Github user clebertsuconic commented on a diff in the pull request:
https://github.com/apache/activemq-artemis/pull/960#discussion_r96241694
--- Diff:
artemis-server/src/main/java/org/apache/activemq/artemis/core/server/impl/QueueImpl.java
---
@@ -3159,6 +3162,9 @@ public void run() {
connection.killMessage(server.getNodeID());
remotingService.removeConnection(connection.getID());
connection.fail(ActiveMQMessageBundle.BUNDLE.connectionsClosedByManagement(connection.getRemoteAddress()));
+ //break once a consumer gets killed. This can
prevent all
+ //consumers to this queue get killed all at once.
+ break;
--- End diff --
This fix was using the definition of the JIRA and literally fixing the
test, not the overal issue.
The proper fix would be to do some calculation with threshold using
consumerSet.size();
I have a small change on QueueIMpl that I think would fix it properly,
including both Kill and Notify.
```java
Set<Consumer> consumersSet = getConsumers();
if (consumersSet.size() == 0 || queueRate < (threshold *
consumersSet.size())) {
if (logger.isDebugEnabled()) {
logger.debug("Insufficient messages received on queue \"" +
getName() + "\" to satisfy slow-consumer-threshold. Skipping inspection of
consumer.");
}
return;
}
for (Consumer consumer : consumersSet) {
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
