Transformations > Machine Learning transformation > Configuring an API proxy
  

Configuring an API proxy

You can configure an API proxy to enforce security for APIs at run time.
The following table describes the types of proxies that are available:
Proxy type
Description
None
Bypasses the proxy server configured at the agent, Spark, or connection level.
Spark
Considers the proxy configured for the Spark engine.
Custom
Considers the proxy configured at the connection level.
Note: A platform proxy is not available and you must configure a Spark proxy instead. A platform proxy considers the proxy configured at the agent level, but mappings that use the Machine Learning transformation refer to the Spark engine for proxy details.
API proxies do not apply if the Machine Learning transformation runs in a serverless runtime environment.

Bypassing the proxy server

To bypass the proxy server, select None as the proxy type in the REST V3 connection. Do not configure a proxy server at the agent or Spark level. For more information, see the help for the appropriate connector.

Configuring a Spark proxy

A Spark proxy replaces a platform proxy. You must configure the Spark proxy in the mapping task. For more information, see the help for the appropriate connector.
    1In the REST V3 connection, select Platform as the proxy type.
    2In the mapping task, set the Spark session property spark.driver.extraJavaOptions to the following value: -Dhttp.proxyHost=<host> -Dhttp.proxyPort=<port> -Dhttp.proxyUser=<user> -Dhttp.proxyPassword=<password>

Configuring a custom proxy

Configure a custom proxy to use the proxy details in the REST V3 connection. Do not configure a proxy server at the agent or Spark level. For more information, see the help for the appropriate connector.
    1In the REST V3 connection, select Custom as the proxy type.
    2Configure the host, post, user, and password.
    3Save the connection.