[
https://issues.apache.org/jira/browse/HDFS-15378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17127656#comment-17127656
]
hemanthboyina commented on HDFS-15378:
--------------------------------------
thanks for the comment [~elgoiri]
i just had taken some time to understand why exactly we are facing the issue
here , and had attached some logs in test and at while getting xceiver count
and xmits count
Datanode (127.0.0.1:38635) has received the command for recovery
{code:java}
2020-06-07 18:55:10,052 [Command processor] INFO datanode.DataNode
(BPOfferService.java:processCommandFromActive(795)) - DatanodeCommand action:
DNA_ERASURE_CODING_RECOVERY
2020-06-07 18:55:10,086 [DataXceiver for client at /127.0.0.1:47330 [Receiving
block BP-1804107793-127.0.1.1-1591536306390:blk_-9223372036854775787_1001]]
INFO datanode.DataNode (DataXceiver.java:writeBlock(747)) - Receiving
BP-1804107793-127.0.1.1-1591536306390:blk_-9223372036854775787_1001 src:
/127.0.0.1:47330 dest: /127.0.0.1:38635 {code}
in Test , Datanode (127.0.0.1:38635) has received check for xciver count and
its xmits count , as the xceiver count was 1 the test went to next line where
we are asserting xmits count without a wait , meanwhile reconstruction work is
happening in parallel
{code:java}
##2020-06-07 18:55:10,575 [Listener at localhost/42653] WARN
hdfs.TestReconstructStripedFile
(TestReconstructStripedFile.java:testErasureCodingWorkerXmitsWeight(577)) -
called test xciver count = 1 DatanodeRegistration(127.0.0.1:38635,
datanodeUuid=d6f1f0ed-be4d-403a-bc00-32d950681c2a, infoPort=38759,
infoSecurePort=0, ipcPort=35273,
storageInfo=lv=-57;cid=testClusterID;nsid=1421432615;c=1591536306390)
2020-06-07 18:55:10,575 [Listener at localhost/42653] WARN datanode.DataNode
(DataNode.java:getXceiverCount(2252)) - called xciver count on test call 1
2020-06-07 18:55:10,576 [Listener at localhost/42653] WARN
hdfs.TestReconstructStripedFile
(TestReconstructStripedFile.java:testErasureCodingWorkerXmitsWeight(580)) -
called test xmits count
2020-06-07 18:55:10,576 [Listener at localhost/42653] WARN datanode.DataNode
(DataNode.java:getXmitsInProgress(2292)) - called xmits on test call3
2020-06-07 18:55:10,637 [Listener at localhost/42653] WARN datanode.DataNode
(DataNode.java:getXmitsInProgress(2292)) - called xmits on test call3
##2020-06-07 18:55:10,662 [StripedBlockReconstruction-0] WARN
datanode.DataNode (StripedBlockReconstructor.java:run(74)) - Reconstruction
happened{code}
upon waiting , the xmits has become zero and on completing the task the xciever
count also become zero
{code:java}
2020-06-07 18:55:10,663 [Block report processor] DEBUG BlockStateChange
(BlockManager.java:processIncrementalBlockReport(4331)) - *BLOCK*
NameNode.processIncrementalBlockReport: from 127.0.0.1:38635 receiving: 0,
received: 1, deleted: 0
2020-06-07 18:55:10,665 [Listener at localhost/39641] WARN datanode.DataNode
(DataNode.java:getXceiverCount(2252)) - called xciver count on test call 0
##2020-06-07 18:55:10,665 [Listener at localhost/42653] WARN
hdfs.TestReconstructStripedFile
(TestReconstructStripedFile.java:testErasureCodingWorkerXmitsWeight(577)) -
called test xciver count = 0 DatanodeRegistration(127.0.0.1:38635,
datanodeUuid=d6f1f0ed-be4d-403a-bc00-32d950681c2a, infoPort=38759,
infoSecurePort=0, ipcPort=35273,
storageInfo=lv=-57;cid=testClusterID;nsid=1421432615;c=1591536306390){code}
> TestReconstructStripedFile#testErasureCodingWorkerXmitsWeight is failing on
> trunk
> ---------------------------------------------------------------------------------
>
> Key: HDFS-15378
> URL: https://issues.apache.org/jira/browse/HDFS-15378
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: hemanthboyina
> Priority: Major
> Attachments: HDFS-15378.001.patch
>
>
> [https://builds.apache.org/job/PreCommit-HDFS-Build/29377/#showFailuresLink]
> [https://builds.apache.org/job/PreCommit-HDFS-Build/29368/]
--
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
