Kudu catalog

You can add Kudu as a catalog in Flink SQL by adding Kudu dependency to your project, registering the Kudu table in Java, and enabling it in the custom environment file.

The Kudu connector comes with a catalog implementation to handle metadata about your Kudu setup and perform table management. By using the Kudu catalog, you can access all the tables already created in Kudu from Flink SQL queries.

The Kudu catalog only allows to create or access existing Kudu tables. Tables using other data sources must be defined in other catalogs, such as in-memory catalog or Hive catalog.

Maven Dependency

<dependency>
   <groupId>org.apache.flink</groupId>
   <artifactId>flink-connector-kudu_2.11</artifactId>
   <version>1.0-csa1.2.0.0</version>
</dependency>

The following example shows how to register and use the Kudu catalog from Java:

String KUDU_MASTERS="host1:port1,host2:port2"
KuduCatalog catalog = new KuduCatalog(KUDU_MASTERS);
tableEnv.registerCatalog("kudu", catalog);
tableEnv.useCatalog("kudu");
tableEnv.listTables();

To use the Kudu Catalog from the SQL client, you have to add it in the YAML configuration file to the catalogs section:

- name: kudu
    type: kudu
    kudu.masters: <kudu_masters>

Required catalog configuration:

Kudu.masters

For more information about the Kudu connector and catalog, see the official documentation.