hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Created: (HBASE-1769) On snapshot, need to reseek outstanding scanners (all need to register and then have their snapshot observer called).
Date Fri, 14 Aug 2009 20:37:14 GMT
On snapshot, need to reseek outstanding scanners (all need to register and then have their
snapshot observer called).
---------------------------------------------------------------------------------------------------------------------

                 Key: HBASE-1769
                 URL: https://issues.apache.org/jira/browse/HBASE-1769
             Project: Hadoop HBase
          Issue Type: Bug
            Reporter: stack
             Fix For: 0.20.1


See testScanAcrossSnapshot in TestMemStore around #126:

{code}
        // Row count is same as column count.
        // On snapshot, need to reseek our scanner.  TODO: HBASE-
        // assertEquals("count=" + count + ", result=" + result,
        //   rowCount, result.size());
{code}

The code is commented out.

This is a new bit of code aded by me fixing hbase-1738.   Its meant to verify we are getting
out the right number of columns.  Well, after adding in new data after a snapshot, we're failing
to return first record because we're seeked to what used to be first on this row -- not the
value that was just uploaded.   Means we may skip value on occasion scanning.  Need to fix
but don't think it critical for RC2 (We used to skip volumes before hbase-1738 in same circumstance).

-- 
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