[ https://issues.apache.org/jira/browse/HADOOP-3197?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12586250#action_12586250 ]
dhruba borthakur commented on HADOOP-3197: ------------------------------------------ It is possible that you have a slow datanode that is participating in this pipeline only. This might have caused the DataStreamer thread to block on the socketWrite. You would also see that this thread is marked as RUNNABLE, probably because it is writing out its bufefr slowly. > Deadlock in DFCClient > --------------------- > > Key: HADOOP-3197 > URL: https://issues.apache.org/jira/browse/HADOOP-3197 > Project: Hadoop Core > Issue Type: Bug > Affects Versions: 0.16.1 > Reporter: André Martin > > The DFS Client hangs - attached the thread dump - looks like a dead lock to > me... > {noformat} > "ResponseProcessor for block blk_-7822837545361798562" prio=10 > tid=0x00002aab993dcc00 nid=0x5241 waiting for monitor entry > [0x000000004365e000..0x000000004365ecc0] > java.lang.Thread.State: BLOCKED (on object monitor) > at > org.apache.hadoop.dfs.DFSClient$DFSOutputStream$ResponseProcessor.run(DFSClient.java:1771) > - waiting to lock <0x00002aaaecf2dd08> (a java.util.LinkedList) > "DataStreamer for file > /seDNS/mapred-out/18A59C65A91D44E5BA24785DF103D1781BB0137E.cache.new block > blk_-7822837545361798562" prio=10 tid=0x00002aab96a46000 nid=0x523f runnable > [0x000000004345c000..0x000000004345cc40] > java.lang.Thread.State: RUNNABLE > at java.net.SocketOutputStream.socketWrite0(Native Method) > at java.net.SocketOutputStream.socketWrite(Unknown Source) > at java.net.SocketOutputStream.write(Unknown Source) > at java.io.BufferedOutputStream.write(Unknown Source) > - locked <0x00002aaaecf2ec50> (a java.io.BufferedOutputStream) > at java.io.DataOutputStream.write(Unknown Source) > - locked <0x00002aaaecf2ec20> (a java.io.DataOutputStream) > at > org.apache.hadoop.dfs.DFSClient$DFSOutputStream$DataStreamer.run(DFSClient.java:1623) > - locked <0x00002aaaecf2dd08> (a java.util.LinkedList) > "BackupJobQueuesThread" prio=10 tid=0x00002aab94b94000 nid=0x7cb2 waiting for > monitor entry [0x000000004244c000..0x000000004244cd40] > java.lang.Thread.State: BLOCKED (on object monitor) > at > org.apache.hadoop.dfs.DFSClient$DFSOutputStream.writeChunk(DFSClient.java:2117) > - waiting to lock <0x00002aaaecf2dd08> (a java.util.LinkedList) > at > org.apache.hadoop.fs.FSOutputSummer.writeChecksumChunk(FSOutputSummer.java:141) > at > org.apache.hadoop.fs.FSOutputSummer.flushBuffer(FSOutputSummer.java:124) > - locked <0x00002aaaecf2e670> (a > org.apache.hadoop.dfs.DFSClient$DFSOutputStream) > at org.apache.hadoop.fs.FSOutputSummer.write(FSOutputSummer.java:58) > - locked <0x00002aaaecf2e670> (a > org.apache.hadoop.dfs.DFSClient$DFSOutputStream) > at > org.apache.hadoop.fs.FSDataOutputStream$PositionCache.write(FSDataOutputStream.java:36) > at java.io.DataOutputStream.writeBytes(Unknown Source) > at > sedns.serializer.file.FileSerializerServer.serializeJobQueuesAndCache(FileSerializerServer.java:723) > - locked <0x00002aaab430fec8> (a java.util.Collections$SynchronizedSet) > at > sedns.pastry.application.ServerApp$BackupJobListThread.run(ServerApp.java:476) > "[EMAIL PROTECTED]" daemon prio=10 tid=0x00002aab94bc7c00 nid=0x7ca7 waiting > on condition [0x0000000041941000..0x0000000041941bc0] > java.lang.Thread.State: TIMED_WAITING (sleeping) > at java.lang.Thread.sleep(Native Method) > at org.apache.hadoop.dfs.DFSClient$LeaseChecker.run(DFSClient.java:597) > at java.lang.Thread.run(Unknown Source) > {noformat} > -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.