I have a cluster with HDP 2.2 setup on linux and using C# client with ODBC driver connectiing to hiveserver2.
What are the limits of the ODBC implementation in terms of queries ? (I know the api ODBC implementation is not completed)
I run the query for example :
select * from table limit 100 -> No problem.
select count(*) from table ->
System.Data.Odbc.OdbcException (0x80131937): ERROR [HY000] [Hortonworks][HiveODBC] (35) Error from Hive: error code: ‘0’ error message: ‘ExecuteStatement finished with operation state: ERROR_STATE’.
select * from table where to_date(column_date) = ‘2015-01-01′ -> works fine
select m.* from (select * from table where to_date(column_date) = ‘2015-01-01) as m
System.Data.Odbc.OdbcException (0x80131937): ERROR [HY000] [Hortonworks][HiveODBC] (35) Error from Hive: error code: ‘0’ error message: ‘ExecuteStatement finished with operation state: ERROR_STATE’.
ODBC seems very inconsistent .. Sub queries seems to never work and some functions not working ?
Note : ODBC driver advanced settings I tried all the options (Native Query etc..)
Any helps ? Is it a known limitation of ODBC driver ?
Best practice or alternative to connect with C# to Hive using HDP on linux ?