You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@hbase.apache.org by "Murali Krishna. P" <mu...@yahoo.com> on 2009/07/02 13:14:47 UTC

HBase jobs hanging

Hi,
        I was trying some mapreduce jobs on my cluster, it stops reading from the table after some point. So, I tried the basic rowcounter job on the same table and the map tasks gets stuck at the same point. Tasks are not getting killed but it is just waiting on the RecordReader.next() call.
  Even if i start the sequential 'count' from the shell, it hangs after counting same number of rows.

Dont see much exceptions on the logs, except the following on one of the region server.
2009-07-02 03:44:54,412 INFO org.apache.hadoop.hbase.regionserver.HRegionServer: Scanner -2204554202283641986 lease expired
2009-07-02 03:57:27,101 INFO org.apache.hadoop.hbase.regionserver.HRegionServer: Scanner -6041936359480242732 lease expired
2009-07-02 03:57:27,293 INFO org.apache.hadoop.hbase.regionserver.HRegionServer: Scanner 6008116931812072259 lease expired
2009-07-02 03:57:29,165 INFO org.apache.hadoop.hbase.regionserver.HRegionServer: Scanner 6927781754287164415 lease expired

Table has 3 regions and I am using the 0.20.0 alpha build. Need help in resolving this scanner issue

Thanks,
Murali Krishna