如何使“ hadoop jar”命令获取相同名称的jar的新版本

Question

I'm new to hadoop and I tried its example within hadoop 2.6.0 for beginning.
First, I recompiled the source code of hadoop-mapreduce-examples-2.6.0.jar and build a new jar file MapReduce-0.0.1.jar

Then I ran the terasort example with this command line

jjin:hadoop$ bin/hadoop jar ~/shared/MapReduce-0.0.1.jar terasort /input /output
15/01/07 12:27:44 INFO terasort.TeraSort: starting
15/01/07 12:27:46 INFO input.FileInputFormat: Total input paths to process : 1
...

After terasort is done, I updated the source of TeraSort.java by modifying the first line log message like this

public int run(String[] args) throws Exception {
LOG.info("starting..."); 
// Update log message by adding '...' to the end of previous one.
Job job = Job.getInstance(getConf());

But after re-run this terasort job, I found the log message didn't change to 'starting...', so it means that the change I made to TeraSort.java doesn't take effect.

The question is how to make hadoop pick up the new MapReduce-0.0.1.jar I build. Thanks

Answer 1

您必须重新编译并重新构建jar，然后指定除第一个作业以外的其他输出目录。