FireboltEngineClass overview

Optional namespaced classes of shared FireboltEngine settings, including pod template, storage, engine config, and rollout and auto-stop defaults.

Overview

FireboltEngineClass is an optional namespaced resource that holds reusable FireboltEngine settings: a pod-template fragment plus defaults for storage, customEngineConfig, rollout, drainCheckEnabled, drainCheckInterval, and autoStop. Engines reference a class by name through spec.engineClassRef. The Firebolt Operator resolves that name namespaced-first: a FireboltEngineClass in the Engine’s namespace if present, otherwise a ClusterFireboltEngineClass of the same name. A namespaced object of the same name is an explicit override of the cluster SKU.

Each inherited field resolves the engine value first, then the namespace FireboltEnginePreset value, then the resolved class value, then the Firebolt Operator’s default. An engine that sets a field owns it. The class fills it in otherwise.

Multiple classes can exist per namespace. The Engine picks one by name, and using a class is optional. See EngineClass configuration inheritance for template precedence, per-Engine overrides, deletion, and rollout behavior.

For the allowlist of user-settable and Firebolt Operator-owned template fields, see the FireboltEngineClass CRD reference.

Inspecting status

kubectl get firengc -n firebolt

NAME                BOUND   READY   AGE
compute-optimized   3       True    24h

Examples

Create a FireboltEngineClass

The FireboltEngineClass can hold shared pod-level properties for engines in the same namespace.

apiVersion: compute.firebolt.io/v1alpha1
kind: FireboltEngineClass
metadata:
  name: default
  namespace: firebolt
spec:
  template:
    metadata:
      labels:
        owner: firebolt
    spec:
      containers:
        - name: engine
          image: ghcr.io/firebolt-db/engine:dev

Apply the manifest:

kubectl apply --server-side -f firebolt-engine-class.yaml

Create a FireboltEngine

The FireboltEngine can reference a namespaced FireboltEngineClass or a cluster ClusterFireboltEngineClass through spec.engineClassRef.

apiVersion: compute.firebolt.io/v1alpha1
kind: FireboltEngine
metadata:
  name: my-engine
  namespace: firebolt
spec:
  instanceRef: demo-instance
  engineClassRef: default
  replicas: 1
  customEngineConfig:
    storage:
      managed_table_storage: s3
      managed_table_bucket_name: my-engine-bucket
      aws:
        endpoint: http://floci.firebolt.svc.cluster.local:4566
        path_style_addressing: true # non-AWS S3-compatible store; ignored for AWS S3
  template:
    spec:
      containers:
        - name: engine
          # floci is zero-auth, but the engine still signs S3 requests — supply any creds via env.
          env:
            - name: AWS_ACCESS_KEY_ID
              value: firebolt
            - name: AWS_SECRET_ACCESS_KEY
              value: firebolt
          resources:
            requests:
              cpu: "2"
              memory: "3Gi"
            limits:
              memory: "3Gi"

Apply the manifest:

kubectl apply --server-side -f engine.yaml

Additional resources

See the FireboltEngineClass CRD reference for the full field surface. See firebolt-engine-class.yaml, instance-with-engines.yaml.