搭建Hadoop2.6.0+Eclipse开发调试环境

上一篇在win7虚拟机下搭建了hadoop2.6.0伪分布式环境。为了开发调试方便，本文介绍在eclipse下搭建开发环境，连接和提交任务到hadoop集群。

1. 环境

Eclipse版本Luna 4.4.1

安装插件hadoop-eclipse-plugin-2.6.0.jar，下载后放到eclipse/plugins目录即可。

2. 配置插件

2.1 配置hadoop主目录

解压缩hadoop-2.6.0.tar.gz到C:Downloadshadoop-2.6.0，在eclipse的Windows->Preferences的Hadoop Map/Reduce中设置安装目录。

2.2 配置插件

打开Windows->Open Perspective中的Map/Reduce，在此perspective下进行hadoop程序开发。

    

打开Windows->Show View中的Map/Reduce Locations，如下图右键选择New Hadoop location…新建hadoop连接。

确认完成以后如下，eclipse会连接hadoop集群。

如果连接成功，在project explorer的DFS Locations下会展现hdfs集群中的文件。

3. 开发hadoop程序

3.1 程序开发

开发一个Sort示例，对输入整数进行排序。输入文件格式是每行一个整数。

package com.ccb;

/**
 * Created by hp on 2015-7-20.
 */

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.Text;
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.output.FileOutputFormat;

public class Sort {

    // 每行记录是一个整数。将Text文本转换为IntWritable类型，作为map的key
    public static class Map extends Mapper<Object, Text, IntWritable, IntWritable> {
        private static IntWritable data = new IntWritable();

        // 实现map函数
        public void map(Object key, Text value, Context context) throws IOException, InterruptedException {
            String line = value.toString();
            data.set(Integer.parseInt(line));
            context.write(data, new IntWritable(1));
        }
    }

    // reduce之前hadoop框架会进行shuffle和排序，因此直接输出key即可。
    public static class Reduce extends Reducer<IntWritable, IntWritable, IntWritable, Text> {

        //实现reduce函数
        public void reduce(IntWritable key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
            for (IntWritable v : values) {
                context.write(key, new Text(""));
            }
        }
    }

    public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();

        // 指定JobTracker地址
        conf.set("mapred.job.tracker", "192.168.62.129:9001");
        if (args.length != 2) {
            System.err.println("Usage: Data Sort <in> <out>");
            System.exit(2);
        }
        System.out.println(args[0]);
        System.out.println(args[1]);

        Job job = Job.getInstance(conf, "Data Sort");
        job.setJarByClass(Sort.class);

        //设置Map和Reduce处理类
        job.setMapperClass(Map.class);
        job.setReducerClass(Reduce.class);

        //设置输出类型
        job.setOutputKeyClass(IntWritable.class);
        job.setOutputValueClass(IntWritable.class);

        //设置输入和输出目录
        FileInputFormat.addInputPath(job, new Path(args[0]));
        FileOutputFormat.setOutputPath(job, new Path(args[1]));
        System.exit(job.waitForCompletion(true) ? 0 : 1);
    }
}

3.2 配置文件

把log4j.properties和hadoop集群中的core-site.xml加入到classpath中。我的示例工程是maven组织，因此放到src/main/resources目录。

