Kogito Serveless Workflow using Knative Functions Example: E-Commerce App

This example aims to port an example of AWS Step functions into the Knative/Serverless Workflow platform.

Prereqs

1. Ko installed
2. Java SDK installed
3. Maven installed
4. Quarkus CLI
5. Knative quickstart plugin
6. Kind
7. PostgreSQL

To edit your workflows:

1. Visual Studio Code with Red Hat Java Plugin installed
2. Serverless Workflow Editor

Creating a Knative cluster

Skip this step if you have already access to a Kubernetes with Knative Serving.

Create a Kind cluster and install Knative Serving by using this command:

kn quickstart kind --install-serving

ko supports loading images directly to Kind local registry:

export KO_DOCKER_REPO=kind.local

Below is the name of the kind cluster created by kn quickstart

export KIND_CLUSTER_NAME=knative 

Installing postgreSQL

The commerce application stores orders and inventories in postgreSQL.

Install kubegres:

kubectl apply -f https://raw.githubusercontent.com/reactive-tech/kubegres/v1.15/kubegres.yaml

kubectl wait deployment -n kubegres-system kubegres-controller-manager --for condition=Available=True --timeout=90s

kubectl apply -f config/infra/postgres

Apply the e-commerce YAML configs

Use ko apply to apply the YAML config for all functions:

ko apply -f config

Building and Deploying the Workflow

Creating the project

To create the project skeleton, run:

kn workflow create --name e-commerce-ksw

This command will create a Maven Quarkus project in the e-commerce-ksw directory with all required Kogito dependencies.

Copy the e-commerce workflow

rm -rf e-commerce-ksw/src/main/resources/* && cp config/sw/* e-commerce-ksw/src/main/resources

Building your project's image and Deploying to Knative

Navigate to your project's directory. For this example:

cd e-commerce-ksw

You can use the Serverless Workflow plug-in for the Knative CLI to build your image with the following command:

kn workflow build --image dev.local/e-commerce-ksw:1.0

Load the produced container image into Kind:

kind load docker-image dev.local/e-commerce-ksw:1.0 --name=knative

Then deploy the workflow as a Knative application:

kn service create -f target/kubernetes/knative.yml

You should see something like "service.serving.knative.dev/e-commerce-ksw created" in the terminal

Check the service is ready:

kn service list

Expected output:

NAME           URL                                              LATEST               AGE   CONDITIONS   READY   REASON
e-commerce-ksw   http://e-commerce-ksw.default.127.0.0.1.sslip.io   e-commerce-ksw-00001   12s   3 OK / 3     True  

Interacting with your application

To interact with the application, you can call the service via command line

curl -v -X POST -H 'Content-Type:application/json' -H 'Accept:application/json' -d '{"workflowdata" : {"order_id":"8dee2","order_info":{"order_date":"2022-01-01T02:30:50Z","customer_id":"id001","order_status":"fillIn","items": [{"item_id":"itemID456","qty":1,"description":"Pencil","unit_price":2.5},{"item_id":"itemID789","qty":1,"description":"Paper","unit_price":4}],"payment":{"merchant_id":"merchantID1234","payment_amount":6.5,"transaction_id":"54c512","transaction_date":"2022-01-01T02:30:50Z","order_id":"8dee2","payment_type":"creditcard"},"inventory":{"transaction_id":"54c512","transaction_date":"2022-01-01T02:30:50Z","order_id":"8dee2","items":["Pencil","Paper"],"transaction_type":"online"}}}}'  http://e-commerce-ksw.default.127.0.0.1.sslip.io/commerce

Resources

• Quarkus Container Images Guide
• Getting Started With Quarkus
• CNCF Serverless Workflow
• Kogito Serverless Workflow