Installing an HDF Cluster for IBM Power Systems
Also available as:
PDF

Configure SAM

About This Task

When you configure Hortonworks Streaming Analytics Manager (SAM), you must provide information about the metadata store database, configure a connection with Schema Registry, and establish the URL for the Druid Supersets.

Steps

  1. In the Customize Services step, navigate to the STREAMLINE CONFIG section of the Streaming Analytics Manager tab.

  2. Select Jar Storage Type. If you plan to enable HA for SAM on this cluster, you must select HDFS.

  3. If you selected HDFS as the Jar Storage Type, configure Jar Storage HDFS URL. This specifies the HDFS location where you want the jars to be stored. For example, hdfs://<<NN_HOST:8020:/hdfs/registry.

  4. Configure jar.storage to the directory on which you want to store .jar files for custom processors.

  5. Set the streamline.dashboard.url to the Superset URL which you can access using Quick Links for Druid.

  6. Configure registry.url to the REST API Endpoint URL for the Registry.

    The format should be http://$FQDN_REGISTRY_HOST:$REGISTRY_PORT/api/v1, where

    • $FQDN_REGISTRY_HOST specifies the host on which you are running Schema Registry and

    • $REGISTRY_PORT specifies the Schema Registry port number, as in the following example:

      http://FQDN_REGISTRY_HOST:7788/api/v1

      You can find the Schema Registry port in the REGISTRY_CONFIG section of the Registry tab.

  7. Configure the STREAMLINE STORAGE configurations based on the database you created to use as a SAM metadata store.

  8. Ensure that the registry storage connector URL has the fully qualified name of the host on which the database was installed and the connector url and default port for the database selected.

Example

MYSQL example:

jdbc:mysql://FQDN_MYSQL:3306/streamline


Postgres Example:

jdbc:postgresql://FQDN_POSTGRES:5432/streamline