[
https://issues.apache.org/jira/browse/AMQ-9202?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17683083#comment-17683083
]
ASF subversion and git services commented on AMQ-9202:
------------------------------------------------------
Commit ed924cddac90b96bdc47b215852a68155d818bcd in activemq's branch
refs/heads/main from Christopher L. Shannon (cshannon)
[ https://gitbox.apache.org/repos/asf?p=activemq.git;h=ed924cdda ]
AMQ-9202 - Make sure Reentrant locks are acquired outside a try block
This is best practice and will prevent unlock from being attempted
inside of a finally block when the thread doesn't actually own the
lock which can happen when the lock attempt throws an exception
such as calling lockInterruptibly()
> Reentrant locks should always be locked outside of a try block
> --------------------------------------------------------------
>
> Key: AMQ-9202
> URL: https://issues.apache.org/jira/browse/AMQ-9202
> Project: ActiveMQ
> Issue Type: Bug
> Components: Broker
> Affects Versions: 5.17.3
> Reporter: Christopher L. Shannon
> Assignee: Christopher L. Shannon
> Priority: Major
> Fix For: 5.18.0, 5.17.4
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> I saw an exception happen today that turned out to be improper use of
> Reentrant locks. Acquiring the lock should always be done before the try
> block so that the finally clause to unlock will never execute unless the lock
> is acquired. More info:
> https://stackoverflow.com/questions/31058681/java-locking-structure-best-pattern
> The issue is inside of AbstractRuntimeConfigurationBroker, for example:
> [https://github.com/apache/activemq/blob/d8ce1d9ff0fa2296f3c56b59602e5cddb6ffe4a9/activemq-runtime-config/src/main/java/org/apache/activemq/plugin/AbstractRuntimeConfigurationBroker.java#L81]
> I saw a IllegalMonitorStateException exception occur due to
> lockInterruptibly() throwing an exception so when the finally block tried to
> unlock it threw the IllegalMonitorStateException since it was not owned by
> the thread.
> I went through the rest of the code and most places are correct but a couple
> spots need to be fixed as well.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
