You can bring your own Horizontal Pod Autoscaler (HPA) plug-in to Gloo Gateway. This way, you can automatically scale gateway proxy pods up and down based on certain thresholds, like memory and CPU consumption.

To allow integration with HPA plug-ins, do not specify any custom replicas in the GlooGatewayParameters resource. This way, your HPA plug-in can scale the proxy based on your autoscaling strategy.

Before you begin

  1. Follow the Get started guide to install Gloo Gateway.

  2. Follow the Sample app guide to create a gateway proxy with an HTTP listener and deploy the httpbin sample app.

  3. Get the external address of the gateway and save it in an environment variable.

      export INGRESS_GW_ADDRESS=$(kubectl get svc -n gloo-system http -o jsonpath="{.status.loadBalancer.ingress[0]['hostname','ip']}")
echo $INGRESS_GW_ADDRESS  
  
      kubectl port-forward deployment/http -n gloo-system 8080:8080

Set up your own HPA plug-in

  1. Deploy the Kubernetes metrics-server in your cluster. The metrics-server retrieves metrics, such as CPU and memory consumption for your workloads. These metrics can be used by the HPA plug-in to determine if the pod must be scaled up or down.

      kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
kubectl -n kube-system patch deployment metrics-server \
 --type=json \
 -p='[{"op":"add","path":"/spec/template/spec/containers/0/args/-","value":"--kubelet-insecure-tls"}]'

    Example output:

      serviceaccount/metrics-server created
clusterrole.rbac.authorization.k8s.io/system:aggregated-metrics-reader created
clusterrole.rbac.authorization.k8s.io/system:metrics-server created
rolebinding.rbac.authorization.k8s.io/metrics-server-auth-reader created
clusterrolebinding.rbac.authorization.k8s.io/metrics-server:system:auth-delegator created
clusterrolebinding.rbac.authorization.k8s.io/system:metrics-server created
service/metrics-server created
deployment.apps/metrics-server configured
apiservice.apiregistration.k8s.io/v1beta1.metrics.k8s.io created

  2. Review the metrics for the http Gateway.

      kubectl top pod -n gloo-system  | grep http

    Example output:

      http-594455765c-5szpq        3m           19Mi

  3. Create a HorizontalPodAutoscaler resource that scales your gateway proxy up to 10 replicas if the memory consumption exceeds 10Mi. You can adjust this value depending on the memory value that you retrieved in the previous step.

      kubectl apply -f- <<EOF
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: hpa
  namespace: gloo-system
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: http
  minReplicas: 1
  maxReplicas: 10
  metrics:
    - type: Resource
      resource:
        name: memory
        target:
          type: AverageValue
          averageValue: 10Mi
EOF

  4. Wait a few minutes for Kubernetes to scale up your gateway proxies. Then, check the number of http pods that were created. Because every pod exceeds the memory threshold that you defined in the HPA policy, Kubernetes scales up the pods to the 10 maximum replicas.

      kubectl top pod -n gloo-system  | grep http

    Example output:

      http-594455765c-5szpq                   3m           19Mi            
http-594455765c-gvf9g                   3m           19Mi            
http-594455765c-h2zb9                   3m           19Mi            
http-594455765c-mghsf                   3m           19Mi            
http-594455765c-n6prj                   3m           19Mi            
http-594455765c-nvs6c                   3m           19Mi            
http-594455765c-srfcz                   3m           19Mi            
http-594455765c-ssvjx                   3m           19Mi            
http-594455765c-tx2z6                   3m           19Mi            
http-594455765c-vxjbp                   3m           19Mi

Cleanup

You can remove the resources that you created in this guide.
  kubectl delete hpa hpa -n gloo-system
kubectl delete -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml