新搭建了个集群，搭建完了，运行

hadoop jar /opt/hadoop-2.4.1/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.4.1.jar pi 10 100

测试一下，却爆出下面的错误：

2014-08-24 02:24:02,456 INFO org.apache.hadoop.yarn.server.resourcemanager.rmapp.RMAppImpl: Application application_1408860487557_0002 failed 2 times due to AM Container for appattempt_1408860487557_0002_000002 exited with  exitCode: 1 due to: Exception from container-launch: org.apache.hadoop.util.Shell$ExitCodeException:  
org.apache.hadoop.util.Shell$ExitCodeException:  
at org.apache.hadoop.util.Shell.runCommand(Shell.java:505)  
at org.apache.hadoop.util.Shell.run(Shell.java:418)  
at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:650)  
at org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:195)   at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:300)  
at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:81)  
at java.util.concurrent.FutureTask.run(FutureTask.java:262)  
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)  
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)  
at java.lang.Thread.run(Thread.java:745)  
Container exited with a non-zero exit code 1  
.Failing this attempt.. Failing the application.  
2014-08-24 02:24:02,457 INFO org.apache.hadoop.yarn.server.resourcemanager.rmapp.RMAppImpl: application_1408860487557_0002 State change from FINAL_SAVING to FAILED  
2014-08-24 02:24:02,457 WARN org.apache.hadoop.yarn.server.resourcemanager.RMAuditLogger: USER=mapred   OPERATION=Application Finished - Failed TARGET=RMAppManager     RESULT=FAILURE  DESCRIPTION=App failed with state: FAILED       PERMISSIONS=Application application_1408860487557_0002 failed 2 times due to AM Container for appattempt_1408860487557_0002_000002 exited with  exitCode: 1 due to: Exception from container-launch: org.apache.hadoop.util.Shell$ExitCodeException:  
org.apache.hadoop.util.Shell$ExitCodeException:  
at org.apache.hadoop.util.Shell.runCommand(Shell.java:505)  
at org.apache.hadoop.util.Shell.run(Shell.java:418)  
at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:650)  
at org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:195)  
at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:300)  
at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:81)  
at java.util.concurrent.FutureTask.run(FutureTask.java:262)  
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)

想到yarn-site.xml中使用了一个以前没配置过的参数：

<property>  
<name>yarn.application.classpath</name>  
<value>  
$HADOOP_CONF_DIR,  
$HADOOP_COMMON_HOME/*,$HADOOP_COMMON_HOME/lib/*,  
$HADOOP_HDFS_HOME/*,$HADOOP_HDFS_HOME/lib/*,  
$HADOOP_MAPRED_HOME/*,$HADOOP_MAPRED_HOME/lib/*,  
$HADOOP_YARN_HOME/*,$HADOOP_YARN_HOME/lib/*  
</value>  
</property>

注释掉这个配置，重新运行MR任务，不报错了，毋庸置疑就是路径的问题了。先记录下来，以后再找具体原因。