[
https://issues.apache.org/jira/browse/CASSANDRA-7066?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14645754#comment-14645754
]
Benedict commented on CASSANDRA-7066:
-------------------------------------
Thanks Stef.
A couple of comments about the description:
* It would be helpful to elaborate "replace the compactions_in_progress system
table" to include a mention of the elimination of temporary file markers.
* It would be helpful to mention that if restoring from backup, these files
should either be deleted (in the case of a full backup), or confirmed not to
reference the files they're restoring
** Perhaps we should offer a tool, like sstablelister, to offline complete any
transaction log files, which can be specified as prep-work for backup restore
* I've noticed a few English speakers (including in this conversation here) use
"loose" as "lose" which no doubt encouraged you to do the same. But it's
incorrect :)
As far as the patch goes, it looks good to me on the whole (though I'll give it
a full review once we've finalised the details). In the meantime, the
incremental checksum at the very least seems like a good idea to detect even
slight corruption of the log file (on the assumption that such corruption might
indicate we've lost the commit record, and had been in the process of it, which
seems to be one of the concerns).
I agree about individual files over-complicating things, however we _could _
perhaps take the maximum last modified of all files with that descriptor, and
then delete the files in ascending order of last modified. This would just help
the log files be robust to backups being copied in with these files still
present and referencing them as old, and also to some potential effects of
problems like CASSANDRA-9908.
> Simplify (and unify) cleanup of compaction leftovers
> ----------------------------------------------------
>
> Key: CASSANDRA-7066
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7066
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Benedict
> Assignee: Stefania
> Priority: Minor
> Labels: benedict-to-commit, compaction
> Fix For: 3.0 alpha 1
>
> Attachments: 7066.txt
>
>
> Currently we manage a list of in-progress compactions in a system table,
> which we use to cleanup incomplete compactions when we're done. The problem
> with this is that 1) it's a bit clunky (and leaves us in positions where we
> can unnecessarily cleanup completed files, or conversely not cleanup files
> that have been superceded); and 2) it's only used for a regular compaction -
> no other compaction types are guarded in the same way, so can result in
> duplication if we fail before deleting the replacements.
> I'd like to see each sstable store in its metadata its direct ancestors, and
> on startup we simply delete any sstables that occur in the union of all
> ancestor sets. This way as soon as we finish writing we're capable of
> cleaning up any leftovers, so we never get duplication. It's also much easier
> to reason about.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
