You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "Clebert Suconic (Jira)" <ji...@apache.org> on 2024/11/14 16:52:00 UTC
[jira] [Updated] (ARTEMIS-5156) Simplify and make LargeMessageFrozenTest more reliable
[ https://issues.apache.org/jira/browse/ARTEMIS-5156?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Clebert Suconic updated ARTEMIS-5156:
-------------------------------------
Description:
LargeMessageFrozenTest uses a TCP Proxy.
The TCP proxy is then turned of in an attempt to force failures.
However I have seen sporadic connection failures on places where I didn't intend to fail. To make things better I'm isolating the usage of the Proxy only on the places where I intend to force a failure, what would make the test scenario more controllable and less likely to fail elsewhere.
was:
As I was writing the test ClusteredLargeMessageInterruptTest I was trying many permutations, tinkering / hacking to reproduce an issue it was reported to me in production. once I found it I created AmqpLargeMessageTest::testDeleteUnreferencedMessage,
and I ended up leaving the test in its last permutation thinking it would be a good thing.. however the test is failing for test issues in itself (consumers not being closed.. and test races). The main purpose of the test is to accumulate messages in the SNF and kill the servers.
So as part of this task I'm simplifying the test making it more reliable.
> Simplify and make LargeMessageFrozenTest more reliable
> ------------------------------------------------------
>
> Key: ARTEMIS-5156
> URL: https://issues.apache.org/jira/browse/ARTEMIS-5156
> Project: ActiveMQ Artemis
> Issue Type: Test
> Reporter: Clebert Suconic
> Assignee: Clebert Suconic
> Priority: Minor
> Labels: pull-request-available
> Fix For: 2.39.0
>
>
> LargeMessageFrozenTest uses a TCP Proxy.
> The TCP proxy is then turned of in an attempt to force failures.
> However I have seen sporadic connection failures on places where I didn't intend to fail. To make things better I'm isolating the usage of the Proxy only on the places where I intend to force a failure, what would make the test scenario more controllable and less likely to fail elsewhere.
--
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