[jira] [Commented] (AMQ-9820) Closed connections leaked in the connection pool

["ASF subversion and git services (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/AMQ-9820?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18050985#comment-18050985 ] 

ASF subversion and git services commented on AMQ-9820:
------------------------------------------------------

Commit 9eb84c56c683576acb94b6b0ca35ad2b147cca53 in activemq's branch refs/heads/main from JB Onofré
[ https://gitbox.apache.org/repos/asf?p=activemq.git;h=9eb84c56c6 ]

Use of GitHub Actions, and enable dependabot (#1497)

* Use of GitHub Actions, and enable dependabot
Keep a backup of Jenkinsfile for now

* Include push/PR on activemq-6.1.x and activemq-5.19.x branches

* Add new runners for the build

* Set Java distribution

* Remove surefire rerun options as deprecated

* Removing JDK 24 for now as the tests will fail due to the SecurityManager

* Remove JDK 21 as the tests will fail due to SecurityManager

* Use separated jobs for build and test (to use different JDK matrix)
Re-add -Dsurefire.rerunFailingTestsCount=3 option to tests

* Update right GitHub Action syntax for options

* Re-add Jenkinsfile with daily schedule, building only on s390x and including Sonar stage

* Fix surefire argLine and set memory configuration

* Temporary ignore JournalCorruptionEofIndexRecoveryTest to test the GitHub Action CI

* Remove heap memory setting and renable KahaDB test

* Clear separation between development cycle and nightly full check

* Adding logs to trigger a new build

* Revert latest for OS versions

* Do not leak connections while constantly checking if the new connection is different from the failed one

* Avoid race conditions by waiting for the log before creating the second consumer.

* [AMQ-9820]: closed connections leaking into the pool when reconnectOnException is used

* try to avoid failures when stats are not fully synchronized

* Increase the timeout to allow statistics to sync up

* switch the latch and the counter, to make sure we wait before asserting the values bellow

* See if we can better fix the ConnectionPool

* Add all platforms for on-demand Jenkins

---------

Co-authored-by: Jean-Louis Monteiro <jl...@tomitribe.com>

> Closed connections leaked in the connection pool
> ------------------------------------------------
>
>                 Key: AMQ-9820
>                 URL: https://issues.apache.org/jira/browse/AMQ-9820
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: ConnectionPool
>            Reporter: Jean-Louis Monteiro
>            Assignee: Matt Pavlovich
>            Priority: Major
>             Fix For: 6.3.0, 5.19.2
>
>          Time Spent: 1h
>  Remaining Estimate: 0h
>
> When using reconnectOnException for the PooledConnectionFactory, it seems that closed connections (on exception) remain in the pool due to wrong validation rules



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@activemq.apache.org
For additional commands, e-mail: issues-help@activemq.apache.org
For further information, visit: https://activemq.apache.org/contact