[
https://issues.apache.org/jira/browse/HIVE-20695?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16641740#comment-16641740
]
Peter Vary commented on HIVE-20695:
-----------------------------------
[~ychena]: I am ok with removing the "half working" locks and setting the
refreshLocalResources to synchronized. If I understand correctly this called
once per query so this most probably not a problem. On the other hand there are
several static methods which are suffering from this synchronization problem.
If we found them - I think - it would be good idea to document these issues by
adding at least a javadoc taking note of this "feature" :)
Otherwise LGTM +1
PS: I would like to keep the habit of committing only after green results. If
there are known issues please file a jira for them and disable those tests.
Thanks,
Peter
> HoS Query fails with hive.exec.parallel=true
> --------------------------------------------
>
> Key: HIVE-20695
> URL: https://issues.apache.org/jira/browse/HIVE-20695
> Project: Hive
> Issue Type: Bug
> Components: Spark
> Affects Versions: 1.2.1
> Reporter: Yongzhi Chen
> Assignee: Yongzhi Chen
> Priority: Major
> Attachments: HIVE-20695.1.patch, HIVE-20695.2.patch
>
>
> Hive queries which fail when running a HiveOnSpark job:
> {noformat}
> ERROR : Failed to execute spark task, with exception
> 'java.lang.Exception(Failed to submit Spark work, please retry later)'
> java.lang.Exception: Failed to submit Spark work, please retry later
> at
> org.apache.hadoop.hive.ql.exec.spark.RemoteHiveSparkClient.execute(RemoteHiveSparkClient.java:186)
> at
> org.apache.hadoop.hive.ql.exec.spark.session.SparkSessionImpl.submit(SparkSessionImpl.java:71)
> at
> org.apache.hadoop.hive.ql.exec.spark.SparkTask.execute(SparkTask.java:107)
> at org.apache.hadoop.hive.ql.exec.Task.executeTask(Task.java:214)
> at
> org.apache.hadoop.hive.ql.exec.TaskRunner.runSequential(TaskRunner.java:99)
> at org.apache.hadoop.hive.ql.exec.TaskRunner.run(TaskRunner.java:79)
> Caused by:
> org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.hdfs.server.namenode.LeaseExpiredException):
> No lease on
> /tmp/hive/dbname/_spark_session_dir/e202c452-8793-4e4e-ad55-61e3d4965c69/somename.jar
> (inode 725730760): File does not exist. [Lease. Holder:
> DFSClient_NONMAPREDUCE_-1981084042_486659, pending creates: 7]
> at
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.checkLease(FSNamesystem.java:3755)
> at
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.analyzeFileState(FSNamesystem.java:3556)
> at
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getAdditionalBlock(FSNamesystem.java:3412)
> at
> org.apache.hadoop.hdfs.server.namenode.NameNodeRpcServer.addBlock(NameNodeRpcServer.java:688)
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
