How do I see the actual Map Reduce job in Hadoop from beyond all abstractions?

Question

I have a Hadoop Cluster. On top of that is HBase. On top of that is Phoenix. When I execute SELECT * FROM TEST LIMIT 1000; from Phoenix... how do I actually see what job was sent to hadoop?

Answer 1

Phoenix does this using parallel scans using Hbase coporcessors and not using map reduce jobs. You can see the details of the query plan by doing

explain SELECT * FROM TEST LIMIT 1000;

If you need more details, route the logs from phoenix to a different log category on hbase server.