the nodes in the
cluster know about the queue. Likewise, when a consumer is created on a queue
on one broker all the other nodes are informed about it. *If* another broker in
the queue has a matching _local_ binding with the same name (i.e. the queue has
actually been created on that broker) then a "redistributor" will be created on
that queue so that messages sent to it can be redistributed if necessary. In
your use-case there is no local binding so the redistributor isn't created.
As I mentioned previously, this issue wasn't caused by this PR. IMO a new
JIRA should be opened and a new PR should be sent to address it. In the
mean-time you can work-around the issue by pre-creating all the queues for
which you need redistribution.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 345590)
Time Spent: 2h (was: 1h 50m)
> Allow message redistribution even with STRICT or OFF message-load-balancing
> semantics
> -------------------------------------------------------------------------------------
>
> Key: ARTEMIS-1925
> URL: https://issues.apache.org/jira/browse/ARTEMIS-1925
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Reporter: Justin Bertram
> Priority: Major
> Time Spent: 2h
> Remaining Estimate: 0h
>
> Currently if the {{message-load-balancing}} is {{STRICT}} or {{OFF}} then
> message redistribution is disabled. Message redistribution should be
> controlled only by the {{redistribtion-delay}}.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
