hadoop:hadoop2.2 ,windows myeclipse环境;
Eclipse调用hadoop运行MR程序其实就是普通的java程序可以提交MR任务到集群执行而已。在Hadoop1中,只需指定jt(jobtracker)和fs(namenode)即可,一般如下:
- Configuration conf = new Configuration();
- conf.set("mapred.job.tracker", "192.168.128.138:9001");
- conf.set("fs.default.name","192.168.128.138:9000");
上面的代码在hadoop1中运行是ok的,完全可以使用java提交任务到集群运行。但是,hadoop2却是没有了jt,新增了yarn。这个要如何使用呢?最简单的想法,同样指定其配置,试试。
- Configuration conf = new YarnConfiguration();
- conf.set("fs.defaultFS", "hdfs://node31:9000");
- conf.set("mapreduce.framework.name", "yarn");
- conf.set("yarn.resourcemanager.address", "node31:8032");
恩,这样配置后,可以运行,首先是下面的错误:
- 2014-04-03 21:20:21,568 ERROR [main] util.Shell (Shell.java:getWinUtilsPath(303)) - Failed to locate the winutils binary in the hadoop binary path
- java.io.IOException: Could not locate executable null\bin\winutils.exe in the Hadoop binaries.
- at org.apache.hadoop.util.Shell.getQualifiedBinPath(Shell.java:278)
- at org.apache.hadoop.util.Shell.getWinUtilsPath(Shell.java:300)
- at org.apache.hadoop.util.Shell.<clinit>(Shell.java:293)
- at org.apache.hadoop.util.StringUtils.<clinit>(StringUtils.java:76)
- at org.apache.hadoop.yarn.conf.YarnConfiguration.<clinit>(YarnConfiguration.java:345)
- at org.fansy.hadoop.mr.WordCount.getConf(WordCount.java:104)
- at org.fansy.hadoop.mr.WordCount.runJob(WordCount.java:84)
- at org.fansy.hadoop.mr.WordCount.main(WordCount.java:47)
这个错误不用管,这个好像是windows调用的时候就会出的错误。
然后是什么权限问题之类的,这个时候就需要去调整下权限,至少我目前是这样做的。调整的权限主要有/tmp 以及运行wordcount的输入、输出目录。命令如下: $HADOOP_HOME/bin/hadoop fs -chmod -R 777 /tmp 。
然后直到你出现了下面的错误,那么,好了,可以说你已经成功了一半了。
- 2014-04-03 20:32:36,596 ERROR [main] security.UserGroupInformation (UserGroupInformation.java:doAs(1494)) - PriviledgedActionException as:Administrator (auth:SIMPLE) cause:java.io.IOException: Failed to run job : Application application_1396459813671_0001 failed 2 times due to AM Container for appattempt_1396459813671_0001_000002 exited with exitCode: 1 due to: Exception from container-launch:
- org.apache.hadoop.util.Shell$ExitCodeException: /bin/bash: line 0: fg: no job control
- at org.apache.hadoop.util.Shell.runCommand(Shell.java:464)
- at org.apache.hadoop.util.Shell.run(Shell.java:379)
- at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:589)
- at org.apache.hadoop.yarn.server.nodemanager.DefaultContainerExecutor.launchContainer(DefaultContainerExecutor.java:195)
- at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:283)
- at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.ContainerLaunch.call(ContainerLaunch.java:79)
- at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
- at java.util.concurrent.FutureTask.run(FutureTask.java:166)
- 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:724)
- .Failing this attempt.. Failing the application.
