cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <j...@apache.org>
Subject [jira] Commented: (CASSANDRA-538) when streaming sstables to other nodes, make sure they don't get compacted before they are streamed
Date Thu, 12 Nov 2009 02:22:41 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-538?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12776799#action_12776799
] 

Jonathan Ellis commented on CASSANDRA-538:
------------------------------------------

it returns null if either the row had a tombstone that no longer needs to be kept, or if there
was a recoverable error (recoverable meaning we can skip to the next row)

> when streaming sstables to other nodes, make sure they don't get compacted before they
are streamed
> ---------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-538
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-538
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Jonathan Ellis
>            Assignee: Jonathan Ellis
>             Fix For: 0.5
>
>         Attachments: 538.patch
>
>
> at one point (pre-ASF codebase?) this was handled with a bootstrap data dir, but anticompaction
no longer puts files there.
> at this point it's probably easier to just make sure streaming keeps a reference to the
sstable objects, which will keep them from being deleted.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message