[ http://issues.apache.org/jira/browse/HADOOP-782?page=comments#action_12455727 ] Mahadev konar commented on HADOOP-782: --------------------------------------
Arun yes, that should be done. Infact I had that in my patch for HADOOP-775. If a tasks is killed/failed the tasktracker does not need to maintain state (files/mapoutputs) for these tasks and should remove these tasks from RunningJobs and clean them up. > TaskTracker.java:killOverflowingTasks & > TaskTracker.java:markUnresponsiveTasks only put the tip in tasksToCleanup > queue, they don't update the runningJobs > ---------------------------------------------------------------------------------------------------------------------------------------------------------- > > Key: HADOOP-782 > URL: http://issues.apache.org/jira/browse/HADOOP-782 > Project: Hadoop > Issue Type: Bug > Components: mapred > Reporter: Arun C Murthy > Assigned To: Arun C Murthy > Fix For: 0.9.0 > > > Both killOverflowingTasks and markUnresponsiveTasks only put the tip in > 'tasksToCleanup' queue, they should also call 'removeTaskFromJob' to update > the 'runningJobs' map to reflect the fact that the task is no longer running. > Thoughts? -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira
