Install AI Server

A step-by-step guide for deploying a DataForge AI Server with Docker Compose.

This chapter covers the installation of a DataForge AI Server using Docker Compose and provides a reference docker compose file to use for deployment.

Deploying AI Server using Docker Compose

1. Prerequisites

To set up a new AI Server, a number of prerequisites have to be met.

Network access to a DataForge Server (gRPC and gRPC-Web endpoints)
Network access to Zabbix server and database
A Linux host with Docker and Docker Compose installed
An AI Server Token from an AI Server on a Zabbix server

You can check whether Docker and Docker Compose are installed by running the following commands:

docker version
docker compose version

2. Docker Compose

The following docker-compose.yaml file defines how the AI Server will start.

You need to modify the docker-compose.yaml file to correctly include the following information:

Connection details to your DataForge server
An AI Server token from your DataForge server (see Zabbix server setup)
Connection details and credentials to your Zabbix database
Connection details to the trapper port of your Zabbix server

services:
    df-ai-server:
        image: images.intellitrend.de/dataforge/dataforge-core/df-ai-server:7.10.12
        restart: always
        environment:
            GOMEMLIMIT: "600MiB"
            DF_AI_SERVER_LOG_LEVEL: "info"
            DF_AI_SERVER_METRICS_IDENTITY: "df_ai_server"
            # DataForge Server connection
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_ADDRESS: "dataforge.intellitrend.de"
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_PORT: "8090"
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_TLS_ENABLE: "false"
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_WEB_ADDRESS: "dataforge.intellitrend.de" # Usually the same as the GRPC_ADDRESS, only with a different port
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_WEB_PORT: "8091"
            DF_AI_SERVER_AISERVER_DATAFORGE_SERVER_GRPC_WEB_TLS_ENABLE: "false"
            # Zabbix Database connection
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_ADDRESS: "[REPLACE_ME]"
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_NAME: "zabbix"
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_PASSWORD: "[REPLACE_ME]"
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_PORT: "5432"
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_USER: "[REPLACE_ME]"
            DF_AI_SERVER_AISERVER_ZABBIX_DATABASE_DRIVER: "postgres"
            # Zabbix Server connection
            DF_AI_SERVER_AISERVER_ZABBIX_SERVER_ADDRESS: "[REPLACE_ME]"
            DF_AI_SERVER_AISERVER_ZABBIX_SERVER_TRAPPER_PORT: "10051"
            # AI Server configuration
            DF_AI_SERVER_AISERVER_STATE_FILE: "/usr/local/intellitrend/df/var/state.json" # Stores the time of the latest metric-data read from the Zabbix database
            DF_AI_SERVER_AISERVER_AI_SERVER_TOKEN: "[REPLACE_ME]" # AI Server token on the Zabbix server
            DF_AI_SERVER_AISERVER_CONFIG_POLL_INTERVAL: 15 # Interval in seconds in which the AI Server checks for new deployments
            DF_AI_SERVER_AISERVER_MODEL_PRIMING_DURATION: 60 # Warm-up time of the model in seconds
        volumes:
            - ./df-ai-runner/data/var/:/usr/local/intellitrend/df/var/:rw

Running the AI Server

To start the AI Server, simply use the following commands, pulling the required docker image from our docker registry.

# Pull image
docker compose pull

# Start in background
docker compose up -d

# Check logs
docker compose logs -f df-ai-runner

The AI Server should connect to the DataForge Server, poll AI deployments, and begin evaluating models.