Nathan Roberts created MAPREDUCE-4728:
-----------------------------------------

             Summary: Interaction between oob heartbeats and damper can cause 
TT to heartbeat with zero delay
                 Key: MAPREDUCE-4728
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4728
             Project: Hadoop Map/Reduce
          Issue Type: Bug
    Affects Versions: 1.0.3
            Reporter: Nathan Roberts


When mapreduce.tasktracker.outofband.heartbeat is true and 
mapreduce.tasktracker.outofband.heartbeat.damper is something largish (like the 
default of 1000000), the TT doesn't wait for tasks to finish before 
heartbeating back to the JT. This causes excessive load on the JT which in-turn 
reduces overall cluster performance.

I believe the problem is that in the following block of code, when 
getHeartbeatInterval() returns 0, we heartbeat back immediately BUT 
finishedCount does not get reset. It looks like nothing ever gets us out of 
this situation so we basically heartbeat without ever sleeping.
 
{code}
        // accelerate to account for multiple finished tasks up-front
        long remaining =
          (lastHeartbeat + getHeartbeatInterval(finishedCount.get())) - now;
        while (remaining > 0) {
          // sleeps for the wait time or
          // until there are *enough* empty slots to schedule tasks
          synchronized (finishedCount) {
            finishedCount.wait(remaining);

            // Recompute
            now = System.currentTimeMillis();
            remaining =
              (lastHeartbeat + getHeartbeatInterval(finishedCount.get())) - now;

            if (remaining <= 0) {
              // Reset count
              finishedCount.set(0);
              break;
            }
          }
        }

{code}

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to