cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-3430) Break Big Compaction Lock apart
Date Fri, 08 Mar 2013 05:06:15 GMT


Jonathan Ellis commented on CASSANDRA-3430:

Pushed updates to  Addressed the above with
the following caveats:

bq. During truncate, we only stop compactions on the main CF. We should also stop it for indexes

We stop them one at a time -- index.truncate goes through CFS.truncate and runWithCompactionsDisabled
again.  I think that's adequate; i.e., we don't need to stop compactions on the index CFs
while we clean out sstables for its parent, or vice versa.

bq. Concerning that wait in runWithCompactionsDisabled, wouldn't it make sense to move it
inside interruptCompactionFor?

My reasoning here was that for DROP CF/KS, we don't actually need to wait for the compactions
to be interrupted, since unmarkCompacting will take care of cleaning up anything that finished
after the drop does.

> Break Big Compaction Lock apart
> -------------------------------
>                 Key: CASSANDRA-3430
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Jonathan Ellis
>            Assignee: Jonathan Ellis
>            Priority: Minor
>              Labels: compaction
>             Fix For: 2.0
>         Attachments: 3430-1.0.txt, 3430-1.1.txt, 3430-v2.txt, 3430-v3.txt

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:

View raw message