You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafodion.apache.org by "Oliver Bucaojit (JIRA)" <ji...@apache.org> on 2015/10/16 03:47:05 UTC

[jira] [Created] (TRAFODION-1531) Hitting WrongRegionException during trickle load testing

Oliver Bucaojit created TRAFODION-1531:
------------------------------------------

             Summary: Hitting WrongRegionException during trickle load testing
                 Key: TRAFODION-1531
                 URL: https://issues.apache.org/jira/browse/TRAFODION-1531
             Project: Apache Trafodion
          Issue Type: Bug
          Components: dtm
    Affects Versions: 1.0 (pre-incubation)
            Reporter: Oliver Bucaojit


Selva and Dennis have been testing trickle loading and an error is causing the logs to be filled and the WrongRegionException to be logged in the HBase server logs.  

Here is a stack trace that is being logged:
2015-10-14 16:46:02,664 WARN org.apache.hadoop.hbase.regionserver.HRegion: Failed getting lock in batch put, row=\x00\x00\x00\x18\x80\x00\x00\x00\x00\x00\x0B\xA0\x80\x00\x00\x01
org.apache.hadoop.hbase.regionserver.WrongRegionException: Requested row out of range for row lock on HRegion
..
..
 
       at org.apache.hadoop.hbase.regionserver.HRegion.doBatchMutate(HRegion.java:3380)
       at org.apache.hadoop.hbase.regionserver.HRegion.put(HRegion.java:2546)
       at org.apache.hadoop.hbase.coprocessor.transactional.TrxRegionEndpoint.commit(TrxRegionEndpoint.java:3754)
       at org.apache.hadoop.hbase.coprocessor.transactional.TrxRegionEndpoint.commit(TrxRegionEndpoint.java:4615)
       at org.apache.hadoop.hbase.coprocessor.transactional.TrxRegionEndpoint.commit(TrxRegionEndpoint.java:663)
       
The root cause is still unknown, a few ideas we have would be:
1) The list of puts method is being used and is batching row(s) to the incorrect region
2) There is an issue with the endkey byte-1 method and the row is being sent to the wrong region from the client

One change that will help debug the issue and keep the database consistent is to add a checkRow() method on the server-side put and delete to make sure the row key belongs in the region.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)