程序执行时会从core-site.xml中获取hdfs地址。

3.3 程序执行

右键选择Run As -> Run Configurations…，在参数中填好输入输出目录，执行Run即可。

 执行日志：

hdfs://192.168.62.129:9000/user/vm/sort_in
hdfs://192.168.62.129:9000/user/vm/sort_out
15/07/27 16:21:36 INFO Configuration.deprecation: session.id is deprecated. Instead, use dfs.metrics.session-id
15/07/27 16:21:36 INFO jvm.JvmMetrics: Initializing JVM Metrics with processName=JobTracker, sessionId=
15/07/27 16:21:36 WARN mapreduce.JobSubmitter: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
15/07/27 16:21:36 WARN mapreduce.JobSubmitter: No job jar file set.  User classes may not be found. See Job or Job#setJar(String).
15/07/27 16:21:36 INFO input.FileInputFormat: Total input paths to process : 3
15/07/27 16:21:36 INFO mapreduce.JobSubmitter: number of splits:3
15/07/27 16:21:36 INFO Configuration.deprecation: mapred.job.tracker is deprecated. Instead, use mapreduce.jobtracker.address
15/07/27 16:21:37 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_local1592166400_0001
15/07/27 16:21:37 INFO mapreduce.Job: The url to track the job: http://localhost:8080/
15/07/27 16:21:37 INFO mapreduce.Job: Running job: job_local1592166400_0001
15/07/27 16:21:37 INFO mapred.LocalJobRunner: OutputCommitter set in config null
15/07/27 16:21:37 INFO mapred.LocalJobRunner: OutputCommitter is org.apache.hadoop.mapreduce.lib.output.FileOutputCommitter
15/07/27 16:21:37 INFO mapred.LocalJobRunner: Waiting for map tasks
15/07/27 16:21:37 INFO mapred.LocalJobRunner: Starting task: attempt_local1592166400_0001_m_000000_0
15/07/27 16:21:37 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
15/07/27 16:21:37 INFO mapred.Task:  Using ResourceCalculatorProcessTree : org.apache.hadoop.yarn.util.WindowsBasedProcessTree@4c90dbc4
15/07/27 16:21:37 INFO mapred.MapTask: Processing split: hdfs://192.168.62.129:9000/user/vm/sort_in/file1:0+25
15/07/27 16:21:37 INFO mapred.MapTask: (EQUATOR) 0 kvi 26214396(104857584)
15/07/27 16:21:37 INFO mapred.MapTask: mapreduce.task.io.sort.mb: 100
15/07/27 16:21:37 INFO mapred.MapTask: soft limit at 83886080
15/07/27 16:21:37 INFO mapred.MapTask: bufstart = 0; bufvoid = 104857600
15/07/27 16:21:37 INFO mapred.MapTask: kvstart = 26214396; length = 6553600
15/07/27 16:21:37 INFO mapred.MapTask: Map output collector class = org.apache.hadoop.mapred.MapTask$MapOutputBuffer
15/07/27 16:21:38 INFO mapred.LocalJobRunner: 
15/07/27 16:21:38 INFO mapred.MapTask: Starting flush of map output
15/07/27 16:21:38 INFO mapred.MapTask: Spilling map output
15/07/27 16:21:38 INFO mapred.MapTask: bufstart = 0; bufend = 56; bufvoid = 104857600
15/07/27 16:21:38 INFO mapred.MapTask: kvstart = 26214396(104857584); kvend = 26214372(104857488); length = 25/6553600
15/07/27 16:21:38 INFO mapred.MapTask: Finished spill 0
15/07/27 16:21:38 INFO mapred.Task: Task:attempt_local1592166400_0001_m_000000_0 is done. And is in the process of committing
15/07/27 16:21:38 INFO mapred.LocalJobRunner: map
15/07/27 16:21:38 INFO mapred.Task: Task 'attempt_local1592166400_0001_m_000000_0' done.
15/07/27 16:21:38 INFO mapred.LocalJobRunner: Finishing task: attempt_local1592166400_0001_m_000000_0
15/07/27 16:21:38 INFO mapred.LocalJobRunner: Starting task: attempt_local1592166400_0001_m_000001_0
15/07/27 16:21:38 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
15/07/27 16:21:38 INFO mapred.Task:  Using ResourceCalculatorProcessTree : org.apache.hadoop.yarn.util.WindowsBasedProcessTree@69e4d7d
15/07/27 16:21:38 INFO mapred.MapTask: Processing split: hdfs://192.168.62.129:9000/user/vm/sort_in/file2:0+15
15/07/27 16:21:38 INFO mapred.MapTask: (EQUATOR) 0 kvi 26214396(104857584)
15/07/27 16:21:38 INFO mapred.MapTask: mapreduce.task.io.sort.mb: 100
15/07/27 16:21:38 INFO mapred.MapTask: soft limit at 83886080
15/07/27 16:21:38 INFO mapred.MapTask: bufstart = 0; bufvoid = 104857600
15/07/27 16:21:38 INFO mapred.MapTask: kvstart = 26214396; length = 6553600
15/07/27 16:21:38 INFO mapred.MapTask: Map output collector class = org.apache.hadoop.mapred.MapTask$MapOutputBuffer
15/07/27 16:21:38 INFO mapred.LocalJobRunner: 
15/07/27 16:21:38 INFO mapred.MapTask: Starting flush of map output
15/07/27 16:21:38 INFO mapred.MapTask: Spilling map output
15/07/27 16:21:38 INFO mapred.MapTask: bufstart = 0; bufend = 32; bufvoid = 104857600
