Skip to content
Config Schema

Operator Guide

The Operator Guide covers configuring and running Rise as a platform.

Use this section if you are:

  • Operating Rise in shared environments
  • Managing backend settings and infrastructure integrations
  • Running Rise on Kubernetes or in production environments

For application developer workflows, start with the User Guide.

Unified container model (one-time deployment re-roll)

Section titled “Unified container model (one-time deployment re-roll)”

Rise models every deployment as one or more containers — a single-container app is the one-container case (an implicit container named app). The on-cluster resource names reflect this: the Deployment is <project>-<deployment_id>-app, its Service is <group>-app (on the app’s own port), and the ingress backend points at that Service.

Upgrading from a release that predates this model recreates the K8s resources for every running deployment on the first reconcile: each app’s Deployment and Service are renamed and the ingress backend moves, so every app restarts once. Existing container images are reused (the synthesized app pulls the deployment’s existing image tag — nothing is rebuilt), and both subdomain and sub-path routing are preserved. Schedule a maintenance window for the upgrade.