Do you have running task trackers and data nodes? Which Nutch job did
you start? Any custom code?
Check the logs of of the four Hadoop daemons, there may be something
there.
On Tue, 01 May 2012 16:26:31 +0100, Dean Pullen
<[email protected]> wrote:
Hi all,
If this is definitely a Hadoop issue, as opposed to it being an issue
caused by Nutch, I'll happily go ask on the Hadoop mailing list...
Anyway, I'm kicking off a nutch ibject job via Hadoop 0.20.2 with
Nutch 1.4.
(I'm using v 0.20.2 because this is the library version included with
Nutch 1.4.)
This is the output:
2012-05-01 16:11:06,869 INFO org.apache.hadoop.mapred.JobTracker:
Initializing job_201205011600_0001
2012-05-01 16:11:06,870 INFO org.apache.hadoop.mapred.JobInProgress:
Initializing job_201205011600_0001
2012-05-01 16:11:07,099 INFO org.apache.hadoop.mapred.JobInProgress:
Input size for job job_201205011600_0001 = 47. Number of splits = 2
2012-05-01 16:11:07,102 INFO org.apache.hadoop.net.NetworkTopology:
Adding a new node: /default-rack/localhost
2012-05-01 16:11:07,102 INFO org.apache.hadoop.mapred.JobInProgress:
tip:task_201205011600_0001_m_000000 has split on
node:/default-rack/localhost
It then does nothing else. The Hadoop job tracker says Total
Submissions = 1, yet states that there are/have been, no running,
completed or failed jobs.
Any ideas as to what's stalling?
Cheers,
Dean Pullen
