You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@cassandra.apache.org by "Jonathan Ellis (JIRA)" <ji...@apache.org> on 2013/04/09 22:28:16 UTC

[jira] [Updated] (CASSANDRA-5449) Make sstable compacting status un/marking less error-prone

     [ https://issues.apache.org/jira/browse/CASSANDRA-5449?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis updated CASSANDRA-5449:
--------------------------------------

    Fix Version/s: 2.0
    
> Make sstable compacting status un/marking less error-prone
> ----------------------------------------------------------
>
>                 Key: CASSANDRA-5449
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5449
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Jonathan Ellis
>            Assignee: Jonathan Ellis
>            Priority: Minor
>             Fix For: 2.0
>
>
> As demonstrated by a393072aac6292412fc465d207c411c4b6b69e0b, it's easy to introduce regressions where we don't unmark the same tables we marked.  This is primarily because the marking and unmarking are usually done by separate methods.  (The opposite problem is also possible -- performAllSSTableOperation unmarks compacting, and so does CompactionTask.execute, which can be part of a pASOp via the scrub path.)
> I suggest making markCompacting return a callable that will wrap the caller-provided code in a try/finally to centralize this.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira