Categories

Versions

You are viewing the RapidMiner Deployment documentation for version 9.8 - Check here for latest version

Docker image for Job Agent

The documentation below describes the following Docker image:

  • Job Agent for RapidMiner Server (rapidminer/rapidminer-execution-jobagent:latest)

This image is available for download on Docker Hub.

Description

This image contains a Job Agent that executes jobs submitted on the RapidMiner Server instance in your RapidMiner platform deployment. For more background, read about our scalable architecture.

Environment variables

The Job Agent Container accepts the following environment variables during startup:

  • RAPIDMINER_SERVER_HOST: the hostname of RapidMiner Server
  • RAPIDMINER_SERVER_PORT: the port of RapidMiner Server (default: 8080)
  • RAPIDMINER_SERVER_PROTOCOL: the protocol (HTTP or HTTPS) used to contact RapidMiner Server
  • JOBAGENT_QUEUE_ACTIVEMQ_URI: the URI of the Active MQ (usually it is embedded in RapidMiner Server). Example value: failover:(tcp://host.of.the.rapidminer.server:5672)
  • JOBAGENT_QUEUE_ACTIVEMQ_USERNAME: the Active MQ username
  • JOBAGENT_QUEUE_ACTIVEMQ_PASSWORD: the Active MQ password
  • JOBAGENT_AUTH_SECRET: The auth secret used to authenticate to RapidMiner Server (a BASE64 encoded string)
  • JOBAGENT_CONTAINER_COUNT: The number of Job Containers assigned to the Job Agent
  • JOBAGENT_CONTAINER_MEMORYLIMIT: The amount of memory provided to the Job Containers
  • RAPIDMINER_JOBAGENT_OPTS: optional parameters which are passed to the Job Agent's Java Virtual Machine (JVM), in the JVM standard format (i.e. -Doptional.parameter.name=optionalParameterValue)
  • JOB_QUEUE: The name of the queue the Job Agent connects to
  • TZ: time zone specification based on the TZ database format (e.g. America/New_York)

Good to know

Please make sure you allocate enough memory for your Job Agent container, taking into account the memory need of your use cases. On Windows hosts, please make sure that the Docker Engine is configured to run with enough memory.