Return-Path: X-Original-To: apmail-hadoop-hdfs-issues-archive@minotaur.apache.org Delivered-To: apmail-hadoop-hdfs-issues-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 57B4D181FE for ; Wed, 24 Jun 2015 02:32:44 +0000 (UTC) Received: (qmail 79983 invoked by uid 500); 24 Jun 2015 02:32:43 -0000 Delivered-To: apmail-hadoop-hdfs-issues-archive@hadoop.apache.org Received: (qmail 79908 invoked by uid 500); 24 Jun 2015 02:32:43 -0000 Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: hdfs-issues@hadoop.apache.org Delivered-To: mailing list hdfs-issues@hadoop.apache.org Received: (qmail 79664 invoked by uid 99); 24 Jun 2015 02:32:43 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 24 Jun 2015 02:32:43 +0000 Date: Wed, 24 Jun 2015 02:32:43 +0000 (UTC) From: "zhouyingchao (JIRA)" To: hdfs-issues@hadoop.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HDFS-8496) Calling stopWriter() with FSDatasetImpl lock held may block other threads MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HDFS-8496?page=3Dcom.atlassian.= jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D14598= 757#comment-14598757 ]=20 zhouyingchao commented on HDFS-8496: ------------------------------------ [~cmccabe], Any comments=EF=BC=9F > Calling stopWriter() with FSDatasetImpl lock held may block other thread= s > -------------------------------------------------------------------------= - > > Key: HDFS-8496 > URL: https://issues.apache.org/jira/browse/HDFS-8496 > Project: Hadoop HDFS > Issue Type: Bug > Affects Versions: 2.6.0 > Reporter: zhouyingchao > Assignee: zhouyingchao > Attachments: HDFS-8496-001.patch > > > On a DN of a HDFS 2.6 cluster, we noticed some DataXceiver threads and h= eartbeat threads are blocked for quite a while on the FSDatasetImpl lock. B= y looking at the stack, we found the calling of stopWriter() with FSDataset= Impl lock blocked everything. > Following is the heartbeat stack, as an example, to show how threads are = blocked by FSDatasetImpl lock: > {code} > java.lang.Thread.State: BLOCKED (on object monitor) > at org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolume= Impl.getDfsUsed(FsVolumeImpl.java:152) > - waiting to lock <0x00000007701badc0> (a org.apache.hadoop.hdfs.= server.datanode.fsdataset.impl.FsDatasetImpl) > at org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolume= Impl.getAvailable(FsVolumeImpl.java:191) > at org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatase= tImpl.getStorageReports(FsDatasetImpl.java:144) > - locked <0x0000000770465dc0> (a java.lang.Object) > at org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHear= tBeat(BPServiceActor.java:575) > at org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerSer= vice(BPServiceActor.java:680) > at org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPSe= rviceActor.java:850) > at java.lang.Thread.run(Thread.java:662) > {code} > The thread which held the FSDatasetImpl lock is just sleeping to wait ano= ther thread to exit in stopWriter(). The stack is: > {code} > java.lang.Thread.State: TIMED_WAITING (on object monitor) > at java.lang.Object.wait(Native Method) > at java.lang.Thread.join(Thread.java:1194) > - locked <0x00000007636953b8> (a org.apache.hadoop.util.Daemon) > at org.apache.hadoop.hdfs.server.datanode.ReplicaInPipeline.stopW= riter(ReplicaInPipeline.java:183) > at org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatase= tImpl.recoverCheck(FsDatasetImpl.java:982) > at org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatase= tImpl.recoverClose(FsDatasetImpl.java:1026) > - locked <0x00000007701badc0> (a org.apache.hadoop.hdfs.server.da= tanode.fsdataset.impl.FsDatasetImpl) > at org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(= DataXceiver.java:624) > at org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteB= lock(Receiver.java:137) > at org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processO= p(Receiver.java:74) > at org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXce= iver.java:235) > at java.lang.Thread.run(Thread.java:662) > {code} > In this case, we deployed quite a lot other workloads on the DN, the loca= l file system and disk is quite busy. We guess this is why the stopWriter t= ook quite a long time. > Any way, it is not quite reasonable to call stopWriter with the FSDataset= Impl lock held. In HDFS-7999, the createTemporary() is changed to call st= opWriter without FSDatasetImpl lock. We guess we should do so in the other = three methods: recoverClose()/recoverAppend/recoverRbw(). > I'll try to finish a patch for this today.=20 -- This message was sent by Atlassian JIRA (v6.3.4#6332)