hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jim Kellerman (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HBASE-922) Bloom filters will not work if enabled after table is created and contains data
Date Sun, 02 Nov 2008 07:49:44 GMT

     [ https://issues.apache.org/jira/browse/HBASE-922?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Jim Kellerman updated HBASE-922:

    Fix Version/s:     (was: 0.19.0)

Moving to 0.20.0 since the best way of resolving this problem is if the HTableDescriptor is
readily available to the client (which means moving HTableDescriptor out of HRegionInfo).

We should add a release note about this problem for 0.19.0, saying that a column's bloom filter
setting should not be changed after the table is created.

> Bloom filters will not work if enabled after table is created and contains data
> -------------------------------------------------------------------------------
>                 Key: HBASE-922
>                 URL: https://issues.apache.org/jira/browse/HBASE-922
>             Project: Hadoop HBase
>          Issue Type: Bug
>    Affects Versions: 0.18.0
>            Reporter: Jim Kellerman
>            Assignee: Jim Kellerman
>            Priority: Blocker
>             Fix For: 0.20.0
> Currently it is possible to enable a bloom filter after a table is created and data has
been stored in it by disabling the table and modifying the column.
> While this correctly sets the attribute in the column descriptor, it does not create
bloom filters for existing data so that on the first compaction, an NPE will be thrown because
the HStore expects each HStoreFile to have a bloom filter.

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

View raw message