我有一个使用旧 API 运行的 hadoop 作业，我将我的实现移至新 API，但在运行它时遇到问题。当作业运行时，不会抛出任何异常，但我从未生成任何输出文件。在旧的 API 下，它会生成带有我排序的结果列表的输出文件。这是正在运行的作业:

Configuration config = new Configuration();
Job job = Job.getInstance(config, "sorting");

job.setOutputKeyClass(IntWritable.class);
job.setOutputValueClass(IntWritable.class);

job.setMapperClass(SortMapper.class);
job.setCombinerClass(SortReducer.class);
job.setReducerClass(SortReducer.class);

job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);

FileInputFormat.setInputPaths(job, new Path(inputFileLocation));
FileOutputFormat.setOutputPath(job, new Path(outputFileLocation));

job.setJarByClass(HadoopTest.class);

long startTime = System.currentTimeMillis();
job.submit();
long endTime = System.currentTimeMillis();

long duration = endTime - startTime;
System.out.println("Duration: " + duration);

这是我的映射器实现:

public static class SortMapper extends MultithreadedMapper<LongWritable, Text, IntWritable, IntWritable> {
    private final static IntWritable one = new IntWritable(1);
    private IntWritable intKey = new IntWritable();

    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        intKey.set(Integer.parseInt(value.toString()));
        context.write(intKey, one);
    }
}

这是我的 reducer 实现:

public static class SortReducer extends Reducer<IntWritable, IntWritable, IntWritable, IntWritable> {
    @Override
    protected void reduce(IntWritable key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
        int sum = 0;
        Iterator<IntWritable> iterator = values.iterator();
        while (iterator.hasNext()) {
            sum += iterator.next().get();
        }
        context.write(key, new IntWritable(sum));
    }
}

日志显示如下(当使用旧 API 运行时，我总是收到关于“无法加载领域映射信息...”和“无法加载 native-hadoop...”的投诉:

2014-03-18 10:19:41.299 java[13311:1d03] Unable to load realm mapping info from SCDynamicStore
14/03/18 10:19:41 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
14/03/18 10:19:41 INFO Configuration.deprecation: session.id is deprecated. Instead, use dfs.metrics.session-id
14/03/18 10:19:41 INFO jvm.JvmMetrics: Initializing JVM Metrics with processName=JobTracker, sessionId=
14/03/18 10:19:41 WARN mapreduce.JobSubmitter: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
14/03/18 10:19:41 WARN mapreduce.JobSubmitter: No job jar file set.  User classes may not be found. See Job or Job#setJar(String).
14/03/18 10:19:41 INFO input.FileInputFormat: Total input paths to process : 2
14/03/18 10:19:41 INFO mapreduce.JobSubmitter: number of splits:2
14/03/18 10:19:42 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_local904621238_0001
14/03/18 10:19:42 WARN conf.Configuration: file:/tmp/hadoop-james.mchugh/mapred/staging/james.mchugh904621238/.staging/job_local904621238_0001/job.xml:an attempt to override final parameter: mapreduce.job.end-notification.max.retry.interval;  Ignoring.
14/03/18 10:19:42 WARN conf.Configuration: file:/tmp/hadoop-james.mchugh/mapred/staging/james.mchugh904621238/.staging/job_local904621238_0001/job.xml:an attempt to override final parameter: mapreduce.job.end-notification.max.attempts;  Ignoring.
14/03/18 10:19:42 WARN conf.Configuration: file:/tmp/hadoop-james.mchugh/mapred/local/localRunner/james.mchugh/job_local904621238_0001/job_local904621238_0001.xml:an attempt to override final parameter: mapreduce.job.end-notification.max.retry.interval;  Ignoring.
14/03/18 10:19:42 WARN conf.Configuration: file:/tmp/hadoop-james.mchugh/mapred/local/localRunner/james.mchugh/job_local904621238_0001/job_local904621238_0001.xml:an attempt to override final parameter: mapreduce.job.end-notification.max.attempts;  Ignoring.
14/03/18 10:19:42 INFO mapreduce.Job: The url to track the job: http://localhost:8080/
14/03/18 10:19:42 INFO mapred.LocalJobRunner: OutputCommitter set in config null

最佳答案

尝试使用 job.waitForCompletion(true); 而不是 job.submit();。由于您在本地运行 mapreduce，因此您应该在 JUnit 杀死本地 jobtracker 之前等待结果。

关于java - Hadoop 不产生任何输出，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/22476852/