用上面出现的错误去google,可以得到这个网页:https://issues.apache.org/jira/browse/MAPREDUCE-5655 。 恩,对的。这个网页就是我们的solution。
我们分为1、2、3步骤吧。
1. 修改MRapps.java 、YARNRunner.java的源码,然后打包替换原来的jar包中的相应class文件,这两个jar我已经打包,可以在这里下载http://download.csdn.net/detail/fansy1990/7143547 。然后替换集群中相应的jar吧,同时需要注意替换Myeclipse中导入的包。额,说起Myeclipse中的jar包,这里还是先上幅jar包的图吧:
2. 修改mapred-default.xml ,添加:(这个只需在eclipse中导入的jar包修改即可,修改后的jar包不用上传到集群)
- <property>
- <name>mapred.remote.os</name>
- <value>Linux</value>
- <description>
- Remote MapReduce framework's OS, can be either Linux or Windows
- </description>
- </property>
(题外话,添加了这个属性后,按说我new一个Configuration后,我使用conf.get("mapred.remote.os")的时候应该是可以得到Linux的,但是我得到的却是null,这个就不清楚是怎么了。)
其文件在:
这时,你再运行程序,额好吧程序基本可以提交了,但是还是报错,查看log,可以看到下面的错误:
- Error: Could not find or load main class org.apache.hadoop.mapreduce.v2.app.MRAppMaster
额,说了这么久,还是把我的wordcount程序贴出来吧:
- package org.fansy.hadoop.mr;
- import java.io.IOException;
- import org.apache.hadoop.conf.Configuration;
- import org.apache.hadoop.fs.FileSystem;
- import org.apache.hadoop.fs.LocatedFileStatus;
- import org.apache.hadoop.fs.Path;
- import org.apache.hadoop.fs.RemoteIterator;
- import org.apache.hadoop.io.LongWritable;
- import org.apache.hadoop.io.Text;
- import org.apache.hadoop.mapred.ClusterStatus;
- import org.apache.hadoop.mapred.JobClient;
- 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;
- import org.apache.hadoop.yarn.conf.YarnConfiguration;
- import org.slf4j.Logger;
- import org.slf4j.LoggerFactory;
- public class WordCount {
- private static Logger log = LoggerFactory.getLogger(WordCount.class);
- public static class WCMapper extends Mapper<LongWritable, Text, LongWritable, Text> {
- public void map(LongWritable key, Text value, Context cxt) throws IOException,InterruptedException {
- // String[] values= value.toString().split("[,| ]");
- cxt.write(key, value);
- }
- }
- public static class WCReducer extends Reducer<LongWritable, Text, LongWritable,Text> {
- public void reduce(LongWritable key, Iterable<Text> values, Context cxt) throws IOException,InterruptedException {
- StringBuffer buff = new StringBuffer();
- for (Text v:values) {
- buff.append(v.toString()+"\t");
- }
- cxt.write(key, new Text(buff.toString()));
- }
- }
- public static void main(String[] args) throws Exception {
- // checkFS();
- String input ="hdfs://node31:9000/input/test.dat";
- String output="hdfs://node31:9000/output/wc003";
- runJob(input,output);
- // runJob(args[0],args[1]);
- // upload();
- }
- /**
- * test operate the hdfs
- * @throws IOException
- */
- public static void checkFS() throws IOException{
- Configuration conf=getConf();
- Path f= new Path("/user");
- FileSystem fs = FileSystem.get(f.toUri(),conf);
- RemoteIterator<LocatedFileStatus> paths=fs.listFiles(f, true);
- while(paths.hasNext()){
- System.out.println(paths.next());
- }
- }
- public static void upload() throws IOException{
- Configuration conf = getConf();
- Path f= new Path("d:\\wordcount.jar");
- FileSystem fs = FileSystem.get(f.toUri(),conf);
- fs.copyFromLocalFile(true, f, new Path("/input/wordcount.jar"));
- System.out.println("done ...");
- }
- /**
- * test the job submit
- * @throws IOException
- * @throws InterruptedException
- * @throws ClassNotFoundException
- */
- public static void runJob(String input,String output) throws IOException, ClassNotFoundException, InterruptedException{
- Configuration conf=getConf();
- Job job = new Job(conf,"word count");
- // job.setJar("hdfs://node31:9000/input/wordcount.jar");
- job.setJobName("wordcount");
- job.setJarByClass(WordCount.class);
- // job.setOutputFormatClass(SequenceFileOutputFormat.class);
- job.setOutputKeyClass(LongWritable.class);
- job.setOutputValueClass(Text.class);
- job.setMapperClass(WCMapper.class);
- job.setCombinerClass(WCReducer.class);
- job.setReducerClass(WCReducer.class);
- FileInputFormat.addInputPath(job, new Path(input));
- // SequenceFileOutputFormat.setOutputPath(job, new Path(args[1]));
- FileOutputFormat.setOutputPath(job, new Path(output));
- System.exit(job.waitForCompletion(true)?0:1);
- }
- private static Configuration getConf() throws IOException{
- Configuration conf = new YarnConfiguration();