15/07/27 16:21:38 INFO mapred.MapTask: kvstart = 26214396(104857584); kvend = 26214384(104857536); length = 13/6553600
15/07/27 16:21:38 INFO mapred.MapTask: Finished spill 0
15/07/27 16:21:38 INFO mapred.Task: Task:attempt_local1592166400_0001_m_000001_0 is done. And is in the process of committing
15/07/27 16:21:38 INFO mapred.LocalJobRunner: map
15/07/27 16:21:38 INFO mapred.Task: Task 'attempt_local1592166400_0001_m_000001_0' done.
15/07/27 16:21:38 INFO mapred.LocalJobRunner: Finishing task: attempt_local1592166400_0001_m_000001_0
15/07/27 16:21:38 INFO mapred.LocalJobRunner: Starting task: attempt_local1592166400_0001_m_000002_0
15/07/27 16:21:38 INFO mapreduce.Job: Job job_local1592166400_0001 running in uber mode : false
15/07/27 16:21:38 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
15/07/27 16:21:38 INFO mapreduce.Job:  map 100% reduce 0%
15/07/27 16:21:38 INFO mapred.Task:  Using ResourceCalculatorProcessTree : org.apache.hadoop.yarn.util.WindowsBasedProcessTree@4e931efa
15/07/27 16:21:38 INFO mapred.MapTask: Processing split: hdfs://192.168.62.129:9000/user/vm/sort_in/file3:0+8
15/07/27 16:21:39 INFO mapred.MapTask: (EQUATOR) 0 kvi 26214396(104857584)
15/07/27 16:21:39 INFO mapred.MapTask: mapreduce.task.io.sort.mb: 100
15/07/27 16:21:39 INFO mapred.MapTask: soft limit at 83886080
15/07/27 16:21:39 INFO mapred.MapTask: bufstart = 0; bufvoid = 104857600
15/07/27 16:21:39 INFO mapred.MapTask: kvstart = 26214396; length = 6553600
15/07/27 16:21:39 INFO mapred.MapTask: Map output collector class = org.apache.hadoop.mapred.MapTask$MapOutputBuffer
15/07/27 16:21:39 INFO mapred.LocalJobRunner: 
15/07/27 16:21:39 INFO mapred.MapTask: Starting flush of map output
15/07/27 16:21:39 INFO mapred.MapTask: Spilling map output
15/07/27 16:21:39 INFO mapred.MapTask: bufstart = 0; bufend = 24; bufvoid = 104857600
15/07/27 16:21:39 INFO mapred.MapTask: kvstart = 26214396(104857584); kvend = 26214388(104857552); length = 9/6553600
15/07/27 16:21:39 INFO mapred.MapTask: Finished spill 0
15/07/27 16:21:39 INFO mapred.Task: Task:attempt_local1592166400_0001_m_000002_0 is done. And is in the process of committing
15/07/27 16:21:39 INFO mapred.LocalJobRunner: map
15/07/27 16:21:39 INFO mapred.Task: Task 'attempt_local1592166400_0001_m_000002_0' done.
15/07/27 16:21:39 INFO mapred.LocalJobRunner: Finishing task: attempt_local1592166400_0001_m_000002_0
15/07/27 16:21:39 INFO mapred.LocalJobRunner: map task executor complete.
15/07/27 16:21:39 INFO mapred.LocalJobRunner: Waiting for reduce tasks
15/07/27 16:21:39 INFO mapred.LocalJobRunner: Starting task: attempt_local1592166400_0001_r_000000_0
15/07/27 16:21:39 INFO util.ProcfsBasedProcessTree: ProcfsBasedProcessTree currently is supported only on Linux.
15/07/27 16:21:39 INFO mapred.Task:  Using ResourceCalculatorProcessTree : org.apache.hadoop.yarn.util.WindowsBasedProcessTree@49250068
15/07/27 16:21:39 INFO mapred.ReduceTask: Using ShuffleConsumerPlugin: org.apache.hadoop.mapreduce.task.reduce.Shuffle@2129404b
15/07/27 16:21:39 INFO reduce.MergeManagerImpl: MergerManager: memoryLimit=652528832, maxSingleShuffleLimit=163132208, mergeThreshold=430669056, ioSortFactor=10, memToMemMergeOutputsThreshold=10
15/07/27 16:21:39 INFO reduce.EventFetcher: attempt_local1592166400_0001_r_000000_0 Thread started: EventFetcher for fetching Map Completion Events
15/07/27 16:21:40 INFO reduce.LocalFetcher: localfetcher#1 about to shuffle output of map attempt_local1592166400_0001_m_000002_0 decomp: 32 len: 36 to MEMORY
15/07/27 16:21:40 INFO reduce.InMemoryMapOutput: Read 32 bytes from map-output for attempt_local1592166400_0001_m_000002_0
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: closeInMemoryFile -> map-output of size: 32, inMemoryMapOutputs.size() -> 1, commitMemory -> 0, usedMemory ->32
15/07/27 16:21:40 INFO reduce.LocalFetcher: localfetcher#1 about to shuffle output of map attempt_local1592166400_0001_m_000000_0 decomp: 72 len: 76 to MEMORY
15/07/27 16:21:40 INFO reduce.InMemoryMapOutput: Read 72 bytes from map-output for attempt_local1592166400_0001_m_000000_0
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: closeInMemoryFile -> map-output of size: 72, inMemoryMapOutputs.size() -> 2, commitMemory -> 32, usedMemory ->104
