I am new to MapR sandbox and exploring HBase Java API to interact with MaprDB Tables. My host environment is Windows and I am building application in Eclipse IDE and want to test against MapR sandbox. My POM file contains dependencies taken from mapR maven repositories. Initially I had not downloaded MapR client on my host. So whenever I used to run the program it gives me error :
2016-08-24 12:12:47,5756 ERROR Cidcache fs/client/fileclient/cc/cidcache.cc:1586 Thread: 12756 MoveToNextCldb: No CLDB entries, cannot run, sleeping 5 seconds!
2016-08-24 12:12:52,5884 ERROR Client fs/client/fileclient/cc/client.cc:831 Thread: 12756 Failed to initialize client for cluster maprdemo, error Cannot send after transport endpoint shutdown(108)
Exception in thread "main" java.io.IOException: Could not create FileClient
Later after following MapR thread :Setting Up the Client - MapR 5.0 Documentation - doc.mapr.com
I downloaded and added MapR client in my MapR_Home (C:/opt/mapr/) directory of my host. On running the configure.bat gave me an error of The system cannot find the path specified.
I ignored it and accidentally ran the program to my surprise the program ran and was able to connect to MapR FS and get the desired result. That made me think does those MapR client just needs to be present in MAPR_HOME directory to make application connect to MaprFS in sandbox? Because I have not followed all the steps mentioned in the thread of Setting up client of MapR after running that configure batch which gave me error. Does that mean program needs just filesystem jars present in that MapR client folder to talk to MapR FS and how does it works?