- conf.set("fs.defaultFS", "hdfs://node31:9000");
- conf.set("mapreduce.framework.name", "yarn");
- conf.set("yarn.resourcemanager.address", "node31:8032");
- // conf.set("mapred.remote.os", "Linux");
- System.out.println(conf.get("mapred.remote.os"));
- // JobClient client = new JobClient(conf);
- // ClusterStatus cluster = client.getClusterStatus();
- return conf;
- }
- }
3. 如何修复上面的报错?按照那个链接的solution,需要修改mapred-default.xml 和yarn-default.xml ,其中mapred-default.xml刚才已经修改过了,这次再次修改,添加:
- <property>
- <name>mapreduce.application.classpath</name>
- <value>
- $HADOOP_CONF_DIR,
- $HADOOP_COMMON_HOME/share/hadoop/common/*,
- $HADOOP_COMMON_HOME/share/hadoop/common/lib/*,
- $HADOOP_HDFS_HOME/share/hadoop/hdfs/*,
- $HADOOP_HDFS_HOME/share/hadoop/hdfs/lib/*,
- $HADOOP_MAPRED_HOME/share/hadoop/mapreduce/*,
- $HADOOP_MAPRED_HOME/share/hadoop/mapreduce/lib/*,
- $HADOOP_YARN_HOME/share/hadoop/yarn/*,
- $HADOOP_YARN_HOME/share/hadoop/yarn/lib/*
- </value>
- </property>
对于yarn-default.xml也是同样的修改,其在hadoop-yarn-common-2.2.0.jar包中,修改内容如下:
- <property>
- <name>mapreduce.application.classpath</name>
- <value>
- $HADOOP_CONF_DIR,
- $HADOOP_COMMON_HOME/share/hadoop/common/*,
- $HADOOP_COMMON_HOME/share/hadoop/common/lib/*,
- $HADOOP_HDFS_HOME/share/hadoop/hdfs/*,
- $HADOOP_HDFS_HOME/share/hadoop/hdfs/lib/*,
- $HADOOP_MAPRED_HOME/share/hadoop/mapreduce/*,
- $HADOOP_MAPRED_HOME/share/hadoop/mapreduce/lib/*,
- $HADOOP_YARN_HOME/share/hadoop/yarn/*,
- $HADOOP_YARN_HOME/share/hadoop/yarn/lib/*
- </value>
- </property>
同样的,上面两个jar包只用替换myeclipse中的jar包即可,不需要替换集群中的。
4. 经过上面的替换,然后再次运行,出现下面的错误:
- Caused by: java.lang.ClassNotFoundException: Class org.fansy.hadoop.mr.WordCount$WCMapper not found
- at org.apache.hadoop.conf.Configuration.getClassByName(Configuration.java:1626)
- at org.apache.hadoop.conf.Configuration.getClass(Configuration.java:1718)
- ... 8 more
额,好吧,我应该不用多少了,这样的错误,应该已经说明我们的myeclipse可以提交任务到hadoop2了,并且可以运行了。好吧最后一步,上传我们打包的wordcount程序的jar文件到$HADOOP_HOME/share/hadoop/mapreduce/lib下面,然后再次运行。(这里上传后不用重启集群)呵呵,最后得到下面的结果:
- 2014-04-03 21:17:34,289 ERROR [main] util.Shell (Shell.java:getWinUtilsPath(303)) - Failed to locate the winutils binary in the hadoop binary path
- java.io.IOException: Could not locate executable null\bin\winutils.exe in the Hadoop binaries.
- at org.apache.hadoop.util.Shell.getQualifiedBinPath(Shell.java:278)
- at org.apache.hadoop.util.Shell.getWinUtilsPath(Shell.java:300)
- at org.apache.hadoop.util.Shell.<clinit>(Shell.java:293)
- at org.apache.hadoop.util.StringUtils.<clinit>(StringUtils.java:76)
- at org.apache.hadoop.yarn.conf.YarnConfiguration.<clinit>(YarnConfiguration.java:345)
- at org.fansy.hadoop.mr.WordCount.getConf(WordCount.java:104)
- at org.fansy.hadoop.mr.WordCount.runJob(WordCount.java:84)
- at org.fansy.hadoop.mr.WordCount.main(WordCount.java:47)
- Linux
- 2014-04-03 21:18:19,853 WARN [main] util.NativeCodeLoader (NativeCodeLoader.java:<clinit>(62)) - Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
- 2014-04-03 21:18:20,499 INFO [main] client.RMProxy (RMProxy.java:createRMProxy(56)) - Connecting to ResourceManager at node31/192.168.0.31:8032
- 2014-04-03 21:18:20,973 WARN [main] mapreduce.JobSubmitter (JobSubmitter.java:copyAndConfigureFiles(149)) - Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
- 2014-04-03 21:18:21,020 INFO [main] input.FileInputFormat (FileInputFormat.java:listStatus(287)) - Total input paths to process : 1
- 2014-04-03 21:18:21,313 INFO [main] mapreduce.JobSubmitter (JobSubmitter.java:submitJobInternal(394)) - number of splits:1
- 2014-04-03 21:18:21,336 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - user.name is deprecated. Instead, use mapreduce.job.user.name
- 2014-04-03 21:18:21,337 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.jar is deprecated. Instead, use mapreduce.job.jar
- 2014-04-03 21:18:21,337 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - fs.default.name is deprecated. Instead, use fs.defaultFS
- 2014-04-03 21:18:21,338 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.output.value.class is deprecated. Instead, use mapreduce.job.output.value.class
