如何在Windows下的eclipse调试Hadoop2.2.0分布式集群

上篇文章， 散仙已经在eclipse中通过local的模式可以正确的调试hadoop2.2，那么本篇，散仙将重点叙述下，如何在eclipse中，真真正正的提交作业到yarn上，开启分布式模式的调试，通过在eclipse上调试，hadoop的MapReduce程序，可以使我们学习Hadoop更加容易，清晰。


如果没有看过，散仙的如何在eclipse中使用local模式调试hadoop的文章，可以先看下上篇，熟悉下基本的问题的解决。

下面进入正题，由于散仙在上篇中，已经使用eclipse成功的使用了local模式的调试，所以本次改成分布式模式的调试，也不算太困难。使用eclipse作为客户端像yarn集群上提交作业，需要将整个项目打包成一个jar，散仙在这里使用的是一个ant脚本，文章最后，散仙会附上来，直接遇到的最大的一个问题如下异常：

<pre name="code" class="java">2014-06-11 17:32:19,761 WARN org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor: Exception from container-launch with container ID: container_1401177251807_0034_01_000001 and exit code: 1 
org.apache.hadoop.util.Shell$ExitCodeException: /bin/bash: line 0: fg: no job control 
 
    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:744)  </pre>
这个问题，在网上已经得到解决，需要下2个patch包，进行打补丁，比较繁琐，散仙，在参考了这位兄弟的文章后，http://blog.csdn.net/fansy1990/article/details/27526167
感觉使用方法解决，比较简洁方便。引起上述异常的主要原因就是，Linux和Windows的环境变量符号不一致导致的问题win上是%而linux上是$所以直接导致了上述原因，当然这个问题再linux上的eclipse是不存在，只有在win上的eclipse中，才会出现，所以我们要做的就是，改变org.apache.hadoop.mapred.YARNRunner类里面的一些方法，来消除此异常。


具体步骤，改写YARNRunner源码中的一些方法（YARNRunner.java源码类在hadoop-mapreduce-client-jobclient的maven项目中的org.apache.hadoop.mapred包下）需要在src下建同样的包名，类名，覆盖原来jar包里面自带的类。

YarnRunner.java的390行 （Apache Hadoop2.2的源码）



<pre name="code" class="java">// Setup the command to run the AM 
    List<String> vargs = new ArrayList<String>(8); 
    vargs.add(Environment.JAVA_HOME.$() + "/bin/java");  </pre>

改为
<pre name="code" class="java">vargs.add("$JAVA_HOME/bin/java");  </pre>
在YarnRunner.java类中，新增一个路径转换的方法
<pre name="code" class="java">private void replaceEnvironment(Map<String, String> environment) { 
      String tmpClassPath = environment.get("CLASSPATH"); 
      tmpClassPath=tmpClassPath.replaceAll(";", ":"); 
      tmpClassPath=tmpClassPath.replaceAll("%PWD%", "\\$PWD"); 
      tmpClassPath=tmpClassPath.replaceAll("%HADOOP_MAPRED_HOME%", "\\$HADOOP_MAPRED_HOME"); 
      tmpClassPath= tmpClassPath.replaceAll("\\\\", "/" ); 
      environment.put("CLASSPATH",tmpClassPath); 
}  </pre>


在YarnRunner.java的在466行添加：
<pre name="code" class="java">replaceEnvironment(environment);  </pre>
通过，这样设置后，原来的异常就得到解决了，散仙在这里分布式测试的例子依旧是hellow world，源码如下：
<pre name="code" class="java">package com.qin.wordcount;

import java.io.IOException;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.YARNRunner;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

/***
*
* Hadoop2.2.0完全分布式测试
* 放WordCount的例子
*
* @author qindongliang
*
* hadoop技术交流群：  376932160
*
*
* */
public class MyWordCount {


/**
* Mapper
*
* **/
private static class WMapper extends Mapper<LongWritable, Text, Text, IntWritable>{


private IntWritable count=new IntWritable(1);
private Text text=new Text();
@Override
protected void map(LongWritable key, Text value,Context context)
throws IOException, InterruptedException {
String values[]=value.toString().split("#");
//System.out.println(values[0]+"========"+values[1]);
count.set(Integer.parseInt(values[1]));
text.set(values[0]);
context.write(text,count);

}

}

/**
* Reducer
*
* **/
private static class WReducer extends Reducer<Text, IntWritable, Text, Text>{

private Text t=new Text();
@Override
protected void reduce(Text key, Iterable<IntWritable> value,Context context)
throws IOException, InterruptedException {
int count=0;
for(IntWritable i:value){
count+=i.get();
}
t.set(count+"");
context.write(key,t);

}

}


/**
* 改动一
* (1)shell源码里添加checkHadoopHome的路径
* (2)974行，FileUtils里面
* **/

public static void main(String[] args) throws Exception{


Configuration conf=new Configuration();

    conf.set("mapreduce.job.jar", "myjob.jar");
conf.set("fs.defaultFS","hdfs://192.168.46.28:9000");
conf.set("mapreduce.framework.name", "yarn"); 
conf.set("yarn.resourcemanager.address", "192.168.46.28:8032");
/**Job任务**/
   //Job job=new Job(conf, "testwordcount");//废弃此API
   Job job=Job.getInstance(conf, "new api");
job.setJarByClass(MyWordCount.class);
System.out.println("模式：  "+conf.get("mapreduce.jobtracker.address"));;
// job.setCombinerClass(PCombine.class);



// job.setNumReduceTasks(3);//设置为3
job.setMapperClass(WMapper.class);
job.setReducerClass(WReducer.class);
job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);



job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);

