[jira] [Commented] (CASSANDRA-17174) Harden resource management on SSTable components to prevent further leaks

["Josh McKenzie (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-17174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17450594#comment-17450594 ] 

Josh McKenzie commented on CASSANDRA-17174:
-------------------------------------------

[Branch|https://github.com/apache/cassandra/commit/532ee78a11668f6e8206b3060b8626fcc0b524e2]

[CI|https://app.circleci.com/pipelines/github/josh-mckenzie/cassandra/141/workflows/cc872579-850d-4db6-ae6b-e80515898615]

> Harden resource management on SSTable components to prevent further leaks
> -------------------------------------------------------------------------
>
>                 Key: CASSANDRA-17174
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17174
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Local/SSTable
>            Reporter: Josh McKenzie
>            Assignee: Josh McKenzie
>            Priority: Normal
>
> We've seen resource leaks pop up w/histogram overflows repeatedly; the code in {{BigTableWriter.openEarly()}} and {{BigTableWriter.openFinal()}} doesn't appropriately catch and handle any exceptions during creation before things are registered with a {{LifecycleTransaction}} so any errors there will leak.
> We should clean that up.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org