Job init fail - job_splitmetainfo file does not exist

[#1]

Hi All,

I have 5 node cluster with HDP 2.3.4.

I use shell script with hive command to create a table and copy some information from the other table.

It works fine when I run it under hive command line, but it show error below when I use workflow to run it.

I found some information on net such as disable permission on dfs and set permission of the staging folder to 777, but they didn't work.

could you help on this?

Error message from cluster management page---------------------------------------------------------------------------------------------------------

Job init failed : org.apache.hadoop.yarn.exceptions.YarnRuntimeException: java.io.FileNotFoundException: File does not exist: hdfs://bigdata01:8020/user/hdfs/.staging/job_1455495681392_0006/job.splitmetainfo
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl$InitTransition.createSplits(JobImpl.java:1568)
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl$InitTransition.transition(JobImpl.java:1432)
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl$InitTransition.transition(JobImpl.java:1390)
at org.apache.hadoop.yarn.state.StateMachineFactory$MultipleInternalArc.doTransition(StateMachineFactory.java:385)
at org.apache.hadoop.yarn.state.StateMachineFactory.doTransition(StateMachineFactory.java:302)
at org.apache.hadoop.yarn.state.StateMachineFactory.access$300(StateMachineFactory.java:46)
at org.apache.hadoop.yarn.state.StateMachineFactory$InternalStateMachine.doTransition(StateMachineFactory.java:448)
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl.handle(JobImpl.java:996)
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl.handle(JobImpl.java:138)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster$JobEventDispatcher.handle(MRAppMaster.java:1346)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster.serviceStart(MRAppMaster.java:1121)
at org.apache.hadoop.service.AbstractService.start(AbstractService.java:193)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster$5.run(MRAppMaster.java:1557)
at java.security.AccessController.doPrivileged(Native Method)
at javax.security.auth.Subject.doAs(Subject.java:422)
at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1657)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster.initAndStartAppMaster(MRAppMaster.java:1553)
at org.apache.hadoop.mapreduce.v2.app.MRAppMaster.main(MRAppMaster.java:1486)
Caused by: java.io.FileNotFoundException: File does not exist: hdfs://bigdata01:8020/user/hdfs/.staging/job_1455495681392_0006/job.splitmetainfo
at org.apache.hadoop.hdfs.DistributedFileSystem$22.doCall(DistributedFileSystem.java:1319)
at org.apache.hadoop.hdfs.DistributedFileSystem$22.doCall(DistributedFileSystem.java:1311)
at org.apache.hadoop.fs.FileSystemLinkResolver.resolve(FileSystemLinkResolver.java:81)
at org.apache.hadoop.hdfs.DistributedFileSystem.getFileStatus(DistributedFileSystem.java:1311)
at org.apache.hadoop.mapreduce.split.SplitMetaInfoReader.readSplitMetaInfo(SplitMetaInfoReader.java:51)
at org.apache.hadoop.mapreduce.v2.app.job.impl.JobImpl$InitTransition.createSplits(JobImpl.java:1563)
... 17 more

Error message from hue--------------------------------------------------------------------------------------------------------------------------------------

            Log Length: 1433
          WARNING: Use "yarn jar" to launch YARN applications.
16/02/15 09:36:22 WARN conf.HiveConf: HiveConf of name hive.server2.enable.impersonation does not exist

Logging initialized using configuration in file:/etc/hive/2.3.4.0-3485/0/hive-log4j.properties
Query ID = yarn_20160215093628_55a46c17-356e-41f9-a9ff-363ab845877f
Total jobs = 1
Launching Job 1 out of 1
Number of reduce tasks determined at compile time: 1
In order to change the average load for a reducer (in bytes):
  set hive.exec.reducers.bytes.per.reducer=<number>
In order to limit the maximum number of reducers:
  set hive.exec.reducers.max=<number>
In order to set a constant number of reducers:
  set mapreduce.job.reduces=<number>
Starting Job = <a href="http://10.195.226.173:8888/jobbrowser/jobs/job_1455495681392_0006">job_1455495681392_0006</a>, Tracking URL = http://bigdata02:8088/proxy/application_1455495681392_0006/
Kill Command = <a href="http://10.195.226.173:8888/filebrowser/view/usr/hdp/2.3.4.0-3485/hadoop/bin/hadoop">/usr/hdp/2.3.4.0-3485/hadoop/bin/hadoop</a> job  -kill <a href="http://10.195.226.173:8888/jobbrowser/jobs/job_1455495681392_0006">job_1455495681392_0006</a>
Hadoop job information for Stage-1: number of mappers: 0; number of reducers: 0
2016-02-15 09:37:01,459 Stage-1 map = 0%,  reduce = 0%
Ended Job = <a href="http://10.195.226.173:8888/jobbrowser/jobs/job_1455495681392_0006">job_1455495681392_0006</a> with errors
Error during job, obtaining debugging information...
FAILED: Execution Error, return code 2 from org.apache.hadoop.hive.ql.exec.mr.MapRedTask
MapReduce Jobs Launched: 
Stage-Stage-1:  HDFS Read: 0 HDFS Write: 0 FAIL
Total MapReduce CPU Time Spent: 0 msec
Failing Oozie Launcher, Main class [org.apache.oozie.action.hadoop.ShellMain], exit code [1]

[#2]

From which user you are running the job?

[#3]

I add user named "hdfs" and use it to run the job.

[#4]

@Tony Chen

make sure your nonroot users have correct permissions. This bug is discussed in https://issues.apache.org/jira/plugins/servlet/mobile#issue/MAPREDUCE-3056

[#5]

@Artem Ervits

It seems that the bug is due to the value of yarn.app.mapreduce.am.staging-dir.

I found the property "yarn.app.mapreduce.am.staging-dir" and the default value is "/user", but I have no idea what value I should use.

[#6]

@Tony Chen

You are using super user hdfs to submit the job. The bug mentioned by Artem is fixed already.

Please submit job as yarn user and try

[#7]

@Tony Chen

Is this correct HDF uri? hdfs://bigdata01:8020/

[#8]

@Neeraj Sabharwal

Thanks.

It works fine when I use yarn user to submit the job.

How do I know the HDF url is correct or not?

[#9]

@Tony Chen You can login to ambari

HDFS --> Config and in filter tab search uri

[#10]

@Tony Chen Stick with yarn user to submit the job.