我正在尝试编写一个 Java 文件来接收 MapReduce 作业的源代码，动态编译它并在 Hadoop 集群上运行该作业。为此，我编写了 3 个方法，称为 compile()、makeJAR() 和 run_Hadoop_Job()。JAR 文件的编译和创建一切正常。但是，当作业提交到 Hadoop 时，一旦作业开始，它就会面临查找所需 Mapper/Reducer 类的问题，并为 Mapper_Class 和 Reducer_Class 抛出 ClassNotFoundException *(java.lang.ClassNotFoundException: reza.rCloud.Mapper_Reducer_Classes $Mapper_Class.class)*. 我知道我引用所需的 Mapper/Reducer 类的方式应该有问题，但是在几次之后我无法弄清楚。非常感谢任何有关如何解决问题的帮助/建议。
关于项目的详细信息：我有一个名为“rCloud_test/src/reza/Mapper_Reducer_Classes.java”的文件，其中包含 Mapper_Class 和 Reducer_Class 的源代码。该文件最终在运行时收到，但现在我将 Hadoop WordCount 示例复制到其中，并将其本地存储在与我的主类文件相同的文件夹中：rCloud_test/src/reza/Platform2.java。
下面你可以看到 Platform2.java 的 main() 方法，这是我这个项目的主要类：
public static void main(String[] args){
System.out.println("Code Execution Started");
String className = "Mapper_Reducer_Classes";
Platform2 myPlatform = new Platform2();
//step 1: compile the received class file dynamically:
boolean compResult = myPlatform.compile(className);
System.out.println(className + ".java compilation result: "+compResult);
//step 2: make a JAR file out of the compiled file:
if (compResult) {
compResult = myPlatform.makeJAR("jar_file", myPlatform.compilation_Output_Folder);
System.out.println("JAR creation result: "+compResult);
}
//step 3: Now let's run the Hadoop job:
if (compResult) {
compResult = myPlatform.run_Hadoop_Job(className);
System.out.println("Running on Hadoop result: "+compResult);
}
导致我所有问题的方法是 run_Hadoop_Job() ，如下所示：
private boolean run_Hadoop_Job(String className){
try{
System.out.println("*Starting to run the code on Hadoop...");
String[] argsTemp = { "project_test/input", "project_test/output" };
Configuration conf = new Configuration();
conf.set("fs.default.name", "hdfs://localhost:54310");
conf.set("mapred.job.tracker", "localhost:54311");
conf.set("mapred.jar", jar_Output_Folder + "/jar_file"+".jar");
conf.set("libjars", required_Execution_Classes);
//THIS IS WHERE IT CAN'T FIND THE MENTIONED CLASSES, ALTHOUGH THEY EXIST BOTH ON DISK
// AND IN THE CREATED JAR FILE:??????
System.out.println("Getting Mapper/Reducer package name: " +
Mapper_Reducer_Classes.class.getName());
conf.set("mapreduce.map.class", "reza.rCloud.Mapper_Reducer_Classes$Mapper_Class");
conf.set("mapreduce.reduce.class", "reza.rCloud.Mapper_Reducer_Classes$Reducer_Class");
Job job = new Job(conf, "Hadoop Example for dynamically and programmatically compiling-running a job");
job.setJarByClass(Platform2.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(IntWritable.class);
FileInputFormat.addInputPath(job, new Path(argsTemp[0]));
FileSystem fs = FileSystem.get(conf);
Path out = new Path(argsTemp[1]);
fs.delete(out, true);
FileOutputFormat.setOutputPath(job, new Path(argsTemp[1]));
//job.submit();
System.out.println("*and now submitting the job to Hadoop...");
System.exit(job.waitForCompletion(true) ? 0 : 1);
System.out.println("Job Finished!");
} catch (Exception e) {
System.out.println("****************Exception!" );
e.printStackTrace();
return false;
}
return true;
}
如果需要，这里是 compile() 方法的源代码：
private boolean compile(String className) {
String fileToCompile = JOB_FOLDER + "/" +className+".java";
JavaCompiler compiler = ToolProvider.getSystemJavaCompiler();
FileOutputStream errorStream = null;
try{
errorStream = new FileOutputStream(JOB_FOLDER + "/logs/Errors.txt");
} catch(FileNotFoundException e){
//if problem creating the file, default wil be console
}
int compilationResult =
compiler.run( null, null, errorStream,
"-classpath", required_Compilation_Classes,
"-d", compilation_Output_Folder,
fileToCompile);
if (compilationResult == 0) {
//Compilation is successful:
