cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aaron Morton (JIRA)" <j...@apache.org>
Subject [jira] Commented: (CASSANDRA-2088) Temp files for failed compactions/streaming not cleaned up
Date Mon, 07 Mar 2011 05:23:59 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-2088?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13003264#comment-13003264
] 

Aaron Morton commented on CASSANDRA-2088:
-----------------------------------------

I'm keen to try this ticket (to learn more about compaction and repair) if it's not already
been worked on. Also if it's ok for me to take a couple of days while I dig into this.

For compaction I'm looking in
- CompactionManager.doCompaction where it creates a new SSTableWriter via cfs.createCompactionWriter()

- CompactionManager.doCleanupCompaction() also uses an SSTableWriter

Are the sorts of failures we're considering for compaction ones that come from the CompactionIterator
or SSTableScanner ?

For repair I'm looking in:
- IncomingStreamReader appears to clean up the temporary pending file in some error situations.
Do we have any more info on the sorts of failures here? e.g. If there is an IOException sending
the re-stream message, or a non checked exception it will fail to cleaup the file. 
- I'm looking into what happens in StreamInSession.finished() closeIfFinished()
- Are we considering failures during the streaming or when processing the data after the stream
has finished?

Any guidance welcome. 

> Temp files for failed compactions/streaming not cleaned up
> ----------------------------------------------------------
>
>                 Key: CASSANDRA-2088
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-2088
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>            Reporter: Stu Hood
>             Fix For: 0.7.4
>
>
> From separate reports, compaction and repair are currently missing opportunities to clean
up tmp files after failures.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message