[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17500478#comment-17500478
]
chen commented on HDFS-7060:
[~cheersyang] Over time, the datanode becomes slower and slower; after
restarting the datanode, it will be stable for a period of time（about 15
days）。if I patch to 2.6.0，can solve this always。
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
>Priority: Major
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16877257#comment-16877257
]
Anes Mukhametov commented on HDFS-7060:
---
got the same issue with cdh 5.16 (hadoop 2.6/2.7)
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
>Priority: Major
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16307926#comment-16307926
]
Weiwei Yang commented on HDFS-7060:
---
Apologies for the late response [~hexiaoqiao] and thanks for your active
feedback : ).
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16305967#comment-16305967
]
He Xiaoqiao commented on HDFS-7060:
---
The last patch is good for me. there are some suggestions for testing:
1. heartbeatsAvgTime latency will be not reduced obviously, especially HDFS
cluster with federation and HA with QJM, since {{HeartbeatsAvgTime}} is a
statistical indicator which is average value about sendheartbeat times to all
{{namenodes}}, if Standby NameNode cost log time to process RPC, actually it
happens frequently when tail editlog or checkpoint, heartbeatsAvgTime indicator
at DataNode will not be reduced as expect.
2. when set DataNode only interact with Active NameNode for testing, and the
result is as expect.
for short, this patch is helpful for me. thanks [~yangjiandan] and
[~cheersyang].
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16303660#comment-16303660
]
He Xiaoqiao commented on HDFS-7060:
---
hi [~yangjiandan] [~cheersyang], we encountered this problems as well, when I
patch to 2.7.1 following [^HDFS-7060.005.patch], I don't get the expected
performance improvement. any suggestions?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16243299#comment-16243299
]
Hudson commented on HDFS-7060:
--
SUCCESS: Integrated in Jenkins build Hadoop-trunk-Commit #13198 (See
[https://builds.apache.org/job/Hadoop-trunk-Commit/13198/])
HDFS-7060. Avoid taking locks when sending heartbeats from the DataNode. (wwei:
rev bb8a6eea52cb1e2c3d0b7f8b49a1bab9e4255acd)
* (edit)
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsDatasetImpl.java
* (edit)
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/fsdataset/impl/FsVolumeImpl.java
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Jiandan Yang
> Labels: BB2015-05-TBR, locks, performance
> Fix For: 3.0.0, 3.1.0
>
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16243263#comment-16243263
]
Weiwei Yang commented on HDFS-7060:
---
Thanks [~elgoiri], please see the attachment [^HDFS Status Post Patch.png].
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR, locks, performance
> Attachments: HDFS Status Post Patch.png, HDFS-7060-002.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch, HDFS-7060.003.patch,
> HDFS-7060.004.patch, HDFS-7060.005.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16242948#comment-16242948
]
Íñigo Goiri commented on HDFS-7060:
---
I went through the failed unit tests and everything is the same OOM issue (as
[~cheersyang] described).
I haven't tested it in our clusters but I assume the metrics in the UI are
reasonable.
It would be nice to attach a screenshot with the UI showing the DFS used.
+1 on [^HDFS-7060.005.patch].
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR, locks, performance
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> HDFS-7060.005.patch, complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16241442#comment-16241442
]
Weiwei Yang commented on HDFS-7060:
---
The UT failures were not related to this patch, they were failing because
jenkins slave was running out of memory
{code}
java.lang.OutOfMemoryError: unable to create new native thread
{code}
I tried to run these failed test cases on my laptop, they were success.
[~elgoiri], please let me know if this looks good to you now.
