Install Hive

Hive is distributed as a container image. Run one instance for each organization and connect it to a PostgreSQL database.

Your deployment platform must provide:

• A stable public address with a secure connection.
• A PostgreSQL database that persists across releases.
• A secret manager for Hive credentials.
• A way to run database migrations before a new release starts serving requests.

Configure the required settings

Every production installation needs these settings:

DATABASE_URL=ecto://USER:PASSWORD@HOST/DATABASE
SECRET_KEY_BASE=<a cryptographically secure 64-byte value>
PHX_HOST=hive.example.com

DATABASE_URL connects Hive to PostgreSQL. SECRET_KEY_BASE protects session data and must remain secret and stable across releases. PHX_HOST is the public host name Hive uses when creating links.

The container listens on port 4000 unless PORT is set. See the configuration reference for all optional settings.

Start with a public instance

Hive is public by default. This lets you confirm that the installation is healthy before depending on an external sign-in provider.

Deploy the published ghcr.io/tuist/hive image through your normal container platform. Run the image's migration command before starting the web process whenever the release changes.

Verify the installation

Open https://hive.example.com/ready. A successful response confirms that Hive can reach its database and accept requests. Then open https://hive.example.com to see the dashboard.

If the readiness check fails, review the migration and application logs through your deployment platform. Confirm that the database address, public host name, and session secret are available to both processes.

Continue setup

1. Configure authentication before making the instance private.
2. Review the role and visibility model.
3. Create the first project.
4. Connect the GitHub integration if projects should ingest issues and releases.
5. Review Deployment options for backups, object storage, secret management, and error reporting.