[ https://issues.apache.org/jira/browse/ARTEMIS-4206?focusedWorklogId=853663&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-853663 ]
ASF GitHub Bot logged work on ARTEMIS-4206: ------------------------------------------- Author: ASF GitHub Bot Created on: 29/Mar/23 11:24 Start Date: 29/Mar/23 11:24 Worklog Time Spent: 10m Work Description: gemmellr commented on code in PR #4418: URL: https://github.com/apache/activemq-artemis/pull/4418#discussion_r1151687346 ########## tests/soak-tests/pom.xml: ########## @@ -260,7 +260,7 @@ <configuration>${basedir}/target/classes/servers/interruptlm</configuration> <args> <arg>--java-options</arg> - <arg>-Djava.rmi.server.hostname=localhost</arg> + <arg>-Djava.rmi.server.hostname=localhost -DARTEMIS_DEBUG_REF=true</arg> Review Comment: Again, seems odd to have the soak tests default to running things in a typically non-representative manner. Perhaps more so in this case than the leak-tests, given none of these tests are actually even looking at/using the effect of the config. Perhaps just add some mechanism to do this easily on request, e.g a property that defaults empty so you can ask the tests to do this without modifying them? ########## artemis-core-client/src/main/java/org/apache/activemq/artemis/api/core/RefCountMessage.java: ########## @@ -16,12 +16,56 @@ */ package org.apache.activemq.artemis.api.core; +import java.io.PrintWriter; +import java.io.StringWriter; +import java.lang.invoke.MethodHandles; +import java.lang.ref.Cleaner; +import java.util.ArrayList; import java.util.HashMap; import java.util.concurrent.atomic.AtomicIntegerFieldUpdater; -// import org.apache.activemq.artemis.utils.collections.ConcurrentHashSet; Issue Time Tracking ------------------- Worklog Id: (was: 853663) Time Spent: 50m (was: 40m) > Unreferenced AMQP Large Messages are not removed > ------------------------------------------------ > > Key: ARTEMIS-4206 > URL: https://issues.apache.org/jira/browse/ARTEMIS-4206 > Project: ActiveMQ Artemis > Issue Type: Bug > Reporter: Clebert Suconic > Assignee: Clebert Suconic > Priority: Major > Fix For: 2.29.0 > > Time Spent: 50m > Remaining Estimate: 0h > > Say you crashed the server after the ack, and before the file.remove, and the > journal. record removal. > The AMQP Large Message may not be removed right away, requiring a restart of > the broker. > At this point this is really caused by ARTEMIS-4193 and only affected 2.29.0 > and no previous versions -- This message was sent by Atlassian Jira (v8.20.10#820010)