Thanks
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> HDFS-7060.005.patch, complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16241433#comment-16241433
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 9m
33s{color} | {color:blue} Docker mode activated. {color} |
|| || || || {color:brown} Prechecks {color} ||
| {color:green}+1{color} | {color:green} @author {color} | {color:green} 0m
0s{color} | {color:green} The patch does not contain any @author tags. {color} |
| {color:red}-1{color} | {color:red} test4tests {color} | {color:red} 0m
0s{color} | {color:red} The patch doesn't appear to include any new or modified
tests. Please justify why no new tests are needed for this patch. Also please
list what manual steps were performed to verify this patch. {color} |
|| || || || {color:brown} trunk Compile Tests {color} ||
| {color:green}+1{color} | {color:green} mvninstall {color} | {color:green} 18m
8s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} compile {color} | {color:green} 0m
54s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} checkstyle {color} | {color:green} 0m
40s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} mvnsite {color} | {color:green} 1m
3s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} shadedclient {color} | {color:green}
11m 40s{color} | {color:green} branch has no errors when building and testing
our client artifacts. {color} |
| {color:green}+1{color} | {color:green} findbugs {color} | {color:green} 1m
56s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} javadoc {color} | {color:green} 0m
51s{color} | {color:green} trunk passed {color} |
|| || || || {color:brown} Patch Compile Tests {color} ||
| {color:green}+1{color} | {color:green} mvninstall {color} | {color:green} 0m
59s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} compile {color} | {color:green} 0m
59s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} javac {color} | {color:green} 0m
59s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} checkstyle {color} | {color:green} 0m
41s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} mvnsite {color} | {color:green} 1m
8s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} whitespace {color} | {color:green} 0m
0s{color} | {color:green} The patch has no whitespace issues. {color} |
| {color:green}+1{color} | {color:green} shadedclient {color} | {color:green}
11m 51s{color} | {color:green} patch has no errors when building and testing
our client artifacts. {color} |
| {color:green}+1{color} | {color:green} findbugs {color} | {color:green} 2m
44s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} javadoc {color} | {color:green} 1m
6s{color} | {color:green} the patch passed {color} |
|| || || || {color:brown} Other Tests {color} ||
| {color:red}-1{color} | {color:red} unit {color} | {color:red} 42m 4s{color}
| {color:red} hadoop-hdfs in the patch failed. {color} |
| {color:red}-1{color} | {color:red} asflicense {color} | {color:red} 0m
25s{color} | {color:red} The patch generated 10 ASF License warnings. {color} |
| {color:black}{color} | {color:black} {color} | {color:black}107m 1s{color} |
{color:black} {color} |
\\
\\
|| Reason || Tests ||
| Unreaped Processes | hadoop-hdfs:6 |
| Failed junit tests | hadoop.hdfs.TestErasureCodingPolicies |
| | hadoop.hdfs.TestSafeModeWithStripedFile |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure110 |
| | hadoop.hdfs.TestSeekBug |
| | hadoop.hdfs.TestParallelRead |
| | hadoop.hdfs.TestErasureCodeBenchmarkThroughput |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure210 |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure070 |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure090 |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure200 |
| | hadoop.hdfs.TestDFSStripedOutputStreamWithFailure040 |
| | hadoop.hdfs.TestFileChecksum |
\\
\\
|| Subsystem || Report/Notes ||
| Docker | Client=17.05.0-ce Server=17.05.0-ce Image:yetus/hadoop:5b98639 |
| JIRA Issue | HDFS-7060 |
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12896312/HDFS-7060.005.patch |
| Optional Tests | asflicense compile javac javadoc mvninstall mvnsite
unit shadedclient findbugs checkstyle |
| uname | Linux 559eb9192c4d 3.13.0-123-generic #172-Ubuntu SMP Mon Jun 26
18:04:35 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux |
| Build tool | maven |
| Personality |
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16241333#comment-16241333
]
Weiwei Yang commented on HDFS-7060:
---
Thank you [~elgoiri], I have uploaded v5 patch to format the comment and
hopefully this time it can trigger a fine jenkins job.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> HDFS-7060.005.patch, complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16240582#comment-16240582
]
Íñigo Goiri commented on HDFS-7060:
---
Thanks [~cheersyang], the comments now make clear the synchronization.
The format of the comments in [^HDFS-7060.004.patch] is a little too short (<60
chars) and the comments could fit into 3 and 2 lines respectively.
