Please file an issue. Sounds like an OOME while writing causes us to exit w/o closing the WAL (You think that the case)? My guess is that in this low memory situation, a close might fail anyways (with another OOME) unless we did some extra gymnastics releasing the little resevoir of memory we keep around to release so cleanup succeeds whenever we see OOME.
St.Ack On Fri, Jul 15, 2011 at 9:32 AM, Andy Sautins <[email protected]> wrote: > > Yesterday we ran into an interesting issue. We were shutting down our > HBase cluster ( 0.90.1 CDH3u0 ) and in the process one of the nodes > encountered a Java heap space exception. The bummer is the log file was > listed as corrupt from hadoop fsck and was unable to be read when re-starting > the database. We were able to recover in our situation by removing the > corrupt log and did not appear to lose any data. > > Has anyone else seen this issue? If I'm reading the situation right it > looks like that a Java heap space error during the WAL checksum write could > leave the WAL corrupt which doesn't seem like desired behavior. > > I'll looking into it further but any thoughts would be appreciated. > > > 2011-07-14 14:54:53,741 FATAL org.apache.hadoop.hbase.regionserver.wal.HLog: > Could not append. Requesting close of hlog > java.io.IOException: Reflection > at > org.apache.hadoop.hbase.regionserver.wal.SequenceFileLogWriter.sync(SequenceFileLogWriter.java:147) > at org.apache.hadoop.hbase.regionserver.wal.HLog.sync(HLog.java:987) > at > org.apache.hadoop.hbase.regionserver.wal.HLog$LogSyncer.run(HLog.java:964) > Caused by: java.lang.reflect.InvocationTargetException > at sun.reflect.GeneratedMethodAccessor1336.invoke(Unknown Source) > at > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) > at java.lang.reflect.Method.invoke(Method.java:597) > at > org.apache.hadoop.hbase.regionserver.wal.SequenceFileLogWriter.sync(SequenceFileLogWriter.java:145) > ... 2 more > Caused by: java.lang.OutOfMemoryError: Java heap space > at > org.apache.hadoop.hdfs.DFSClient$DFSOutputStream$Packet.<init>(DFSClient.java:2375) > at > org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.writeChunk(DFSClient.java:3271) > at > org.apache.hadoop.fs.FSOutputSummer.writeChecksumChunk(FSOutputSummer.java:150) > at > org.apache.hadoop.fs.FSOutputSummer.flushBuffer(FSOutputSummer.java:132) > at > org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.sync(DFSClient.java:3354) > at > org.apache.hadoop.fs.FSDataOutputStream.sync(FSDataOutputStream.java:97) > at > org.apache.hadoop.io.SequenceFile$Writer.syncFs(SequenceFile.java:944) > ... 6 more > >