- 2014-04-03 21:18:21,338 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapreduce.combine.class is deprecated. Instead, use mapreduce.job.combine.class
- 2014-04-03 21:18:21,339 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapreduce.map.class is deprecated. Instead, use mapreduce.job.map.class
- 2014-04-03 21:18:21,339 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.job.name is deprecated. Instead, use mapreduce.job.name
- 2014-04-03 21:18:21,339 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapreduce.reduce.class is deprecated. Instead, use mapreduce.job.reduce.class
- 2014-04-03 21:18:21,340 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.input.dir is deprecated. Instead, use mapreduce.input.fileinputformat.inputdir
- 2014-04-03 21:18:21,340 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.output.dir is deprecated. Instead, use mapreduce.output.fileoutputformat.outputdir
- 2014-04-03 21:18:21,342 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.map.tasks is deprecated. Instead, use mapreduce.job.maps
- 2014-04-03 21:18:21,343 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.output.key.class is deprecated. Instead, use mapreduce.job.output.key.class
- 2014-04-03 21:18:21,343 INFO [main] Configuration.deprecation (Configuration.java:warnOnceIfDeprecated(840)) - mapred.working.dir is deprecated. Instead, use mapreduce.job.working.dir
- 2014-04-03 21:18:21,513 INFO [main] mapreduce.JobSubmitter (JobSubmitter.java:printTokens(477)) - Submitting tokens for job: job_1396463733942_0003
- 2014-04-03 21:18:21,817 INFO [main] impl.YarnClientImpl (YarnClientImpl.java:submitApplication(174)) - Submitted application application_1396463733942_0003 to ResourceManager at node31/192.168.0.31:8032
- 2014-04-03 21:18:21,859 INFO [main] mapreduce.Job (Job.java:submit(1272)) - The url to track the job: http://node31:8088/proxy/application_1396463733942_0003/
- 2014-04-03 21:18:21,860 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1317)) - Running job: job_1396463733942_0003
- 2014-04-03 21:18:31,307 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1338)) - Job job_1396463733942_0003 running in uber mode : false
- 2014-04-03 21:18:31,311 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1345)) - map 0% reduce 0%
- 2014-04-03 21:19:02,346 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1345)) - map 100% reduce 0%
- 2014-04-03 21:19:11,416 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1345)) - map 100% reduce 100%
- 2014-04-03 21:19:11,425 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1356)) - Job job_1396463733942_0003 completed successfully
- 2014-04-03 21:19:11,552 INFO [main] mapreduce.Job (Job.java:monitorAndPrintJob(1363)) - Counters: 43
- File System Counters
- FILE: Number of bytes read=11139
- FILE: Number of bytes written=182249
- FILE: Number of read operations=0
- FILE: Number of large read operations=0
- FILE: Number of write operations=0
- HDFS: Number of bytes read=8646
- HDFS: Number of bytes written=10161
- 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)=29330
- Total time spent by all reduces in occupied slots (ms)=5825
- Map-Reduce Framework
- Map input records=235
- Map output records=235
- Map output bytes=10428
- Map output materialized bytes=11139
- Input split bytes=98
- Combine input records=235
- Combine output records=235
- Reduce input groups=235
- Reduce shuffle bytes=11139
- Reduce input records=235
- Reduce output records=235
- Spilled Records=470
- Shuffled Maps =1
- Failed Shuffles=0
- Merged Map outputs=1
- GC time elapsed (ms)=124
- CPU time spent (ms)=21920
- Physical memory (bytes) snapshot=299376640
- Virtual memory (bytes) snapshot=1671372800
- Total committed heap usage (bytes)=152834048
- 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=8548
- File Output Format Counters
- Bytes Written=10161
上面你看到Linux,是因为我使用了conf.set("mapred.remote.os", "Linux"); 不过在实际运行的时候却不需要设置。
另外,如果是linux系统部署的tomcat调用hadoop2集群运行MR程序的话,应该不需要替换其jar吧的,这个还有待验证。
哈,总算搞定了。这个问题也算是困扰了我好久了,期间几次想要冲破,结果都是无果而归,甚是郁闷。额,其实这个也不算是原创了,哎,国外在02/Dec/13 18:35这个时间点就搞定了。不过,我搜了好久,都没有中文的相关介绍。(如果有的话,那就是我搜索能力的问题了,居然没有搜到,哎)。
分享,成长,快乐
转载请注明blog地址:http://blog.csdn.net/fansy1990