I'm OK with this once we get a clean Yetus run and the minor comment is fixed.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16240324#comment-16240324
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 0m
0s{color} | {color:blue} Docker mode activated. {color} |
| {color:red}-1{color} | {color:red} docker {color} | {color:red} 0m
7s{color} | {color:red} Docker failed to build yetus/hadoop:tp-32080. {color} |
\\
\\
|| Subsystem || Report/Notes ||
| JIRA Issue | HDFS-7060 |
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12896193/HDFS-7060.004.patch |
| Console output |
https://builds.apache.org/job/PreCommit-HDFS-Build/21964/console |
| Powered by | Apache Yetus 0.7.0-SNAPSHOT http://yetus.apache.org |
This message was automatically generated.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16240293#comment-16240293
]
Weiwei Yang commented on HDFS-7060:
---
bq. is it worth making these assumptions explicit in a comment in each of these
classes?
Added some comments in v5 patch, also fixed the checkstyle warning.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16240291#comment-16240291
]
Weiwei Yang commented on HDFS-7060:
---
Hi [~goirix]
Thanks for commenting on this issue. The patch was original posted by
[~brahmareddy] and we applied to our production cluster then rebased to trunk
as 003 patch. My colleague [~yangjiandan] and I have reviewed the patch and
agreed following
# {{FsDatasetImpl#statsLock}} is only synchronized on a bunch of read
operations, and these values returned from {{FsVolumeList}} are thread safe so
it is not necessary
# Changes in {{FsVolumeImpl}} removes unnecessary dataset locks because all
those metrics operated are atomic. This helps to minimize the locking in HB
path which reduces the HB latency a lot.
for short change #1 is nice to have and change #2 is where the magic happens.
Hope this makes sense to you and let me know your thought. Thanks.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, HDFS-7060.004.patch,
> complete_failed_qps.png, sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16238362#comment-16238362
]
Íñigo Goiri commented on HDFS-7060:
---
[^HDFS-7060.003.patch] removed both {{statsLock}} and the synchronization on
{{dataset}}.
I'm all in for removing the locks just for heart beating (we've seen similar
issues internally).
However, what are the implications here?
* {{BlockPoolSlice}} maintains the DFS used and the number of blocks. There
might be some minor disadjustment but the data should converge; I cannot
foresee corruption here.
* {{FsVolumeImpl}} seems to mostly rely on {{BlockPoolSlice}} and the rest also
seems thread safe. So again some possible disadjustment but no corruption.
Any other implications of the removal of these two synchronization points?
Is it worth making these assumptions explicit in a comment in each of these
classes?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
>Priority: Major
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16224451#comment-16224451
]
Weiwei Yang commented on HDFS-7060:
---
Hi [~yangjiandan]
Your latest patch looks good to me. Thanks for uploading the comparison result
before and after applying the patch, looks pretty good. I checked the failed
UTs and tested locally, it seems to work fine might not relate to this patch,
please double check. Also please fix the minor checkstyle issue. [~jojochuang],
[~szetszwo], [~cmccabe], [~kihwal], [~jnp], [~brahmareddy], do you see any
issue with this patch?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch, HDFS-7060.003.patch, complete_failed_qps.png,
> sendHeartbeat.png
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16224402#comment-16224402
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 0m
15s{color} | {color:blue} Docker mode activated. {color} |
|| || || || {color:brown} Prechecks {color} ||
| {color:green}+1{color} | {color:green} @author {color} | {color:green} 0m
0s{color} | {color:green} The patch does not contain any @author tags. {color} |
| {color:red}-1{color} | {color:red} test4tests {color} | {color:red} 0m
0s{color} | {color:red} The patch doesn't appear to include any new or modified
tests. Please justify why no new tests are needed for this patch. Also please
list what manual steps were performed to verify this patch. {color} |
|| || || || {color:brown} trunk Compile Tests {color} ||
| {color:green}+1{color} | {color:green} mvninstall {color} | {color:green} 16m
14s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} compile {color} | {color:green} 0m
49s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} checkstyle {color} | {color:green} 0m
38s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} mvnsite {color} | {color:green} 0m
55s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} shadedclient {color} | {color:green}
11m 10s{color} | {color:green} branch has no errors when building and testing
our client artifacts. {color} |
| {color:green}+1{color} | {color:green} findbugs {color} | {color:green} 1m
49s{color} | {color:green} trunk passed {color} |
| {color:green}+1{color} | {color:green} javadoc {color} | {color:green} 0m
47s{color} | {color:green} trunk passed {color} |
|| || || || {color:brown} Patch Compile Tests {color} ||
| {color:green}+1{color} | {color:green} mvninstall {color} | {color:green} 0m
52s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} compile {color} | {color:green} 0m
47s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} javac {color} | {color:green} 0m
47s{color} | {color:green} the patch passed {color} |
| {color:orange}-0{color} | {color:orange} checkstyle {color} | {color:orange}
0m 35s{color} | {color:orange} hadoop-hdfs-project/hadoop-hdfs: The patch
generated 1 new + 94 unchanged - 0 fixed = 95 total (was 94) {color} |
| {color:green}+1{color} | {color:green} mvnsite {color} | {color:green} 0m
50s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} whitespace {color} | {color:green} 0m
0s{color} | {color:green} The patch has no whitespace issues. {color} |
| {color:green}+1{color} | {color:green} shadedclient {color} | {color:green}
10m 49s{color} | {color:green} patch has no errors when building and testing
our client artifacts. {color} |
| {color:green}+1{color} | {color:green} findbugs {color} | {color:green} 1m
54s{color} | {color:green} the patch passed {color} |
| {color:green}+1{color} | {color:green} javadoc {color} | {color:green} 0m
46s{color} | {color:green} the patch passed {color} |
|| || || || {color:brown} Other Tests {color} ||
| {color:red}-1{color} | {color:red} unit {color} | {color:red}109m 27s{color}
| {color:red} hadoop-hdfs in the patch failed. {color} |
| {color:red}-1{color} | {color:red} asflicense {color} | {color:red} 0m
19s{color} | {color:red} The patch generated 3 ASF License warnings. {color} |
| {color:black}{color} | {color:black} {color} | {color:black}158m 40s{color} |
{color:black} {color} |
\\
\\
|| Reason || Tests ||
| Failed junit tests |
hadoop.hdfs.server.datanode.TestDataNodeVolumeFailureToleration |
| | hadoop.hdfs.server.federation.router.TestRouterRpc |
| | hadoop.hdfs.server.namenode.TestNamenodeCapacityReport |
| Timed out junit tests | org.apache.hadoop.hdfs.TestLeaseRecovery2 |
\\
\\
|| Subsystem || Report/Notes ||
| Docker | Client=17.05.0-ce Server=17.05.0-ce Image:yetus/hadoop:5b98639 |
| JIRA Issue | HDFS-7060 |
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12894663/HDFS-7060.003.patch |
| Optional Tests | asflicense compile javac javadoc mvninstall mvnsite
unit shadedclient findbugs checkstyle |
| uname | Linux 2ae54fe8fca4 3.13.0-119-generic #166-Ubuntu SMP Wed May 3
12:18:55 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux |
| Build tool | maven |
| Personality | /testptch/patchprocess/precommit/personality/provided.sh |
| git revision | trunk / 9711b78 |
| maven | version: Apache Maven 3.3.9 |
| Default Java | 1.8.0_131 |
| findbugs | v3.1.0-RC1 |
| checkstyle |
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16221789#comment-16221789
]
Jiandan Yang commented on HDFS-7060:
-
We have deploy Hadoop including patch of HDFS-7060-003.patch, sendHeartbeat
latency decreased significantly, and QPS of complete_failed also down
!sendHeartbeat.jpg|datanode_sendHeartbeat_latency!
!complete_failed_qps.jpg|complete_failed_qps!
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060-003.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16221765#comment-16221765
]
Jiandan Yang commented on HDFS-7060:
-
upload HDFS-7060-003.patch
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060-003.patch,
> HDFS-7060.000.patch, HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16212722#comment-16212722
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 0m
0s{color} | {color:blue} Docker mode activated. {color} |
| {color:red}-1{color} | {color:red} patch {color} | {color:red} 0m 8s{color}
| {color:red} HDFS-7060 does not apply to trunk. Rebase required? Wrong Branch?
