Using distributed cache on output of first hadoop job

Question

I am currently running two jobs such that job2 is dependent on job1. I plan to use output of job1 and pass it as distributed cache in job2. This will be used as reference in job2 to process another set of data. I have written load functions in the 2 jobs where i have done job configurations. Then i make job2 dependent on job1 and submit both it to JobControl. The problem is when i use distributed cache inside my load function in job2, the file is not yet created, and i will get a Exception. How do i handle this?

I couldn't see any initialize() etc method which i can overload instead of using my load() functions. I am using the new apis.

MyTranslationBuilderDriver job1 = new MyTranslationBuilderDriver();
    job1.load(args, "translator/path");
    MyTranslatorDriver job2 = new MyTranslatorDriver();
    job2.load(args, "path");
    JobControl jbcntrl=new JobControl("jbcntrl");
    ControlledJob cjob1 = new ControlledJob(job1.job, null);
    ControlledJob cjob2 = new ControlledJob(job2.job, null);
    cjob2.addDependingJob(cjob1);
    jbcntrl.addJob(cjob2);
    jbcntrl.run();
    Thread thread = new Thread(jbcntrl);
      thread.start();
      while (!jbcntrl.allFinished()) {
        try {
          Thread.sleep(10000);
        } catch (InterruptedException e) {
          e.printStackTrace();
        }
      }
      jbcntrl.stop();
}