I see this in RS logs on high repeat rate in TRUNK as of recent update: .... 2010-09-26 04:18:50,271 WARN org.apache.hadoop.hbase.regionserver.wal.HLog: HDFS pipeline error detected. Found 2 replicas but expecting 3 replicas. Requesting close of hlog. 2010-09-26 04:18:50,271 WARN org.apache.hadoop.hbase.regionserver.wal.HLog: HDFS pipeline error detected. Found 2 replicas but expecting 3 replicas. Requesting close of hlog. 2010-09-26 04:18:50,303 INFO org.apache.hadoop.hbase.regionserver.wal.SequenceFileLogWriter: Using syncFs -- HDFS-200 2010-09-26 04:18:50,339 WARN org.apache.hadoop.hbase.regionserver.wal.HLog: HDFS pipeline error detected. Found 2 replicas but expecting 3 replicas. Requesting close of hlog. 2010-09-26 04:18:50,728 INFO org.apache.hadoop.hbase.regionserver.wal.HLog: Roll /hbase/.logs/su187,60020,1285474482586/10.20.20.187%3A60020.1285474728759, entries=4, filesize=6464. New hlog /hbase/.logs/su187,60020,1285474482586/10.20.20.187%3A60020.1285474730271 ...
It slows everything down. Anyone have an idea why this might have just started happening? Thanks, St.Ack
