[
https://issues.apache.org/jira/browse/HDFS-6098?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13932485#comment-13932485
]
Arpit Agarwal edited comment on HDFS-6098 at 3/12/14 10:08 PM:
---------------------------------------------------------------
The synchronization of {{FsDatasetImpl#bpSlices}} looks wrong. In different
locations it is modified without a lock or synchronized via
{{FsDatasetImpl#dataset}}. Also reads are unsynchronized.
was (Author: arpitagarwal):
The synchronization of {{FsDatasetImpl#bpSlices}} looks wront. In different
locations it is modified without a lock or synchronized via
{{FsDatasetImpl#dataset}}. Also reads are unsynchronized.
> ConcurrentModificationException exception during DataNode shutdown
> ------------------------------------------------------------------
>
> Key: HDFS-6098
> URL: https://issues.apache.org/jira/browse/HDFS-6098
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: datanode
> Affects Versions: 2.4.0
> Reporter: Arpit Agarwal
>
> Exception hit during DN shutdown while running
> {{TestWebHdfsWithMultipleNameNodes}}:
> {code}
> java.util.ConcurrentModificationException: null
> at java.util.HashMap$HashIterator.nextEntry(HashMap.java:894)
> at java.util.HashMap$EntryIterator.next(HashMap.java:934)
> at java.util.HashMap$EntryIterator.next(HashMap.java:932)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl.shutdown(FsVolumeImpl.java:251)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeList.shutdown(FsVolumeList.java:249)
> at
> org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetImpl.shutdown(FsDatasetImpl.java:1376)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.shutdown(DataNode.java:1301)
> at
> org.apache.hadoop.hdfs.MiniDFSCluster.shutdownDataNodes(MiniDFSCluster.java:1523)
> at
> org.apache.hadoop.hdfs.MiniDFSCluster.shutdown(MiniDFSCluster.java:1498)
> at
> org.apache.hadoop.hdfs.MiniDFSCluster.shutdown(MiniDFSCluster.java:1482)
> at
> org.apache.hadoop.hdfs.web.TestWebHdfsWithMultipleNameNodes.shutdownCluster(TestWebHdfsWithMultipleNameNodes.java:99)
> {code}
--
This message was sent by Atlassian JIRA
(v6.2#6252)