See https://wiki.apache.org/hadoop/HowToContribute for help. {color} |
\\
\\
|| Subsystem || Report/Notes ||
| JIRA Issue | HDFS-7060 |
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12708051/HDFS-7060-002.patch |
| Console output |
https://builds.apache.org/job/PreCommit-HDFS-Build/21760/console |
| Powered by | Apache Yetus 0.6.0-SNAPSHOT http://yetus.apache.org |
This message was automatically generated.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16212333#comment-16212333
]
Jiandan Yang commented on HDFS-7060:
-
[~xinwei] [~brahmareddy] [~jojochuang] We encountered the same
problem(branch-2.8.2), BPServiceActor#offerService blocked because
sendHeartBeat waited for FSDataset lock, and blockReceivedAndDeleted was delay
about 60s, and eventually client can not close file and threw Exception
"Unable to close file because the last blockxxx does not have enough number of
replicas”
I think HDFS-7060 can solve our problem very well. Does this patch have any
problem? Why does it merge into trunk.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15783413#comment-15783413
]
Wei-Chiu Chuang commented on HDFS-7060:
---
BlockPoolSplice objects in FsVolumeImpl are stored in ConcurrentHashMap, so
ConcurrentModificationException is not possible. FsVolumeImpl objects are
stored in CopyOnWriteArrayList, so adding/removing volumes does not throw
ConcurrentModificationException either.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15783407#comment-15783407
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 0m
0s{color} | {color:blue} Docker mode activated. {color} |
| {color:red}-1{color} | {color:red} patch {color} | {color:red} 0m 5s{color}
| {color:red} HDFS-7060 does not apply to trunk. Rebase required? Wrong Branch?
See https://wiki.apache.org/hadoop/HowToContribute for help. {color} |
\\
\\
|| Subsystem || Report/Notes ||
| JIRA Issue | HDFS-7060 |
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12708051/HDFS-7060-002.patch |
| Console output |
https://builds.apache.org/job/PreCommit-HDFS-Build/17973/console |
| Powered by | Apache Yetus 0.5.0-SNAPSHOT http://yetus.apache.org |
This message was automatically generated.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15783405#comment-15783405
]
Wei-Chiu Chuang commented on HDFS-7060:
---
I am trying to revive this thread.
The heartbeat issue with dataset lock will only get more severe into the
future, given that datanodes are becoming more dense and multiple NameNode
support in Hadoop 3.
It's common to see DataNodes installing 12 disks, which means 12 dataset locks
per heartbeat per NameNode. Even a 1 second delay in grabing a dataset lock due
to other concurrent activities holding dataset lock results in 12 second delay
in heartbeat, and more NameNodes means even more lock contention.
I aggree [~szetszwo]'s concern about ConcurrentModificationException is valid.
But modifying BlockPoolSlice can be done by holding the BlockPoolSlice object
lock, like [~jzhuge] suggested.
getDfsUsed is, essentially an approximation of disk usage, and I don't feel the
added gain is sufficient to justify the latency in heartbeats.
At the least, I think we can have a NoLock version of getDfsUsed, which must
acquire dataset lock before being called. That way we can eliminate most lock
contention.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
>
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15215462#comment-15215462
]
Hadoop QA commented on HDFS-7060:
-
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem || Runtime || Comment ||
| {color:blue}0{color} | {color:blue} reexec {color} | {color:blue} 0m 0s
{color} | {color:blue} Docker mode activated. {color} |
| {color:red}-1{color} | {color:red} patch {color} | {color:red} 0m 5s {color}
| {color:red} HDFS-7060 does not apply to trunk. Rebase required? Wrong Branch?
See https://wiki.apache.org/hadoop/HowToContribute for help. {color} |
\\
\\
|| Subsystem || Report/Notes ||
| JIRA Patch URL |
https://issues.apache.org/jira/secure/attachment/12708051/HDFS-7060-002.patch |
| JIRA Issue | HDFS-7060 |
| Console output |
https://builds.apache.org/job/PreCommit-HDFS-Build/14976/console |
| Powered by | Apache Yetus 0.2.0 http://yetus.apache.org |
This message was automatically generated.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15215413#comment-15215413
]
John Zhuge commented on HDFS-7060:
--
If we assume dataset lock protects {{bpSlices}}. The following functions access
{{bpSlices}} but they and their callers do NOT lock dataset. Did I miss
anything? Is this a bug?