15/07/27 16:21:40 INFO reduce.LocalFetcher: localfetcher#1 about to shuffle output of map attempt_local1592166400_0001_m_000001_0 decomp: 42 len: 46 to MEMORY
15/07/27 16:21:40 INFO reduce.InMemoryMapOutput: Read 42 bytes from map-output for attempt_local1592166400_0001_m_000001_0
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: closeInMemoryFile -> map-output of size: 42, inMemoryMapOutputs.size() -> 3, commitMemory -> 104, usedMemory ->146
15/07/27 16:21:40 INFO reduce.EventFetcher: EventFetcher is interrupted.. Returning
15/07/27 16:21:40 INFO mapred.LocalJobRunner: 3 / 3 copied.
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: finalMerge called with 3 in-memory map-outputs and 0 on-disk map-outputs
15/07/27 16:21:40 INFO mapred.Merger: Merging 3 sorted segments
15/07/27 16:21:40 INFO mapred.Merger: Down to the last merge-pass, with 3 segments left of total size: 128 bytes
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: Merged 3 segments, 146 bytes to disk to satisfy reduce memory limit
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: Merging 1 files, 146 bytes from disk
15/07/27 16:21:40 INFO reduce.MergeManagerImpl: Merging 0 segments, 0 bytes from memory into reduce
15/07/27 16:21:40 INFO mapred.Merger: Merging 1 sorted segments
15/07/27 16:21:40 INFO mapred.Merger: Down to the last merge-pass, with 1 segments left of total size: 136 bytes
15/07/27 16:21:40 INFO mapred.LocalJobRunner: 3 / 3 copied.
15/07/27 16:21:40 INFO Configuration.deprecation: mapred.skip.on is deprecated. Instead, use mapreduce.job.skiprecords
15/07/27 16:21:40 INFO mapred.Task: Task:attempt_local1592166400_0001_r_000000_0 is done. And is in the process of committing
15/07/27 16:21:40 INFO mapred.LocalJobRunner: 3 / 3 copied.
15/07/27 16:21:40 INFO mapred.Task: Task attempt_local1592166400_0001_r_000000_0 is allowed to commit now
15/07/27 16:21:40 INFO output.FileOutputCommitter: Saved output of task 'attempt_local1592166400_0001_r_000000_0' to hdfs://192.168.62.129:9000/user/vm/sort_out/_temporary/0/task_local1592166400_0001_r_000000
15/07/27 16:21:40 INFO mapred.LocalJobRunner: reduce > reduce
15/07/27 16:21:40 INFO mapred.Task: Task 'attempt_local1592166400_0001_r_000000_0' done.
15/07/27 16:21:40 INFO mapred.LocalJobRunner: Finishing task: attempt_local1592166400_0001_r_000000_0
15/07/27 16:21:40 INFO mapred.LocalJobRunner: reduce task executor complete.
15/07/27 16:21:40 INFO mapreduce.Job:  map 100% reduce 100%
15/07/27 16:21:41 INFO mapreduce.Job: Job job_local1592166400_0001 completed successfully
15/07/27 16:21:41 INFO mapreduce.Job: Counters: 38
    File System Counters
        FILE: Number of bytes read=3834
        FILE: Number of bytes written=1017600
        FILE: Number of read operations=0
        FILE: Number of large read operations=0
        FILE: Number of write operations=0
        HDFS: Number of bytes read=161
        HDFS: Number of bytes written=62
        HDFS: Number of read operations=41
        HDFS: Number of large read operations=0
        HDFS: Number of write operations=10
    Map-Reduce Framework
        Map input records=14
        Map output records=14
        Map output bytes=112
        Map output materialized bytes=158
        Input split bytes=339
        Combine input records=0
        Combine output records=0
        Reduce input groups=13
        Reduce shuffle bytes=158
        Reduce input records=14
        Reduce output records=14
        Spilled Records=28
        Shuffled Maps =3
        Failed Shuffles=0
        Merged Map outputs=3
        GC time elapsed (ms)=10
        CPU time spent (ms)=0
        Physical memory (bytes) snapshot=0
        Virtual memory (bytes) snapshot=0
        Total committed heap usage (bytes)=1420296192
    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=48
    File Output Format Counters 
        Bytes Written=62

4. 可能出现的问题

4.1 权限问题，无法访问HDFS

修改集群hdfs-site.xml配置，关闭hadoop集群的权限校验。

<property>

<name>dfs.permissions</name>

<value>false</value>

</property>

4.2 出现NullPointerException异常

在环境变量中配置%HADOOP_HOME%为C:Downloadhadoop-2.6.0

下载winutils.exe和hadoop.dll到C:Downloadhadoop-2.6.0in

注意：网上很多资料说的是下载hadoop-common-2.2.0-bin-master.zip，但很多不支持hadoop2.6.0版本。需要下载支持hadoop2.6.0版本的程序。

4.3 程序执行失败

需要执行Run on Hadoop，而不是Java Application。