Major Description
On scoundrel's cluster, I made him set his MemStore size at 256MB and hbase.hregion.memstore.block.multiplier to 8 and it seems that it overruns the integer so we get stuff like :
2009-12-17 21:52:34,314 INFO org.apache.hadoop.hbase.regionserver.HRegion: Blocking updates for 'IPC Server handler 79 on 60020' on region text_storage,,1261107769109: memstore size 0 is >= than blocking -2.0g size 2009-12-17 21:52:44,317 DEBUG org.apache.hadoop.hbase.regionserver.HRegion: Flush requested on text_storage,,1261107769109 2009-12-17 21:52:54,320 DEBUG org.apache.hadoop.hbase.regionserver.HRegion: Flush requested on text_storage,,1261107769109 2009-12-17 21:53:04,323 DEBUG org.apache.hadoop.hbase.regionserver.HRegion: Flush requested on text_storage,,1261107769109 2009-12-17 21:53:14,325 DEBUG org.apache.hadoop.hbase.regionserver.HRegion: Flush requested on text_storage,,1261107769109 ...