[ http://issues.apache.org/jira/browse/HADOOP-787?page=comments#action_12456199 ] Runping Qi commented on HADOOP-787: -----------------------------------
I think the jobtracker should save the old jobs to DFS and read them back when they are needed. > job tracker should limit the number of completed/failed jobs in RAM > ------------------------------------------------------------------- > > Key: HADOOP-787 > URL: http://issues.apache.org/jira/browse/HADOOP-787 > Project: Hadoop > Issue Type: Bug > Components: mapred > Affects Versions: 0.9.0 > Reporter: Owen O'Malley > Assigned To: Arun C Murthy > Fix For: 0.10.0 > > > The job tracker currently runs out of memory when a user runs a lot of small > jobs back to back. To bound the amount of RAM that the job tracker consumes > recording completed/failed jobs, I think that the JobTracker should limit the > number of "remembered" jobs per a user to > jobtracker.max.remembered.task.per.user (default 100). -- 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
