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 2013/09/24 17:49:06 UTC

[jira] [Commented] (CASSANDRA-6087) Node OOMs on commit log replay when starting up

    [ https://issues.apache.org/jira/browse/CASSANDRA-6087?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13776413#comment-13776413 ] 

Jonathan Ellis commented on CASSANDRA-6087:
-------------------------------------------

committed
                
> Node OOMs on commit log replay when starting up
> -----------------------------------------------
>
>                 Key: CASSANDRA-6087
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6087
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Oleg Anastasyev
>            Assignee: Oleg Anastasyev
>             Fix For: 1.2.11, 2.0.2
>
>         Attachments: CassandraDaemon.txt
>
>
> After some activity on batchlogs and hints and CFs restarted nodes without finished draining. On startup it OOMs. 
> Investigating it found, that memtables occupied all available RAM, because MeteredFlusher is started later in setup code, so memtables cannot flush. 
> Fixed by starting metered flusher before commit log replay starts

--
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