Chapter 1. Getting started with functions
Function lifecycle management includes creating and deploying a function, after which it can be invoked. You can do all of these operations on OpenShift Serverless using the kn func
tool.
1.1. Prerequisites
To enable the use of OpenShift Serverless Functions on your cluster, you must complete the following steps:
The OpenShift Serverless Operator and Knative Serving are installed on your cluster.
NoteFunctions are deployed as a Knative service. If you want to use event-driven architecture with your functions, you must also install Knative Eventing.
-
You have the
oc
CLI installed. -
You have the Knative (
kn
) CLI installed. Installing the Knative CLI enables the use ofkn func
commands which you can use to create and manage functions. - You have installed Docker Container Engine or Podman version 3.4.7 or higher.
- You have access to an available image registry, such as the OpenShift Container Registry.
- If you are using Quay.io as the image registry, you must ensure that either the repository is not private, or that you have followed the OpenShift Container Platform documentation on Allowing pods to reference images from other secured registries.
- If you are using the OpenShift Container Registry, a cluster administrator must expose the registry.
1.2. Creating, deploying, and invoking a function
On OpenShift Serverless, you can use the kn func
to create, deploy, and invoke a function.
Procedure
Create a function project:
$ kn func create -l <runtime> -t <template> <path>
Example command
$ kn func create -l typescript -t cloudevents examplefunc
Example output
Created typescript function in /home/user/demo/examplefunc
Navigate to the function project directory:
Example command
$ cd examplefunc
Build and run the function locally:
Example command
$ kn func run
Deploy the function to your cluster:
$ kn func deploy
Example output
Function deployed at: http://func.example.com
Invoke the function:
$ kn func invoke
This invokes either a locally or remotely running function. If both are running, the local one is invoked.