[jira] Assigned: (HADOOP-1625) "could not move files" exception in DataXceiver

["Raghu Angadi (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-1625?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi reassigned HADOOP-1625:
------------------------------------

    Assignee: Raghu Angadi

> "could not move files" exception in DataXceiver
> -----------------------------------------------
>
>                 Key: HADOOP-1625
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1625
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>    Affects Versions: 0.14.0
>            Reporter: Konstantin Shvachko
>            Assignee: Raghu Angadi
>            Priority: Blocker
>             Fix For: 0.14.0
>
>
> I am running TestDFSIO with the new crcs.
> The test fails with the following exception repeated several times
> 07/07/17 19:15:50 WARN fs.DFSClient: Error while writing.
> java.io.EOFException
>     at java.io.DataInputStream.readByte(DataInputStream.java:243)
>     at org.apache.hadoop.dfs.DFSClient$DFSOutputStream.endBlock(DFSClient.java:1663)
>     at org.apache.hadoop.dfs.DFSClient$DFSOutputStream.close(DFSClient.java:1735)
>     at org.apache.hadoop.fs.FSDataOutputStream$PositionCache.close(FSDataOutputStream.java:49)
>     at org.apache.hadoop.fs.FSDataOutputStream.close(FSDataOutputStream.java:64)
>     at org.apache.hadoop.io.SequenceFile$Writer.close(SequenceFile.java:773)
>     at org.apache.hadoop.fs.TestDFSIO.createControlFile(TestDFSIO.java:129)
>     at org.apache.hadoop.fs.TestDFSIO.main(TestDFSIO.java:353)
> The data-node log also contains an exception.
> 07/07/17 19:15:50 ERROR dfs.DataNode: DataXCeiver
> java.io.IOException: could not move files for blk_-2838788366095905360 from tmp to C:\ndfs\data\current\blk_-2838788366095905360
>     at org.apache.hadoop.dfs.FSDataset$FSDir.addBlock(FSDataset.java:93)
>     at org.apache.hadoop.dfs.FSDataset$FSDir.addBlock(FSDataset.java:82)
>     at org.apache.hadoop.dfs.FSDataset$FSVolume.addBlock(FSDataset.java:327)
>     at org.apache.hadoop.dfs.FSDataset.finalizeBlock(FSDataset.java:593)
>     at org.apache.hadoop.dfs.DataNode$DataXceiver.writeBlock(DataNode.java:962)
>     at org.apache.hadoop.dfs.DataNode$DataXceiver.run(DataNode.java:734)
>     at java.lang.Thread.run(Thread.java:595)
> In debugger I can see that the meta-data file rename fails.
> The data-node runs on windows (if it is relevant).
> I mark it is as a blocker until the reverse is proven.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.