Details
-
Bug
-
Status: Resolved
-
Blocker
-
Resolution: Not A Problem
-
3.0.0-beta1
-
None
-
None
-
None
Description
On the deployment of hadoop 3 cluster (based on current trunk branch) with default settings, the MR job will get failed as following exceptions:
2017-08-16 13:00:03,846 INFO mapreduce.Job: Job job_1502913552390_0001 running in uber mode : false 2017-08-16 13:00:03,847 INFO mapreduce.Job: map 0% reduce 0% 2017-08-16 13:00:03,864 INFO mapreduce.Job: Job job_1502913552390_0001 failed with state FAILED due to: Application application_1502913552390_0001 failed 2 times due to AM Container for appattempt_1502913552390_0001_000002 exited with exitCode: 1 Failing this attempt.Diagnostics: [2017-08-16 13:00:02.963]Exception from container-launch. Container id: container_1502913552390_0001_02_000001 Exit code: 1 Stack trace: ExitCodeException exitCode=1: at org.apache.hadoop.util.Shell.runCommand(Shell.java:994) at org.apache.hadoop.util.Shell.run(Shell.java:887) at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:1212) at org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:295) at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.launchContainer(ContainerLaunch.java:455) at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:275) at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:90) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:745)
This is because mapreduce related jar are not added into yarn setup by default. To make MR job run successful, we need to add following configurations to yarn-site.xml now:
<property>
<name>yarn.application.classpath</name>
<value>
...
<HADOOP_HOME>/share/hadoop/mapreduce/*,
<HADOOP_HOME>/share/hadoop/mapreduce/lib/*
...
</value>
But this config is not necessary for previous version of Hadoop. We should fix this issue before beta release otherwise it will be a regression for configuration changes.
This could be more like a YARN issue (if so, we should move), depends on how we fix it finally.
Attachments
Issue Links
- is related to
-
-
- Resolved
-