FLUME-1163 (https://issues.apache.org/jira/browse/FLUME-1163) states that this
problem has been fixed in Flume 1.2.0. However, my tests indicate that the fix
works sometimes, but does not work most of the time (i.e. it didn't rename the
.tmp file). The exceptions occurred when it did not work are shown below. It
seems that the Filesystem has been closed when it tries to close and rename the
.tmp file.
2012-08-02 16:09:16,844 WARN hdfs.BucketWriter: failed to close() HDFSWriter
for file
(hdfs://xxx.yyy.com/user/yongli/flume/2012/08_02_16.events.1343938150590.tmp).
Exception follows.
java.io.IOException: DFSOutputStream is closed
at
org.apache.hadoop.hdfs.DFSClient$DFSOutputStream.sync(DFSClient.java:3666)
at
org.apache.hadoop.fs.FSDataOutputStream.sync(FSDataOutputStream.java:97)
at
org.apache.flume.sink.hdfs.HDFSDataStream.close(HDFSDataStream.java:103)
at
org.apache.flume.sink.hdfs.BucketWriter.doClose(BucketWriter.java:256)
at
org.apache.flume.sink.hdfs.BucketWriter.access$400(BucketWriter.java:50)
at org.apache.flume.sink.hdfs.BucketWriter$3.run(BucketWriter.java:242)
at org.apache.flume.sink.hdfs.BucketWriter$3.run(BucketWriter.java:239)
at
org.apache.flume.sink.hdfs.BucketWriter.runPrivileged(BucketWriter.java:127)
at org.apache.flume.sink.hdfs.BucketWriter.close(BucketWriter.java:239)
at
org.apache.flume.sink.hdfs.HDFSEventSink$3.call(HDFSEventSink.java:796)
at
org.apache.flume.sink.hdfs.HDFSEventSink$3.call(HDFSEventSink.java:793)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
at java.util.concurrent.FutureTask.run(FutureTask.java:138)
at
java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:619)
2012-08-02 16:09:16,853 INFO hdfs.BucketWriter: Renaming hdfs:// xxx.yyy.com
/user/yongli/flume/2012/08_02_16.events.1343938150590.tmp to
hdfs://xxx.yyy.com/user/yongli/flume /2012/08_02_16.events.1343938150590
2012-08-02 16:09:16,853 WARN hdfs.HDFSEventSink: Exception while closing
hdfs:// xxx.yyy.com /user/yongli/flume /2012/08_02_16.events. Exception follows.
java.io.IOException: Filesystem closed
at org.apache.hadoop.hdfs.DFSClient.checkOpen(DFSClient.java:264)
at org.apache.hadoop.hdfs.DFSClient.rename(DFSClient.java:800)
at
org.apache.hadoop.hdfs.DistributedFileSystem.rename(DistributedFileSystem.java:216)
at
org.apache.flume.sink.hdfs.BucketWriter.renameBucket(BucketWriter.java:377)
at
org.apache.flume.sink.hdfs.BucketWriter.doClose(BucketWriter.java:275)
at
org.apache.flume.sink.hdfs.BucketWriter.access$400(BucketWriter.java:50)
at org.apache.flume.sink.hdfs.BucketWriter$3.run(BucketWriter.java:242)
at org.apache.flume.sink.hdfs.BucketWriter$3.run(BucketWriter.java:239)
at
org.apache.flume.sink.hdfs.BucketWriter.runPrivileged(BucketWriter.java:127)
at org.apache.flume.sink.hdfs.BucketWriter.close(BucketWriter.java:239)
at
org.apache.flume.sink.hdfs.HDFSEventSink$3.call(HDFSEventSink.java:796)
at
org.apache.flume.sink.hdfs.HDFSEventSink$3.call(HDFSEventSink.java:793)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
at java.util.concurrent.FutureTask.run(FutureTask.java:138)
at
java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:619)
2012-08-02 16:09:16,856 INFO instrumentation.MonitoredCounterGroup: Component
type: SINK, name: hdfs-sink stopped
