Configure a Proxy for Snowflake’s Spark Connector on AWS

Configure the Spark connector's use_proxy, proxy_host and proxy_port options, with the documented AWS restrictions on protocol, account and bypass settings.

For the Snowflake Spark connector on AWS, configure the reader with the connector’s own proxy options:

options = {
    **snowflake_options,
    'use_proxy': 'true',
    'proxy_host': 'proxy.example',
    'proxy_port': '8080',
    'proxy_protocol': 'http',
}
reader = spark.read.format('snowflake').options(**options)

snowflake_options holds the application’s endpoint and authentication settings. Add the intended table or query before calling load.

use_proxy defaults to false. When enabled, it requires proxy_host and proxy_port as separate fields. proxy_user and proxy_password provide an intermediary account. Without proxy authentication, a permitted public-proxy route can use host and port candidates from the HTTP proxy list while retaining Snowflake’s account settings.

Snowflake documents the account fields, proxy_protocol and non_proxy_hosts as AWS-only. proxy_protocol was added in connector 2.11.1 and defaults to http. non_proxy_hosts separates hosts with a URL-escaped pipe (%7C) and supports an asterisk wildcard.

Use a connector matching the project’s Spark and Scala versions. These data-source options have their own names and formats; JDBC properties and java.net.Proxy objects are not interchangeable inputs.

Allakkat attuumassuteqartut