cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kurt Greaves (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-13142) Upgradesstables cancels compactions unnecessarily
Date Mon, 26 Jun 2017 12:11:00 GMT


Kurt Greaves commented on CASSANDRA-13142:

[~krummas] I've updated the branch although haven't addressed your second point yet. I agree
we should block until all compactions for a given CF are finished, however I'm having trouble
identifying a way to get currently running compaction futures. Is there some novel way to
achieve this or should I just go the route of using {{CompactionInfo.Holder}}?

Also regarding {{cfs.markAllCompacting()}}, I've changed it to take operations to interrupt,
however the main change being that it now uses {{getUncompactingSSTables()}} rather than {{getPermittedToCompactSSTables()}}.
I believe this is OK as at the moment it's only ever called after all compactions are stopped,
so these should return the same assuming all types are interruptible, otherwise it will return
the subset of SSTables that are not compacting and didn't get forcibly stopped. Worth checking
my logic here though.

And yeah those last three points were broken because I changed tract from "uninterruptibles"
to "interruptibles" and confused myself... heh. Should be addressed now.

> Upgradesstables cancels compactions unnecessarily
> -------------------------------------------------
>                 Key: CASSANDRA-13142
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Kurt Greaves
>            Assignee: Kurt Greaves
>         Attachments: 13142-v1.patch
> Since at least 1.2 upgradesstables will cancel any compactions bar validations when run.
This was originally determined as a non-issue in CASSANDRA-3430 however can be quite annoying
(especially with STCS) as a compaction will output the new version anyway. Furthermore, as
per CASSANDRA-12243 it also stops things like view builds and I assume secondary index builds
as well which is not ideal.
> We should avoid cancelling compactions unnecessarily.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message