[
https://issues.apache.org/jira/browse/HDFS-16070?focusedWorklogId=611782&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-611782
]
ASF GitHub Bot logged work on HDFS-16070:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 16/Jun/21 07:39
Start Date: 16/Jun/21 07:39
Worklog Time Spent: 10m
Work Description: zhengchenyu commented on a change in pull request #3105:
URL: https://github.com/apache/hadoop/pull/3105#discussion_r652427364
##########
File path:
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/DataNode.java
##########
@@ -2394,16 +2396,22 @@ void transferBlock(ExtendedBlock block, DatanodeInfo[]
xferTargets,
int numTargets = xferTargets.length;
if (numTargets > 0) {
- final String xferTargetsString =
- StringUtils.join(" ", Arrays.asList(xferTargets));
- LOG.info("{} Starting thread to transfer {} to {}", bpReg, block,
- xferTargetsString);
+ if (transferringBlock.contains(block)) {
+ LOG.warn(
Review comment:
Just like I said, when datanode's io is busy, DataTransfer will stuck
for long time, the pendingReconstructBlock will timeout. Then a duplicated
DataTransfer thread for same block will start, and the duplicated DataTransfer
thread will make io busier. In fact, we found our datanode's io is busy for
long time util we restart datanode. So I think we should avoid the duplicated
DataTransfer.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 611782)
Time Spent: 3h (was: 2h 50m)
> DataTransfer block storm when datanode's io is busy.
> ----------------------------------------------------
>
> Key: HDFS-16070
> URL: https://issues.apache.org/jira/browse/HDFS-16070
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 3.3.0, 3.2.1
> Reporter: zhengchenyu
> Priority: Major
> Labels: pull-request-available
> Time Spent: 3h
> Remaining Estimate: 0h
>
> When I speed up the decommission, I found that some datanode's io is busy,
> then I found host's load is very high, and ten thousands data transfer thread
> are running.
> Then I find log like below.
> {code}
> # setup datatranfer log
> 2021-06-08 13:42:37,620 INFO org.apache.hadoop.hdfs.server.datanode.DataNode:
> DatanodeRegistration(10.201.4.49:9866,
> datanodeUuid=6c55b7cb-f8ef-445b-9cca-d82b5b077ed1, infoPort=9864,
> infoSecurePort=0, ipcPort=9867,
> storageInfo=lv=-57;cid=CID-37e80bd5-733a-4d7b-ba3d-b46269573c72;nsid=215490653;c=1584525570797)
> Starting thread to transfer
> BP-852924019-10.201.1.32-1584525570797:blk_-9223372036449848858_30963611 to
> 10.201.7.52:9866
> 2021-06-08 13:52:36,345 INFO org.apache.hadoop.hdfs.server.datanode.DataNode:
> DatanodeRegistration(10.201.4.49:9866,
> datanodeUuid=6c55b7cb-f8ef-445b-9cca-d82b5b077ed1, infoPort=9864,
> infoSecurePort=0, ipcPort=9867,
> storageInfo=lv=-57;cid=CID-37e80bd5-733a-4d7b-ba3d-b46269573c72;nsid=215490653;c=1584525570797)
> Starting thread to transfer
> BP-852924019-10.201.1.32-1584525570797:blk_-9223372036449848858_30963611 to
> 10.201.7.31:9866
> 2021-06-08 14:02:37,197 INFO org.apache.hadoop.hdfs.server.datanode.DataNode:
> DatanodeRegistration(10.201.4.49:9866,
> datanodeUuid=6c55b7cb-f8ef-445b-9cca-d82b5b077ed1, infoPort=9864,
> infoSecurePort=0, ipcPort=9867,
> storageInfo=lv=-57;cid=CID-37e80bd5-733a-4d7b-ba3d-b46269573c72;nsid=215490653;c=1584525570797)
> Starting thread to transfer
> BP-852924019-10.201.1.32-1584525570797:blk_-9223372036449848858_30963611 to
> 10.201.16.50:9866
> # datatranfer done log
> 2021-06-08 13:54:08,134 INFO org.apache.hadoop.hdfs.server.datanode.DataNode:
> DataTransfer, at bd-tz1-hadoop-004049.zeus.lianjia.com:9866: Transmitted
> BP-852924019-10.201.1.32-1584525570797:blk_-9223372036449848858_30963611
> (numBytes=7457424) to /10.201.7.52:9866
> 2021-06-08 14:10:47,170 INFO org.apache.hadoop.hdfs.server.datanode.DataNode:
> DataTransfer, at bd-tz1-hadoop-004049.zeus.lianjia.com:9866: Transmitted
> BP-852924019-10.201.1.32-1584525570797:blk_-9223372036449848858_30963611
> (numBytes=7457424) to /10.201.16.50:9866
> {code}
> You will see last datatranfser thread was done on 13:54:08, but next
> datatranfser was start at 13:52:36.
> If datatranfser was not done in 10min(pending timeout + check interval), then
> next datatranfser for same block will be running. Then disk and network are
> heavy.
> Note: decommission ec block will trigger this problem easily, becuase every
> ec internal block are unique.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
