[ http://issues.apache.org/jira/browse/HADOOP-185?page=comments#action_12377439 ]
Doug Cutting commented on HADOOP-185: ------------------------------------- Might it be cleaner to make the launchingTasks field private to ExpireLaunchingTasks, and have access to it through syncronized methods? That would make it much easier to assure oneself that it is always accessed in a thread safe manner, as access would be more localized, no? > tasks are lost during pollForNewTask > ------------------------------------ > > Key: HADOOP-185 > URL: http://issues.apache.org/jira/browse/HADOOP-185 > Project: Hadoop > Type: Bug > Components: mapred > Versions: 0.1.1 > Reporter: Owen O'Malley > Assignee: Owen O'Malley > Priority: Critical > Fix For: 0.2 > Attachments: launch-time-out.patch > > There is the potential for "losing" tasks that are assigned by the JobTracker > to a TaskTracker, but that fail during returning the result (usually due to a > RPC timeout). In this case, the Job becomes "wedged" in that the tasks will > never run and never time out. -- 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
