[MapReduce-user] reduce tasks pending - no progress for few hours

On 06-Dec-2011, at 9:32 PM, Keren Ouaknine wrote:

Hello,

Please find details below. I would like to resume the pending tasks (not sure why they went into pending state in the first place).
I had a look at the logs, there were some tasks failures (3), but successful on a second attempt.

Job Setup: Successful
Status: Running
Started at: Tue Dec 06 05:41:39 EST 2011
Running for: 5hrs, 18mins, 49sec
Job Cleanup: Pending

Kind % Complete Num Tasks Pending Running Complete Killed Failed/Killed
Task Attempts
map 100.00%

1260 0 0 1260 0 0 / 6
reduce 9.99%


90 81 0 9 0 3 / 0

Thanks,
Keren

--
Keren Ouaknine
Cell: +972 54 2565404
Web: www.kereno.com

On Tue, Dec 6, 2011 at 11:26 AM, Harsh J wrote:

Keren,

How many tasktrackers are you running, and are they all still up?

On 06-Dec-2011, at 9:32 PM, Keren Ouaknine wrote:

Hello,

Please find details below. I would like to resume the pending tasks (not
sure why they went into pending state in the first place).
I had a look at the logs, there were some tasks failures (3), but
successful on a second attempt.
*
Job Setup:* Successful<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=setup&pagenum=1&state=completed>
*Status:* Running
*Started at:* Tue Dec 06 05:41:39 EST 2011
*Running for:* 5hrs, 18mins, 49sec
*Job Cleanup:* Pending

Kind% CompleteNum TasksPendingRunningCompleteKilledFailed/Killed
Task Attempts<http://10.239.24.12:50030/jobfailures.jsp?jobid=job_201111302144_0187>
map<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=map&pagenum=1>
100.00%
1260001260<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=map&pagenum=1&state=completed>
00 / 6<http://10.239.24.12:50030/jobfailures.jsp?jobid=job_201111302144_0187&kind=map&cause=killed>
reduce<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=reduce&pagenum=1>
9.99%

9081<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=reduce&pagenum=1&state=pending>
09<http://10.239.24.12:50030/jobtasks.jsp?jobid=job_201111302144_0187&type=reduce&pagenum=1&state=completed>
0 3<http://10.239.24.12:50030/jobfailures.jsp?jobid=job_201111302144_0187&kind=reduce&cause=failed>/ 0
Thanks,
Keren

--
Keren Ouaknine
Cell: +972 54 2565404
Web: www.kereno.com

On 06-Dec-2011, at 10:46 PM, Keren Ouaknine wrote:

Hello Harsh,

1. All tasktracker are up, they are 10 available. What do you reckon?

2. On one node, there is additionally a process called Child. It seems like a tasktracker, so I am not sure why both Child and tasktracker are running on that node?

Thanks,
Keren

On Tue, Dec 6, 2011 at 11:26 AM, Harsh J wrote:
Keren,

How many tasktrackers are you running, and are they all still up?



--
Keren Ouaknine
Cell: +972 54 2565404
Web: www.kereno.com