question Re: Pig job keeps running without progress in Archives of Support Questions (Read Only)

Pig job keeps running without progress

salamemder — Sun, 18 Aug 2019 12:38:26 GMT

I'm running Hortonworks version 2.3.2 on VirtualBox, just following dadoop tutorial: how to process data with apache pig (http://zh.hortonworks.com/hadoop-tutorial/how-to-process-data-with-apache-pig/).

Have uploaded batter.csv file to the exact location as per the instruction, and just executed the simplest load Pig statement: batting = LOAD 'Batting.csv' USING PigStorage(',');

The job starts running, but never proceeds. Tried to kill the job, cannot either.

Could anyone please give me some hint of the possible problem.

Used to run the same tutorial last year on the older version successfully.

Re: Pig job keeps running without progress

aervits — Tue, 23 Feb 2016 19:53:21 GMT

Your mapreduce service is down or execute in pig ambari view with tez, check the box for tez. Start the mapreduce2 service first

Re: Pig job keeps running without progress

nsabharwal — Tue, 23 Feb 2016 19:56:03 GMT

@Olivia Shen

Couple of things to check

1) MapReduce - there is alert in your screenshot. Make sure that Core services are up

2) Another job is consuming all the cluster resources

Re: Pig job keeps running without progress

nsabharwal — Tue, 23 Feb 2016 19:58:21 GMT

@Olivia Shen See this thread https://community.hortonworks.com/questions/14795/why-after-submitting-a-pig-job-does-it-fail-to-sta.html

Re: Pig job keeps running without progress

pankaj_singh — Tue, 23 Feb 2016 21:34:07 GMT

What is the Status of application started by pig script. You can view it onYarn->ResourceManager UI.

Re: Pig job keeps running without progress

salamemder — Thu, 25 Feb 2016 16:15:32 GMT

Thank you for the link to the thread. I switched to another PC that is more powerful, then problem solved. I guess this was because the cluster was busy

Re: Pig job keeps running without progress

nsabharwal — Thu, 25 Feb 2016 16:32:06 GMT

@Olivia Shen

Please accept the best answer to close the thread.