String path="hdfs://192.168.46.28:9000/qin/output";
FileSystem fs=FileSystem.get(conf);
Path p=new Path(path);
if(fs.exists(p)){
fs.delete(p, true);
System.out.println("输出路径存在，已删除！");
}
FileInputFormat.setInputPaths(job, "hdfs://192.168.46.28:9000/qin/input");
FileOutputFormat.setOutputPath(job,p );
System.exit(job.waitForCompletion(true) ? 0 : 1); 




}


}
</pre>
在运行的时候，需要注意把，hadoop集群上的配置文件core-site.xml,hdfs-site.xml,mapred-site.xml,yarn-site.xml文件拷贝到src的根目录下，最好也放一个log4j.xml方便查看日志。并在mapred-site.xml里面，添加如下属性：
<pre name="code" class="xml"> <name>mapred.remote.os</name>

<value>Linux</value>

<description>RemoteMapReduce framework's OS, can be either Linux orWindows</description>

</property></pre>
然后，把项目打成jar包，运行提交作业，散仙的控制台打印内容如下：
<pre name="code" class="java">模式：  hp1:8021
输出路径存在，已删除！
INFO - RMProxy.createRMProxy(56) | Connecting to ResourceManager at /192.168.46.28:8032
WARN - JobSubmitter.copyAndConfigureFiles(149) | Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
INFO - FileInputFormat.listStatus(287) | Total input paths to process : 1
INFO - JobSubmitter.submitJobInternal(394) | number of splits:1
INFO - Configuration.warnOnceIfDeprecated(840) | user.name is deprecated. Instead, use mapreduce.job.user.name
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.jar is deprecated. Instead, use mapreduce.job.jar
INFO - Configuration.warnOnceIfDeprecated(840) | fs.default.name is deprecated. Instead, use fs.defaultFS
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.output.value.class is deprecated. Instead, use mapreduce.job.output.value.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.mapoutput.value.class is deprecated. Instead, use mapreduce.map.output.value.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapreduce.map.class is deprecated. Instead, use mapreduce.job.map.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.job.name is deprecated. Instead, use mapreduce.job.name
INFO - Configuration.warnOnceIfDeprecated(840) | mapreduce.reduce.class is deprecated. Instead, use mapreduce.job.reduce.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapreduce.inputformat.class is deprecated. Instead, use mapreduce.job.inputformat.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.input.dir is deprecated. Instead, use mapreduce.input.fileinputformat.inputdir
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.output.dir is deprecated. Instead, use mapreduce.output.fileoutputformat.outputdir
INFO - Configuration.warnOnceIfDeprecated(840) | mapreduce.outputformat.class is deprecated. Instead, use mapreduce.job.outputformat.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.map.tasks is deprecated. Instead, use mapreduce.job.maps
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.output.key.class is deprecated. Instead, use mapreduce.job.output.key.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.mapoutput.key.class is deprecated. Instead, use mapreduce.map.output.key.class
INFO - Configuration.warnOnceIfDeprecated(840) | mapred.working.dir is deprecated. Instead, use mapreduce.job.working.dir
INFO - JobSubmitter.printTokens(477) | Submitting tokens for job: job_1402492118962_0004
INFO - YarnClientImpl.submitApplication(174) | Submitted application application_1402492118962_0004 to ResourceManager at /192.168.46.28:8032
INFO - Job.submit(1272) | The url to track the job: http://hp1:8088/proxy/application_1402492118962_0004/
INFO - Job.monitorAndPrintJob(1317) | Running job: job_1402492118962_0004
INFO - Job.monitorAndPrintJob(1338) | Job job_1402492118962_0004 running in uber mode : false
INFO - Job.monitorAndPrintJob(1345) |  map 0% reduce 0%
INFO - Job.monitorAndPrintJob(1345) |  map 100% reduce 0%
INFO - Job.monitorAndPrintJob(1345) |  map 100% reduce 100%
INFO - Job.monitorAndPrintJob(1356) | Job job_1402492118962_0004 completed successfully
INFO - Job.monitorAndPrintJob(1363) | Counters: 43
File System Counters
FILE: Number of bytes read=58
FILE: Number of bytes written=159667
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=147
HDFS: Number of bytes written=27
HDFS: Number of read operations=6
HDFS: Number of large read operations=0
HDFS: Number of write operations=2
Job Counters
Launched map tasks=1
Launched reduce tasks=1
Data-local map tasks=1
Total time spent by all maps in occupied slots (ms)=6155
Total time spent by all reduces in occupied slots (ms)=4929
Map-Reduce Framework
Map input records=4
Map output records=4
Map output bytes=44
Map output materialized bytes=58
Input split bytes=109
Combine input records=0
Combine output records=0
Reduce input groups=3
Reduce shuffle bytes=58
Reduce input records=4
Reduce output records=3
Spilled Records=8
Shuffled Maps =1
Failed Shuffles=0
Merged Map outputs=1
GC time elapsed (ms)=99
CPU time spent (ms)=1060
Physical memory (bytes) snapshot=309071872
Virtual memory (bytes) snapshot=1680531456
Total committed heap usage (bytes)=136450048
Shuffle Errors
BAD_ID=0
CONNECTION=0
IO_ERROR=0
WRONG_LENGTH=0
WRONG_MAP=0
WRONG_REDUCE=0
File Input Format Counters
Bytes Read=38
File Output Format Counters
Bytes Written=27
</pre>

作业在8088界面上显示情况如下：



wordcount的执行结果，也正确，至此，我们的eclipse调试hadoop2.2分布式集群，已经成功了，大家可以去试一试了。