{code}
FsVolumeImpl#addBlockPool (this is a write access to bpSlices)
FsVolumeImpl#addBlockPool
FsDatasetImpl#addVolume
FsVolumeImpl#getBlockPoolUsed
FsVolumeList#getBlockPoolUsed
FsDatasetImpl#getBlockPoolUsed
FsVolumeImpl#getBlockPoolSlice
(too many callers, hard to track down !)
FsVolumeImpl#getBlockPoolList
DirectoryScanner.ReportCompiler#call
FsVolumeImpl#checkDirs
FsVolumeImpl#checkDirs
FsDatasetImpl#checkDataDir
...
{code}
A few suggestions regarding lock usage:
* All locks including intrinsic locks should include docs that clearly state
what resources they protect.
* Locks should be used in the classes they are defined. Very hard to reason if
their usages are scattered around.
If we want to protect {{bpSlices}} with lock, can we just use {{bpSlices}}
intrinsic lock instead of dataset lock?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
>
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118300#comment-15118300
]
Kihwal Lee commented on HDFS-7060:
--
We have recently tried it on a big cluster. I can't tell how much it helps, but
when the nodes were hit with heavy I/O storm, some of datanodes were slowed
down enough to be "dead" for minutes. I am guessing it could have been worse
without this change. Maybe it will be better with the lifeline protocol?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118535#comment-15118535
]
Tsz Wo Nicholas Sze commented on HDFS-7060:
---
Without holding the dataset lock, we may hit ConcurrentModificationException
since the bpSlices map could be modified by reconfiguring the volumes in
runtime. We may use a different lock to protect bpSlices. ConcurrentHashMap
is not enough (or not very useful) since it does not synchronize iterations.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118556#comment-15118556
]
Tsz Wo Nicholas Sze commented on HDFS-7060:
---
> ... Maybe it will be better with the lifeline protocol?
I agree that we need HDFS-9239: datanode lifeline protocol.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15023692#comment-15023692
]
Colin Patrick McCabe commented on HDFS-7060:
I took another look at this, and it seems reasonable. Every time I think I've
found a potential race condition issue, it turns out that the variable being
read is volatile or atomic. :)
I agree that if we get a little more testing, this would be a good improvement
to make.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15018828#comment-15018828
]
Nathan Roberts commented on HDFS-7060:
--
Has anyone given any further thought on this patch? It seems safe to me and it
eliminates serious stability issues when datanodes' disks get very busy.
_getUsed()_ returns a value that is calculated by a DU process that ran for a
long time anyway, (so it's always somewhat out-of-sync). It's not very
difficult to load up a datanode with disk-intensive tasks that prevent the
datanode from getting a heartbeat in for several minutes, eventually being
declared dead by the NN. We've seen this take out entire clusters with large
Map/Reduce merges, as well as very large shuffles.
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15019286#comment-15019286
]
Haohui Mai commented on HDFS-7060:
--
Hi [~nroberts], we're being conservative as you never know whether it will lead
to new race conditions. Can you please test it in your cluster?
> Avoid taking locks when sending heartbeats from the DataNode
>
>
> Key: HDFS-7060
> URL: https://issues.apache.org/jira/browse/HDFS-7060
> Project: Hadoop HDFS
> Issue Type: Improvement
>Reporter: Haohui Mai
>Assignee: Xinwei Qin
> Labels: BB2015-05-TBR
> Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
> HDFS-7060.001.patch
>
>
> We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
> when the DN is under heavy load of writes:
> {noformat}
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
> - locked <0x000780612fd8> (a java.lang.Object)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
> at
> org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: BLOCKED (on object monitor)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
> - waiting to lock <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
>java.lang.Thread.State: RUNNABLE
> at java.io.UnixFileSystem.createFileExclusively(Native Method)
> at java.io.File.createNewFile(File.java:1006)
> at
> org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
> - locked <0x000780304fb8> (a
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
> at
> org.apache.hadoop.hdfs.server.datanode.BlockReceiver.(BlockReceiver.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
> at
> org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
> at
> org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
> at java.lang.Thread.run(Thread.java:744)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14516265#comment-14516265
]
Xinwei Qin commented on HDFS-7060:
---
[~jnp] Thanks for your comment. This test was done before applying HDFS-7999.
