2.0 KiB
2.0 KiB
Coolify Deployment Reference
This note captures the current production deployment inputs so the automated Coolify workflow can be wired up consistently.
Compose bundle
- Use
docker-compose.prod.ymlas the base service definition. - The compose file expects the following variables supplied by the deployment runner (either through an
.envfile or Coolify secret variables):DATABASE_HOSTDATABASE_PORT(defaults to5432if omitted)DATABASE_USERDATABASE_PASSWORDDATABASE_NAME- Optional
APT_CACHE_URLbuild arg when a proxy is required.
- Expose port
8003for the FastAPI service and5432for PostgreSQL.
Runtime expectations
- Application container runs with
ENVIRONMENT=production,LOG_LEVEL=WARNING, and enables large import/export limits via:CALMINER_EXPORT_MAX_ROWSCALMINER_IMPORT_MAX_ROWSCALMINER_EXPORT_METADATACALMINER_IMPORT_STAGING_TTL
- PostgreSQL container requires persistent storage (volume
postgres_data). - The app health check hits
http://localhost:8003/health.
Coolify integration inputs (to provision secrets later)
- Coolify instance URL (e.g.
https://coolify.example.com) and target project/app identifier. - API token or CLI credentials with permission to trigger deployments.
- SSH key or repository token already configured in Coolify to pull this repository.
- Optional: webhook/event endpoint if we want to observe deployment status from CI.
Manual deployment checklist (current state)
- Ensure the compose environment variables above are defined in Coolify under the application settings.
- Rebuild the application image with the latest commit (using
docker-compose.prod.yml). - Trigger a deployment in Coolify so the platform pulls the new image and restarts the service.
- Confirm the health check passes and review logs in Coolify.
These details will feed into the new deploy-coolify workflow so it can authenticate, trigger the deployment, and surface logs automatically.