[
https://issues.apache.org/jira/browse/ARTEMIS-4362?focusedWorklogId=870748&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-870748
]
ASF GitHub Bot logged work on ARTEMIS-4362:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 13/Jul/23 09:54
Start Date: 13/Jul/23 09:54
Worklog Time Spent: 10m
Work Description: gtully commented on code in PR #4552:
URL: https://github.com/apache/activemq-artemis/pull/4552#discussion_r1262321833
##########
artemis-server/src/main/java/org/apache/activemq/artemis/core/server/impl/QueueImpl.java:
##########
@@ -179,6 +179,14 @@ public class QueueImpl extends CriticalComponentImpl
implements Queue {
private volatile boolean queueDestroyed = false;
+ // Variable to control if we should print a flow controlled message or not.
+ // Once it was flow controlled, we will stop warning until it's cleared
once again
+ private volatile boolean pageFlowControlled = false;
+
+ private volatile long pageFlowControlledLastLog = 0;
+
+ private static final long PAGE_FLOW_CONTROL_PRINT_INTERVAL = 60_000L;
Review Comment:
I think this will need to be configurable.... possibly combining with what
we do for flow control of a producer, when an address is blocked, and a new
producer tries to send... it is analogous to a consumer on a blocked queue.
Both are reaching configured limits that we expect to be met in normal
operation, however for a new producer or consumer - the limits will influence
expected behaviour, they cannot send or consume... and may revert to the logs
to see why.
we could have flowControllLoggingPeriod - that can be tweaked to configure
the interval and disable. it could be applicable to both producers and
consumers.
Periodic will work, b/c there will be some correlation between new consumer
being added and the period. Is it true that will always attempt a dispatch on
adding a new consumer?
Issue Time Tracking
-------------------
Worklog Id: (was: 870748)
Time Spent: 1h (was: 50m)
> Produce a logger when consumers are holding too many messages not allowing
> deposing
> -----------------------------------------------------------------------------------
>
> Key: ARTEMIS-4362
> URL: https://issues.apache.org/jira/browse/ARTEMIS-4362
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Reporter: Clebert Suconic
> Priority: Major
> Time Spent: 1h
> Remaining Estimate: 0h
>
> We should add a log.warn whenever the system can't read more messages because
> consumers are holding too many messages in pending state.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
