[jira] [Updated] (CASSANDRA-4409) Only consider whole row tombstone in collation controller

["Sylvain Lebresne (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-4409?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sylvain Lebresne updated CASSANDRA-4409:
----------------------------------------

    Attachment: 0001-Use-only-top-level-row-deletion-to-avoid-sstable-durin.txt

Patch attached to fix. I will note that in practice this was not really a bug because the deletionInfo used were those of the columnIterator, and at that point those deletionInfo should only contain whole row tombstone, not range tombstone. Yet, the code was misleading and could have trigger a bug if the code change underneath.
                
> Only consider whole row tombstone in collation controller
> ---------------------------------------------------------
>
>                 Key: CASSANDRA-4409
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-4409
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 1.2
>            Reporter: Sylvain Lebresne
>            Assignee: Sylvain Lebresne
>            Priority: Trivial
>             Fix For: 1.2
>
>         Attachments: 0001-Use-only-top-level-row-deletion-to-avoid-sstable-durin.txt
>
>
> CollationController has that optimization that if it has already seen a row tombstone more recent that a sstable max timestamp, it skips the sstable.  However, this was not updated correctly while introducing range tombstone and currently the code might skip a sstable based on the timestamp of a tombstone that does not cover the full row.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira