Use Case: Impfpass

This is a digital Big-Data Vaccination Certificate-Prototype-App and an Examination performance.

The Base of this Application was another Big-Data-Application implemented by our Big-Data-Professor Prof. Dr.-Ing. habil. Dennis Pfisterer.

Some Parts of his parts, are still present in this Readme and in the Code of this Application.

Git Clone

This Project is using Git submodules. To clone the whole project please proceed as follows:

git clone https://github.com/hschaeufler/big_data_use_case.git
git submodule init
git submodule update

Initialize

If you are running on Windows and have Hyper-V, we recommend to use Hyper-V as default driver:

minikube config set driver hyperv

It's recommended also that you give minikube a little more Memory. With the following commands you can allocate CPUs and Memory to the Kubernets cluster.

minikube config set memory 8192
minikube config set cpus 4

When you have already started kubernetes, you need to delete the current Cluster und restart it.

minikube delete
minikube start

Don't forget to enable the ingress plugin

minikube addons enable ingress

Prerequisites

A running Strimzi.io Kafka operator

helm repo add strimzi http://strimzi.io/charts/
helm install my-kafka-operator strimzi/strimzi-kafka-operator
kubectl apply -f https://farberg.de/talks/big-data/code/helm-kafka-operator/kafka-cluster-def.yaml

Maybe you get the following Error

C:\workspaces\bigdata\big-data-use-case>helm install my-kafka-operator strimzi/strimzi-kafka-operator
Error: failed to download "strimzi/strimzi-kafka-operator" (hint: running `helm repo update` may help)

Then use helm repo update

Instead of the old kafka-cluster-def.yaml you have to use an upgraded version:

kubectl apply -f helm-kafka-operator\kafka-cluster-def.yaml

A running Hadoop cluster with YARN (for checkpointing)

helm repo add stable https://charts.helm.sh/stable
helm install --namespace=default --set hdfs.dataNode.replicas=1 --set yarn.nodeManager.replicas=1 --set hdfs.webhdfs.enabled=true my-hadoop-cluster stable/hadoop

Deploy

To develop using Skaffold, use skaffold dev.

Open the App

For opening the App, please get the IP-Address if the Ingress using following command:

kubectl get ingress

Then you have to add the IP-Address with the Hostname vac.book to your etc/host-File. In Windows it's under following Location: C:\Windows\System32\drivers\etc.

It should look like this. Please use the correct Ip-Address.

# Vacbook
172.19.67.118 vac.book

After this you can open the App in your Browser https://vac.book

Certificate

Because some Progressive Web App-Features like the Media Devices API (for getting Access to the Webcam) requires a Secure Context (https), a self-signed Certificate was used for the ingress. The following Parameters was used to sign this Certificate. It's valid for 10.000 days.

openssl req -x509 -newkey rsa:4096 -keyout key.pem -out cert.pem -days 10000 -nodes

Then you can open the App with following Adress in your Browser: https://vac.book

Troubleshooting

Maybe following Commands will help you, when you have Problems running the Use-Case.

Get a overview over the running pods and services:

kubectl get all

Get More Infos about a pod or service:

kubectl describe podname

Command for reading the Log-File

kubectl logs podname

Getting Cluster Events

kubectl get events

Get Endpoints for checking DNS-Errors

kubectl get endpoints

Helm

When you have problems with hadoop or kafka you can uninstall and install the operator again. To uninstall a release, you first need the name of it. With the following command you can list all releases in the current namespace. Then you can use the name to uninstall it.

helm list
helm delete my-hadoop-cluster

Mysql DB

Connect to the MySQL-Databse with following command:

kubectl get pods
kubectl exec --stdin --tty [podname] -- /bin/bash
mysql -u root -p

Docker

Build a Dockerfile in current directory (.) and tags it (-t container-name)

 docker build -t container-name .

Run the builded Container in detached mode (-d) and connect port (p 3000:3000)

 docker run -dp 3000:3000 container-name

Stop a running container container

docker stop container-name

Show all running container

docker ps

Closing Note

Be careful when you use this App. It is a Prototype, not a Production-Ready Application. For example, we do not have any Html and JavaScript-Sanitation.