Uploaded image for project: 'Pentaho Data Integration - Kettle'
  1. Pentaho Data Integration - Kettle
  2. PDI-15984

Hive, Hive2, Cloudera Impala, and Impala options for Connections have Inconsistent User Interface Options



    • Story Points:
    • Notice:
      When an issue is open, the "Fix Version/s" field conveys a target, not necessarily a commitment. When an issue is closed, the "Fix Version/s" field conveys the version that the issue was fixed in.


      JNDI was omitted from the Hive, Hive 2, and Impala options when the user tries to create a connection in Spoon. After checking internally, we've found that this is because there are connection pooling compatibility issues with thrift, which is part of those drivers.

      However, when we look throughout out product, the inclusion or exclusion of the JNDI option for those options are not handled consistently. For example, in PUC, JNDI is included, but in Spoon it is not.

      This causes confusion because users aren't quite sure whether they can and cannot use JNDI, and it is also not clearly documented that there are issues with this particular driver. We do have a recommendation to not use connection pooling for the Impala JDBC driver in these two places, but not for Hive. And then, only in context with Analyzer:

      We are building a KCS to document this, but the inconsistency in the user interface can cause issues, particularly since including connection pooling could cause serious errors.

      The customer has also recommended that a warning be added about this here:

      This issue could save our users considerable time and pain. A ticket that was filed that explains this issue a bit more clearly appears in the internal comments.




            Unassigned Unassigned
            cbrathwaite Chantel Brathwaite (Inactive)
            0 Vote for this issue
            3 Start watching this issue