Knative functions

Knative functions are used to create, build, and deploy stateless, event-driven functions as Knative Services by using the func CLI. Knative functions can run any container on any Kubernetes cluster, whether the container is built around a serverless function or other application code. Knative functions can handle network routing, event triggers and autoscaling.

Advantages of Knative functions and typical use cases are:

• They are stateless and event-driven, which means they can respond quickly and efficiently to changes in data or user requests.
  • They are autoscaled by Knative based on demand, which means they can scale up or down as needed, even to zero when idle.
  • They support traffic management and progressive rollout, which means they can route requests to different versions of an application based on rules or percentages.
  • They are developer-friendly software, which means they provide a simple programming model that abstracts away complex Kubernetes concepts and tasks.
  • They offer flexibility and control, which means they allow developers to customize their functions with environment variables, annotations, secrets, config maps etc..

To create a serverless function with func CLI and Knative, you need to follow these steps:

• Install Knative Serving and Eventing components on your Kubernetes cluster.
• Install func CLI on your local machine.
• Create a new project directory and initialize a function using func create command. You can specify the language, runtime, template and name of your function. For example:

  mkdir hello
  cd hello
  func create --runtime python --template http --name hello

• Write your function code in the generated file (e.g., hello.py). You can use any dependencies or libraries you need by adding them to the requirements.txt file.
• Build your function into a container image using func build command. You need to provide a registry where your image will be pushed. For example:

  func build -r <registry>/<username>/hello

• Deploy your function to your cluster using func deploy command. You need to provide the same registry as before and a namespace where your function will run. For example:

  func deploy -r <registry>/<username>/hello -n default

• Invoke your function using func invoke command or by sending an HTTP request to its URL. For example:

  func invoke hello
  curl http://hello.default.example.com

---

Related: Cloud events with python
To create a Python CloudEvents function using func CLI, you need to follow these steps:

• Create a new project directory and initialize a function using func create command with --template events option. This will generate a project that responds to a CloudEvent over HTTP. For example:

  mkdir hello-events
  cd hello-events
  func create --runtime python --template events --name hello-events

• Write your function code in the generated file (e.g., hello-events.py). You can access the CloudEvent data and context as parameters of your function. For example:

  def main(event, context):
    print(f"Hello, {event['name']}!")
    print(f"Context: {context}")

• Build and deploy your function to your cluster using func build and func deploy commands as before. For example:

  func build -r <registry>/<username>/hello-events
  func deploy -r <registry>/<username>/hello-events -n default

• Invoke your function by sending a CloudEvent to its URL using curl or another tool. For example:

  curl -v "http://hello-events.default.example.com" \
    -X POST \
    -H "Ce-Id: say-hello" \
    -H "Ce-Specversion: 1.0" \
    -H "Ce-Type: greeting" \
    -H "Ce-Source: not-sendoff" \
    -H "Content-Type: application/json" \
    -d '{"name":"Knative"}'

---

References

• https://knative.dev/docs/functions/
• https://learn.microsoft.com/en-us/azure/azure-functions/create-first-function-cli-python
• https://stackoverflow.com/questions/57149078/firebase-cloud-functions-using-python