You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-dev@hadoop.apache.org by "Ravi Gummadi (JIRA)" <ji...@apache.org> on 2009/04/27 12:20:30 UTC

[jira] Created: (HADOOP-5748) ProcessTree.destroy() is sleeping for 5 seconds holding the task slot

ProcessTree.destroy() is sleeping for 5 seconds holding the task slot
---------------------------------------------------------------------

                 Key: HADOOP-5748
                 URL: https://issues.apache.org/jira/browse/HADOOP-5748
             Project: Hadoop Core
          Issue Type: Bug
          Components: mapred
            Reporter: Ravi Gummadi


Currently, in ProcessTree.destroy(), after sending SIGTERM to the task JVM, TT sleeps for 5 seconds(default value of mapred.tasktracker.tasks.sleeptime-before-sigkill) before sending SIGKILL. This seems to be blocking the task slot(not getting released) for 5 seconds. We should avoid this so that another task could be launched in that slot immediately.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.