Hunk Cloudera 5.7 Not seeing MapReduce jobs in YARN Resource Manager

[vavkkishore_usa]

Dear All,

I installed Cloudera Quickstart VM 5.7 and installed Hunk by downloading splunk-6.4.2-00f5bb3fa822-Linux-x86_64.tar

Then created a virtual index -

Downloaded from Hunk website sample data file "Hunkdata.json.gz" - copied it in HDFS at this location /data/Hunkdata.json.gz

Did searches "index=ponyindex"

I see the output

My question:
I don't see any MapReduce programs running in YARN resource manager.

Where are MapReduce jobs running on Hadoop cluster - I should see an application_id in YARN Resource Manager but I don't see

Please help me where to look for YARN application jobs.

Thanks,

[rdagan_splunk · Accepted solution]

To run a MapReduce you will need to run this combo:
index=XYZ | stats count (or any Splunk reporting command) + be in Smart mode
Your
index=XYZ by itself will not generate a MapReduce job. Also if you are in Verbose mode you will not generate MapReduce jobs

[vavkkishore_usa]

Thanks rdagan for your response

[ddrillic]

I always start by going to -

And then to -

Please look for the first error in the search.log file.

[vavkkishore_usa]

Hi ddrillic ,
Thanks for your reply.

Yes I analyzed the search.log file and there is no error. I am able to see the results successfully.

I see the log files "stream" and in stackoverflow read Hunk will do MapReduce streaming.

I got a doubt is Hunk using Local MapReduce and not running the MR through YARN. To clarify this I raised this question and looking forward to help me why I am not seeing YARN application id or MapReduce job on the cluster though results are shown in the Splunk UI.

Thanks,