You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Michael Semb Wever (Jira)" <ji...@apache.org> on 2023/01/03 19:39:00 UTC
[jira] [Commented] (CASSANDRA-18130) Log hardware and container params during test runs to help troubleshoot intermittent failures
[ https://issues.apache.org/jira/browse/CASSANDRA-18130?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17654165#comment-17654165 ]
Michael Semb Wever commented on CASSANDRA-18130:
------------------------------------------------
Initial attempt at improving this is found in this patch here: https://github.com/thelastpickle/cassandra/blob/mck/17869/trunk/.jenkins/Jenkinsfile#L291-L301
It's only touches on what is requested here, just reporting disk usage. But it's approach in keeping reports per agent is new, and can be re-used for poorman's monitoring.
Currently the patch is part of CASSANDRA-17869, but I plan on opening a new jira to cover this rewrite of the Jenkinsfile.
> Log hardware and container params during test runs to help troubleshoot intermittent failures
> ---------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-18130
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18130
> Project: Cassandra
> Issue Type: Task
> Components: Test/dtest/java, Test/dtest/python, Test/unit
> Reporter: Josh McKenzie
> Assignee: Josh McKenzie
> Priority: Normal
>
> {color:#000000}We’ve long had flakiness in our containerized ASF CI environment that we don’t see in circleci. The environment itself is both containerized and heterogenous, so there are differences in both the hardware environment and the software environment in which it executes. For reference, see: [https://github.com/apache/cassandra-builds/blob/trunk/ASF-jenkins-agents.md#current-agents]{color}
> {color:#000000} {color}
> {color:#000000}We should log a variety of hardware, container, and software environment details to help get to the bottom of where some test failures may be occurring. As we don’t have shell access to the machines it’ll be easier to have this information logged / retrieved during test runs than to try and profile each host independently.{color}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org