We are running datameer on top of cloudera hadoop platform and I am seeing that reduce job is running for almost two days now. Is there a way to fine tune the reduce job to fast? - there is enough reduce slot avaiable and cluster is free with slots. please advice.
Submit Host Address:
Job-ACLs: All users are allowed
Job Setup: Successful
Started at: Mon Dec 09 13:12:10 EST 2013
Running for: 68hrs, 32mins, 14sec
Job Cleanup: Pending
0 / 0
2 / 0
Seems that the reduce task has failed twice already. See the "2/0" thing. Can you drill down to one of the failed reduce task attempt and look at its logs?