Cloudera Docs

Migrating stored Hive queries after Unified Analytics removal

Migrate or delete stored Hive queries in Cloudera Data Warehouse on premises 1.5.5 Service Pack 3 and higher releases following the removal of Unified Analytics.

In Cloudera Data Services1.5.5 SP3 and higher releases, Cloudera Data Warehouse removes the Unified Analytics feature. Consequently, affected Virtual Warehouses revert to standard Impala Virtual Warehouses, and the Hive engine is no longer available within them.

While the runtime upgrade to Impala is seamless, queries saved in Data Explorer as a Hive snippet type while using Unified Analytics will fail to load or run. The system triggers the Snippet type hive is not configured error. This occurs because Hue stores the engine type used at the time of query creation.

Migration options

Depending on the number of queries and their syntax, you can use one of the three methods to manage your stored queries.
  1. If a stored Hive query uses syntax specific only to Hive that is not supported by Impala, the query no longer functions in the upgraded warehouse. Delete the query.
    1. Go to the saved queries folder in Data Explorer.
    2. Select the Hive-specific query.
    3. Delete the query from the system.
  2. For individual queries that use syntax compatible with both Hive and Impala, manually update the snippet type.
    1. Open the existing Hive-saved query in the Data Explorer query editor.
    2. Copy the SQL text.
    3. Create a new query snippet and select Impala as the engine type.
    4. Paste the SQL text and save the query.
    5. Delete the original Hive version of the query.
  3. If you have a large number of saved queries, perform a batch update by exporting and editing the query metadata.
    1. Open Data Explorer and go to the Documents section.
    2. Select the Hive queries created in Unified Analytics and export them as a JSON file.
    3. Open the exported JSON file in a standard text editor. Use the search-and-replace function to change all instances of Hive to Impala and hive to impala.
    4. Save the edited JSON file and import it back into Data Explorer.
    5. Verify that the new Impala queries function correctly, then delete the Unified Analytics (Hive) saved queries.