jsturtevant / azure-functions-kubernetes-chart

A chart for Azure Functions running on Kubernetes

Helm Chart for Azure Functions on Kubernetes

Installing

There are a few settings you should override:

• functionApp.name - name of your function apps your using
• resources.requests.cpu - required for Pod Autoscaler to work
• image.repository - Should be your image with your function app running in it (see http://www.jamessturtevant.com/posts/Running-the-Azure-Functions-runtime-in-kubernetes/ for creating a docker image with the azure functions runtime in it)
• scale.maxReplicas - max number of replicas
• scale.minReplicas - min number of replicas
• scale.cpuUtilizationPercentage - cpu percentage

An example would be:

helm install --set functionApp.name=sampleapp \
             --set resources.requests.cpu=200m \
             --set image.repository=vyta/functions \
             --set scale.maxReplicas=10 \
             --set scale.minReplicas=1 \
             --set scale.cpuUtilizationPercentage=50 \
             ./az-func-k8

Debugging

If you want to see what install might look like:

helm install --debug --dry-run --set resources.requests.cpu=200m ./az-func-k8

Test the scaling

Test that the scaling works:

docker run -it busybox /bin/sh
#
\#: 		while true; do wget -q -O- http://<your-ipaddress>/api/httpfunction?name=testingload; done 

In a separate terminal run:

kubectl get hpa
kubectl get deploy <deployment-name>