You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "Francesco Nigro (Jira)" <ji...@apache.org> on 2021/05/16 07:36:00 UTC
[jira] [Updated] (ARTEMIS-3298) CriticalAnalyzer alert with
replication and many clients
[ https://issues.apache.org/jira/browse/ARTEMIS-3298?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Francesco Nigro updated ARTEMIS-3298:
-------------------------------------
Attachment: storage_manager_crash.log
> CriticalAnalyzer alert with replication and many clients
> --------------------------------------------------------
>
> Key: ARTEMIS-3298
> URL: https://issues.apache.org/jira/browse/ARTEMIS-3298
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Major
> Attachments: storage_manager_crash.log
>
>
> While performing load testing with a single pair broker, the critical analyzer kick-in on master broker, shutting it down.
> The broker version used is https://github.com/apache/activemq-artemis/commit/fe361d3faeba0a8ff12f16a8fed45bb39d4c6cbd
> Next runs are performed using by switching to LOG policy to understand if the issue is legit or not.
> On a 16 cores client machine, the critical number of producers/consumers seems 16-16, sending durable 100 bytes messages.
> The journal configuration (for both master and slave) is MAPPED without datasync.
> I'm going to update the description as soon as I try with other journal configurations.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)