Deployment

Deployment guide

actinia can be deployed in different environments depending on the intended use case. Docker is suitable for a quick local or server-based setup, Helm is intended for Kubernetes or OpenShift deployments, and pip can be used for custom Python-based installations.

Docker

actinia requires a running Valkey database. For a quick deployment, Valkey can be started together with actinia using Docker Compose. Run the following commands from the docker directory of the actinia-core repository. Build and start the services:

docker-compose -f docker-compose.yml build
docker-compose -f docker-compose.yml up

To run the services in the background:

docker-compose -f docker-compose.yml up -d

The desired actinia version can be configured in the docker-compose.yml file.

For more details, see the Docker deployment documentation.

Helm chart

actinia can be deployed in Kubernetes or OpenShift environments using the official Helm chart.

The Helm chart and installation instructions are available in the actinia Helm charts repository.

pip and PyPI

actinia-core can also be installed from PyPI:

pip install actinia-core

This installation option requires additional setup, including a running Valkey server and an actinia configuration file.

For a complete installation example, see the OSGeoLive actinia installation script together with the required actinia configuration files.

For actinia plugins published on PyPI, see the PyPI search results for actinia.

After deployment

After deploying actinia, configure authentication and job processing according to your environment:

• Authentication
• Job queues and workers

For extensions and additional functionality, see Modules and plugins.