Explanation
This chart packages the APEx Dispatch API FastAPI service for Kubernetes environments that use Knative Serving. Its main goal is to make the service easy to deploy, configure, render, and validate while the application is being developed and tested.
What The Chart Describes
The chart creates a small set of resources around the application:
- a Knative
Serviceto run the FastAPI container - a
ConfigMapfor non-secret runtime configuration - an optional
Secretfor sensitive values such as client credentials and backend configuration
This keeps the deployment model simple and focused on the needs of a service that is iterated on frequently.
Why Knative Serving
The chart uses serving.knative.dev/v1 Service resources instead of a plain
Kubernetes Deployment because the target runtime is Knative Serving.
That choice gives the chart a deployment model centered on:
- revision-based updates
- request-driven scaling
- Knative-native autoscaling settings such as
minScale,maxScale, andcontainerConcurrency
For this chart, Knative is part of the deployment contract rather than an optional add-on.
Configuration Approach
The chart separates configuration by intent:
- non-secret application settings go into a
ConfigMap - secret material is read from a Kubernetes
Secret
This mirrors the FastAPI application's runtime needs while keeping local rendering and test scenarios straightforward. It also lets teams either create the secret from Helm values or bind the deployment to an existing secret already managed elsewhere.
Scope
This chart is currently scoped for development and testing purposes.
In practice, that means it is intended to support activities such as:
- local or shared environment validation
- integration testing of rendered manifests
- early deployment testing on a Knative cluster
- experimenting with runtime values, scaling settings, and secrets wiring
It is not documented as a production-hardened deployment package at this stage. Production concerns such as stricter operational defaults, stronger rollout guarantees, platform-specific hardening, and environment policy integration should be treated as follow-on work.
Why Probes Are Optional By Default
The chart leaves probes disabled by default because the FastAPI health endpoint checks dependencies beyond simple process startup. For development and testing environments, it is often more useful to bring the service up first and then enable readiness and liveness checks once the backing dependencies are wired correctly.