Woodgrove Bank

This repo shows off a banking app built using .NET, Kubernetes, and Azure.

Setup

There are a variety of ways the Woodgrove Bank apps can be set up on Kubernetes or on Azure PaaS.

Kubernetes

Presuming a local Kubernetes setup. In the case of this document, Docker Desktop was used, along with the Kubernetes that ships with it. A "Reset Kubernetes Cluster" was performed to reset the cluster.

1. Working Namespace

   Create a new namespace in your Kubernetes cluster for your deployment by executing this command. This way if you make a mistake you can easily reset by deleting the namespace and starting over.

   kubectl create namespace woodgrovebank01
   kubectl config set-context --current --namespace=woodgrovebank01 
   

2. If you haven't yet already installed ingress-nginx, do so next:

   kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.1.1/deploy/static/provider/baremetal/deploy.yaml

3. Per the Orleans Kubernetes docs, for RBAC-enabled clusters, the Kubernetes service account for the pods may also need to be granted the required access. rolebinding.yaml performs this routine on the Kubernetes cluster.

   kubectl apply -f ./k8s/rolebinding.yaml
   kubectl apply -f ./k8s/dnsutils.yaml

4. Install KEDA to enable silo auto-scaling with the experimental Orleans Scaler:

   helm repo add kedacore https://kedacore.github.io/charts
   helm repo update
   helm install keda kedacore/keda --namespace woodgrovebank01

5. Create a new Azure Storage account (or just use Azurite)

   You can either use Azurite to emulate the Azure Storage Service running right in your Kubernetes instance (not recommended for production usage, this is only for dev-time deployments). If you'd like to use Azurite, execute this command:

   Note: Do NOT execute both of the below commands. Only use one or the other.

   kubectl apply -f ./k8s/azurite.yaml
   kubectl create secret generic storage-connection-strings --from-literal=clustering="UseDevelopmentStorage=true;DevelopmentStorageProxyUri=http://azurite"

   Or, if you've created your own Azure Storage Account in the cloud, execute this command:

    kubectl create secret generic storage-connection-strings --from-literal=clustering="<connection-string-from-portal>"
   

6. Build the containers using docker-compose.

   docker-compose build

7. Now install all the services.

   kubectl apply -f ./k8s/api.yaml
   kubectl apply -f ./k8s/admin.yaml
   kubectl apply -f ./k8s/simulations.yaml
   kubectl apply -f ./k8s/dashboard.yaml

8. Browse to the various microservices you just deployed.

• Open your browser to the Woodgrove Bank Employee App.
• Open another browser tab to the Swagger UI page for the Woodgrove Bank API.
• Open a third browser tab to the The Orleans Dashboard.

1. (Optional) Run the data-generating app to generate fake data.

cd BogusGenerator
dotnet run

This console app will generate some fake data. As you use the Swagger UI page and the Bogus generator to create fake data, you'll see more grains appear in the cluster and in the front end interface. Add one round of customers and accounts.

Auto-scaling the Orleans Cluster by Grain Count

Now you'll add the Orleans Scaler (experimental) to the Kubernetes cluster see how you can scale your silos according to the number of a specific type of grain you expect to see in the Orleans cluster, per silo.

• Deploy the scaler

  kubectl apply -f ./k8s/scaler.yaml
  

• Watch the logs for the scaler. Here, I pass them to jq so they come out nicely-formatted on-screen.

  kubectl logs -f <orleans-scaler-pod-name> | jq
  

  Now you should see JSON-formatted logs on-screen. You may need to install jq, first, but if you don't want to, just omit the | jq part and you should be good.