[
https://issues.apache.org/jira/browse/MAPREDUCE-6704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Bibin A Chundatt updated MAPREDUCE-6704:
----------------------------------------
Attachment: temp.patch
[~tangzhankun]
Thank you for attaching logs. Had quick look are the launch_container.sh
{{HADOOP_MAPRED_HOME}} patch is set based nodemanager which should have worked
in all cases.As per my understanding could fail only when the
{{HADOOP_MAPRED_HOME}} where the container do launch is different.
Could you try based on {{temp.patch}} .Currently i don't have docker setup to
test my analysis.
{code}
+++
b/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/ContainerExecutor.java
@@ -320,11 +320,11 @@ public void writeLaunchEnv(OutputStream out, Map<String,
String> environment,
ContainerLaunch.ShellScriptBuilder.create();
Set<String> whitelist = new HashSet<>();
- whitelist.add(ApplicationConstants.Environment.HADOOP_YARN_HOME.name());
- whitelist.add(ApplicationConstants.Environment.HADOOP_COMMON_HOME.name());
- whitelist.add(ApplicationConstants.Environment.HADOOP_HDFS_HOME.name());
- whitelist.add(ApplicationConstants.Environment.HADOOP_CONF_DIR.name());
- whitelist.add(ApplicationConstants.Environment.JAVA_HOME.name());
+ String[] nmWhiteList = conf.get(YarnConfiguration.NM_ENV_WHITELIST,
+ YarnConfiguration.DEFAULT_NM_ENV_WHITELIST).split(",");
+ for (String param : nmWhiteList) {
+ whitelist.add(param);
+ }
{code}
[~templedf] {{ContainerExecutor#writeLaunchEnv}} is handling only the default
whitelist params as per current code.Looks like bug in yarn.
The above change should allow taking the {{HADOOP_MAPRED_HOME}} from docker
env during container launch.
{{launcher_container.sh}} mapred_home_patch will get exported as following.
{noformat}
export HADOOP_MAPRED_HOME=${HADOOP_MAPRED_HOME:-"/home/hadoopbuild"}
{noformat}
instead of
{noformat}
export HADOOP_MAPRED_HOME="/home/hadoopbuild"
{noformat}
> Container fail to launch for mapred application
> -----------------------------------------------
>
> Key: MAPREDUCE-6704
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6704
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Bibin A Chundatt
> Assignee: Bibin A Chundatt
> Priority: Blocker
> Attachments: 0001-MAPREDUCE-6704.patch, 0001-YARN-5026.patch,
> ClusterSetup.html, MAPREDUCE-6704.0002.patch, MR-6704-branch2.8.tar.gz,
> MR-6704-trunk.tar.gz, SingleCluster.html, container-whitelist-env-wip.patch,
> temp.patch
>
>
> Container fail to launch for mapred application.
> As part for launch script {{HADOOP_MAPRED_HOME}} default value is not set
> .After
> https://github.com/apache/hadoop/commit/9d4d30243b0fc9630da51a2c17b543ef671d035c
> {{HADOOP_MAPRED_HOME}} is not able to get from {{builder.environment()}}
> since {{DefaultContainerExecutor#buildCommandExecutor}} sets inherit to false.
> {noformat}
> 16/05/02 09:16:05 INFO mapreduce.Job: Job job_1462155939310_0004 failed with
> state FAILED due to: Application application_1462155939310_0004 failed 2
> times due to AM Container for appattempt_1462155939310_0004_000002 exited
> with exitCode: 1
> Failing this attempt.Diagnostics: Exception from container-launch.
> Container id: container_1462155939310_0004_02_000001
> Exit code: 1
> Stack trace: ExitCodeException exitCode=1:
> at org.apache.hadoop.util.Shell.runCommand(Shell.java:946)
> at org.apache.hadoop.util.Shell.run(Shell.java:850)
> at
> org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:1144)
> at
> org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:227)
> at
> org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.launchContainer(ContainerLaunch.java:385)
> at
> org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:281)
> at
> org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:89)
> 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)
> Container exited with a non-zero exit code 1. Last 4096 bytes of stderr :
> Java HotSpot(TM) 64-Bit Server VM warning: ignoring option UseSplitVerifier;
> support was removed in 8.0
> Error: Could not find or load main class
> org.apache.hadoop.mapreduce.v2.app.MRAppMaster
> Container exited with a non-zero exit code 1. Last 4096 bytes of stderr :
> Java HotSpot(TM) 64-Bit Server VM warning: ignoring option UseSplitVerifier;
> support was removed in 8.0
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-h...@hadoop.apache.org
