Overview

OpenFunction is a cloud-native open source FaaS (Function as a Service) platform aiming to enable users to focus on their business logic without worrying about the underlying runtime environment and infrastructure. Users only need to submit business-related source code in the form of functions.

OpenFunction features but not limited to the following:

• Convert business-related function source code to runnable application source code.
• Generate a deployable container image from the converted application source code.
• Deploy the generated container image to the underlying runtime environment such as K8s, and automatically scale up and down according to business traffic, and scale to 0 when there is no traffic.
• Provide event management functions for trigger functions.
• Provide additional functions to manage function versions, ingress management etc.

---

Prerequisites

The current version of OpenFunction requires that you have a Kubernetes cluster with version >=1.18.6.

In addition, you need to deploy several dependencies for the OpenFunction Builder and Serving.

You can refer to the Installation Guide to setup OpenFunction Builder and Serving.

Builder

You need to install at least one of the following options for builders:

• Currently, OpenFunction Builder uses Tekton and Cloud Native Buildpacks to build container images, you need to install Tekton.

Serving

You need to install at least one of the following options for the serving component:

• Currently, OpenFunction Serving relies on Knative, so you need to install Knative Serving.
• Another Serving runtime Dapr + KEDA will be supported soon.

Tekton and Knative

You can deploy Tekton and Knative follow this command.

sh hack/deploy-tekon-and-knative.sh

You deploy Tekton and Knative follow this command if you having poor network connectivity to GitHub/Googleapis.

sh hack/deploy-tekon-and-knative.sh --poor-network

If you want to use NodePort to expose the Tekton dashboard service, follow this command.

sh hack/deploy-tekon-and-knative.sh --tekton-dashboard-nodeport <port>

CustomResourceDefinitions

The core function of OpenFunction is to enable users to develop, run, and manage business applications as execution units of code functions. OpenFunction implements the following custom resource definitions (CRDs):

• Function, defines a function.
• Builder, defines a function builder.
• Serving, defines a function workload.

QuickStart

Install

You can install the OpenFunction platform by the following command:

• Install the latest stable version

kubectl apply -f https://github.com/OpenFunction/OpenFunction/releases/download/v0.1.0/release.yaml

• Install the development version

kubectl apply -f https://raw.githubusercontent.com/OpenFunction/OpenFunction/main/config/bundle.yaml

Note: When using non-default namespaces, make sure that the ClusterRoleBinding in the namespace is adapted.

Sample: Run a function.

If you have already installed the OpenFunction platform, follow the steps below to run a sample function.

1. Creating a secret

   In order to access your container registry, you need to create a secret. You can create this secret by editing the username and password fields in config/samples/registry-account.yaml, and then apply it.

   kubectl apply -f config/samples/registry-account.yaml

2. Creating functions

   For sample function below, modify the spec.image field in config/samples/function-sample.yaml to your own container registry address:

   apiVersion: core.openfunction.io/v1alpha1
   kind: Function
   metadata:
     name: function-sample
   spec:
     image: "<your registry name>/sample-go-func:latest"

   Use the following command to create this Function:

   kubectl apply -f config/samples/function-sample.yaml

3. Result observation

   You can observe the process of a function with the following command:

   kubectl get functions.core.openfunction.io
   
   NAME              AGE
   function-sample   5s

   You can also observe the process of a builder in the Tekton Dashboard.

   Finally, you can observe the final state of the function workload in the Serving:

   kubectl get servings.core.openfunction.io
    
   NAME                      AGE
   function-sample-serving   15s

   You can now find out the service entry of the function with the following command:

   kubectl get ksvc
    
   NAME                           URL                                                                 LATESTCREATED                        LATESTREADY                          READY   REASON
   function-sample-serving-ksvc   http://function-sample-serving-ksvc.default.<external-ip>.nip.io   function-sample-serving-ksvc-00001   function-sample-serving-ksvc-00001   True

   Or get the service address directly with the following command:

   where<external-ip>indicates the external address of your gateway service

   kubectl get ksvc function-sample-serving-ksvc -o jsonpath={.status.url}
    
   http://function-sample-serving-ksvc.default.<external-ip>.nip.io

   Access the above service address via commands such as curl:

   curl http://function-sample-serving-ksvc.default.<external-ip>.nip.io
    
   Hello, World!

Removal

You can uninstall the components of OpenFunction by executing the following command:

kubectl delete -f config/bundle.yaml

Development

You can get help on developing this project by visiting Development Guide.

Roadmap

Here you can find OpenFunction's roadmap.