Installation of Knative on a GKE Autopilot cluster

install-knative.sh

# Install Knative
kubectl apply -f https://github.com/knative/serving/releases/download/knative-v1.0.0/serving-crds.yaml
kubectl apply -f https://github.com/knative/serving/releases/download/knative-v1.0.0/serving-core.yaml

# Autopilot Mod: Remove annotation
kubectl patch -n knative-serving deployment activator --type=json -p='[{"op":"remove","path":"/spec/template/metadata/annotations/cluster-autoscaler.kubernetes.io~1safe-to-evict"}]'
kubectl patch -n knative-serving deployment autoscaler --type=json -p='[{"op":"remove","path":"/spec/template/metadata/annotations/cluster-autoscaler.kubernetes.io~1safe-to-evict"}]'
kubectl patch -n knative-serving deployment domainmapping-webhook --type=json -p='[{"op":"remove","path":"/spec/template/metadata/annotations/cluster-autoscaler.kubernetes.io~1safe-to-evict"}]'
kubectl patch -n knative-serving deployment webhook --type=json -p='[{"op":"remove","path":"/spec/template/metadata/annotations/cluster-autoscaler.kubernetes.io~1safe-to-evict"}]'

# Autopilot Mod: Provision Spare Capacity
kubectl create -f https://gist.githubusercontent.com/WilliamDenniss/3cff8658411ee3342c3e35017c22898c/raw/94aee82768a0dd3ff1f789bc8e5ae02f7ff2b408/knative-hello-world-balloon.yaml

# Install Kourier
kubectl apply -f https://github.com/knative/net-kourier/releases/download/knative-v1.0.0/kourier.yaml
kubectl patch configmap/config-network \
  --namespace knative-serving \
  --type merge \
  --patch '{"data":{"ingress-class":"kourier.ingress.networking.knative.dev"}}'

# Configure DNS
kubectl apply -f https://github.com/knative/serving/releases/download/knative-v1.0.0/serving-default-domain.yaml

knative-hello-world-balloon.yaml

apiVersion: scheduling.k8s.io/v1
kind: PriorityClass
metadata:
  name: balloon-priority
value: -10
preemptionPolicy: Never
globalDefault: false
description: "Balloon pod priority."
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: balloon
spec:
  replicas: 1
  selector:
    matchLabels:
      app: balloon-pod
  template:
    metadata:
      labels:
        app: balloon-pod
    spec:
      priorityClassName: balloon-priority
      terminationGracePeriodSeconds: 0
      containers:
      - name: ubuntu
        image: ubuntu
        command: ["sleep"]
        args: ["infinity"]
        resources:
            requests:
              cpu: 725m
              memory: 750Mi

You're a baller.
Thank you.

apart from the safe-to-evict annotation that blocks cluster-autoscaler, there are also PDBs that prevent rolling redeploy even when the HPA is set to minScale 2. To allow a rolling redeploy:

kubectl -n knative-serving patch horizontalpodautoscaler activator -p '{"spec": {"minReplicas": 2}}'
kubectl -n knative-serving patch horizontalpodautoscaler webhook -p '{"spec": {"minReplicas": 2}}'
kubectl -n knative-serving patch poddisruptionbudget activator-pdb -p '{"spec": {"minAvailable": "50%"}}'
kubectl -n knative-serving patch poddisruptionbudget webhook-pdb -p '{"spec": {"minAvailable": "50%"}}'