doytsujin / moniker

naming for cloud resources

Moniker

This project is a Frigga replacement aimed at making naming strategies within Spinnaker more flexible.

How it works

Every resource is assigned a Moniker which exposes app, stack, detail, and cluster fields. If, for example, a Moniker is assigned using Frigga, we can say that cluster = app-stack-detail. However, Frigga isn't required.

Every cloud provider, Spinnaker account, or even resource can decide how Monikers get applied/derived. What this means is that we can support either deriving and assigning a Moniker directly via the resource's name (what Frigga does), or store the Moniker in the resource's metadata. As an example, the new Kubernetes provider stores Monikers in annotations:

metadata:
  annotations:
    "moniker.spinnaker.io/application": my-application-name
    "moniker.spinnaker.io/cluster": canary-frontend
  name: frontend-test-cache

Notice that the resource's name, cluster and application are completely decoupled.

The logic for assigning & deriving Monikers only needs to happen in the relevant Clouddriver integration; the rest of Spinnaker only sees the Moniker that was assigned to a resource. This gives the rest of Spinnaker one vocabulary for looking up resources by their relationships (in particular deciding which application and cluster a resource belongs to).