Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
HBase >> mail # user >> Scanner returning keys out of order

Copy link to this message
Scanner returning keys out of order
Hi all,

we are seeing very strange behavior of HBase (version 0.90.6-cdh3u5) in
the following scenario:

  1) Open scanner and start scanning.
  2) Check order of returned keys (simple test if next key is
lexigraphically greater than the previous one).
  3) The check may occasionally fail.

When we investigated this a bit, we saw major compaction occurring on
the scanned region precisely in the time of the check failure.
Interesting is that the compaction occurred on a column family that was
not actually involved in the scanning. Has anyone else seen this? Is it
possible that this is fixed in any higher version of HBase?