return true;
} else {
//Compilation Failed:
return false;
}
}
以及 makeJAR() 方法的源代码：
private boolean makeJAR(String outputFileName, String inputDirectory) {
Manifest manifest = new Manifest();
manifest.getMainAttributes().put(Attributes.Name.MANIFEST_VERSION,
"1.0");
JarOutputStream target = null;
try {
target = new JarOutputStream(new FileOutputStream(
jar_Output_Folder+ "/"
+ outputFileName+".jar" ), manifest);
add(new File(inputDirectory), target);
} catch (Exception e) { return false; }
finally {
if (target != null)
try{
target.close();
} catch (Exception e) { return false; }
}
return true;
}
private void add(File source, JarOutputStream target) throws IOException
{
BufferedInputStream in = null;
try
{
if (source.isDirectory())
{
String name = source.getPath().replace("\\", "/");
if (!name.isEmpty())
{
if (!name.endsWith("/"))
name += "/";
JarEntry entry = new JarEntry(name);
entry.setTime(source.lastModified());
target.putNextEntry(entry);
target.closeEntry();
}
for (File nestedFile: source.listFiles())
add(nestedFile, target);
return;
}
JarEntry entry = new JarEntry(source.getPath().replace("\\", "/"));
entry.setTime(source.lastModified());
target.putNextEntry(entry);
in = new BufferedInputStream(new FileInputStream(source));
byte[] buffer = new byte[1024];
while (true)
{
int count = in.read(buffer);
if (count == -1)
break;
target.write(buffer, 0, count);
}
target.closeEntry();
}
finally
{
if (in != null)
in.close();
}
}
最后是用于访问文件的固定参数：
private String JOB_FOLDER = "/Users/reza/My_Software/rCloud_test/src/reza/rCloud";
private String HADOOP_SOURCE_FOLDER = "/Users/reza/My_Software/hadoop-0.20.2";
private String required_Compilation_Classes = HADOOP_SOURCE_FOLDER + "/hadoop-0.20.2-core.jar";
private String required_Execution_Classes = required_Compilation_Classes + "," +
"/Users/reza/My_Software/ActorFoundry_dist_ver/lib/commons-cli-1.1.jar," +
"/Users/reza/My_Software/ActorFoundry_dist_ver/lib/commons-logging-1.1.1.jar";
public String compilation_Output_Folder = "/Users/reza/My_Software/rCloud_test/dyn_classes";
private String jar_Output_Folder = "/Users/reza/My_Software/rCloud_test/dyn_jar";
运行 Platform2 后，磁盘上的项目结构如下：
rCloud_test/classes/reza/rCloud/Platform2.class：包含 Platform2 类 rCloud_test/dyn_classes/reza/rCloud/ 包含 Mapper_Reducer_Classes.class、Mapper_Reducer_Classes$Mapper_Class.class 和 Mapper_Reducer_Classes$Reducer_Class.class 的类。 jar 包含创建的 jar 文件
修订版：这是 rCloud_test/src/reza/rCloud/Mapper_Reducer_Classes.java 的源代码：
package reza.rCloud;
import java.io.IOException;
import java.lang.InterruptedException;
import java.util.StringTokenizer;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
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.util.GenericOptionsParser;
public class Mapper_Reducer_Classes {
/**
* The map class of WordCount.
*/
public static class Mapper_Class
extends Mapper<Object, Text, Text, IntWritable> {
private final static IntWritable one = new IntWritable(1);
private Text word = new Text();
public void map(Object key, Text value, Context context)
throws IOException, InterruptedException {
StringTokenizer itr = new StringTokenizer(value.toString());
while (itr.hasMoreTokens()) {
word.set(itr.nextToken());
context.write(word, one);
}
}
}
/**
* The reducer class of WordCount
*/
public static class Reducer_Class
extends Reducer<Text, IntWritable, Text, IntWritable> {
public void reduce(Text key, Iterable<IntWritable> values, Context context)
throws IOException, InterruptedException {
int sum = 0;
for (IntWritable value : values) {
sum += value.get();
}
context.write(key, new IntWritable(sum));
}
}
}