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14514511#comment-14514511
]
Jitendra Nath Pandey commented on HDFS-7060:
bq. We create 400 threads to write 20 million files to a Hadoop cluster with 3
DNs. Every DN has already about 90 million blocks.
[~xinwei], this is a good test. One question, did you test before applying
HDFS-7999 patch or after?
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14481372#comment-14481372
]
Brahma Reddy Battula commented on HDFS-7060:
[~cmccabe] and [~wheat9] can you please check once..Hopefully this should get
in hadoop-2.7...
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14392672#comment-14392672
]
Xinwei Qin commented on HDFS-7060:
---
[~cmccabe] , [~wheat9] Thanks a lot for your comments.
We create 400 threads to write 20 million files to a Hadoop cluster with 3 DNs.
Every DN has already about 90 million blocks.
Without HDFS-7060 patch, heartbeat always delays about several hundreds of
seconds or even longer to make DN dead.
With HDFS-7060 patch, heartbeat only sometimes delays about 50s and the DN
never dead.
As can be seen from above, making the DN heartbeat lockless can reduce the
delay time of heartbeat significantly, especially in large-scale concurrent
read and write scenario. So, if the lock is unnecessary, removing it will be a
good idea.
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14391982#comment-14391982
]
Brahma Reddy Battula commented on HDFS-7060:
Thanks a lot for taking look into this issue..sure we'll update performance
numbers..
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14391538#comment-14391538
]
Colin Patrick McCabe commented on HDFS-7060:
[~Xinwei Qin], [~brahmareddy], do you have information on how much performance
improvement this patch gives, assuming that HDFS-7999 has already been applied?
I think this is a good improvement to do in any case, but I'm curious how
important it is to get this patch in soon. For example, should we try to get
it into 2.7? Having some performance numbers would help make that call. Thanks
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Commented] (HDFS-7060) Avoid taking locks when sending heartbeats from the DataNode
[
https://issues.apache.org/jira/browse/HDFS-7060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14391604#comment-14391604
]
Haohui Mai commented on HDFS-7060:
--
I agree with Colin that it needs to be well thought out to make sure that we
don't introduce new race conditions.
I think we need clear explanations (possibly some refactoring) before we can
put it into branch-2.
Avoid taking locks when sending heartbeats from the DataNode
Key: HDFS-7060
URL: https://issues.apache.org/jira/browse/HDFS-7060
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: Haohui Mai
Assignee: Xinwei Qin
Attachments: HDFS-7060-002.patch, HDFS-7060.000.patch,
HDFS-7060.001.patch
We're seeing the heartbeat is blocked by the monitor of {{FsDatasetImpl}}
when the DN is under heavy load of writes:
{noformat}
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.getDfsUsed(FsVolumeImpl.java:115)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.getStorageReports(FsDatasetImpl.java:91)
- locked 0x000780612fd8 (a java.lang.Object)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.sendHeartBeat(BPServiceActor.java:563)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.offerService(BPServiceActor.java:668)
at
org.apache.hadoop.hdfs.server.datanode.BPServiceActor.run(BPServiceActor.java:827)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: BLOCKED (on object monitor)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:743)
- waiting to lock 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.createFileExclusively(Native Method)
at java.io.File.createNewFile(File.java:1006)
at
org.apache.hadoop.hdfs.server.datanode.DatanodeUtil.createTmpFile(DatanodeUtil.java:59)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.BlockPoolSlice.createRbwFile(BlockPoolSlice.java:244)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.createRbwFile(FsVolumeImpl.java:195)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:753)
- locked 0x000780304fb8 (a
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl)
at
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.createRbw(FsDatasetImpl.java:60)
at
org.apache.hadoop.hdfs.server.datanode.BlockReceiver.init(BlockReceiver.java:169)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:621)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.opWriteBlock(Receiver.java:124)
at
org.apache.hadoop.hdfs.protocol.datatransfer.Receiver.processOp(Receiver.java:71)
at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:232)
at java.lang.Thread.run(Thread.java:744)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
