k8s-experiments

Experimenting with k8s for scalable GatewayD deployment

Resources

The following resources were used to create this project:

• Kubernetes
• MicroK8s

Setup

1. Install MicroK8s on your machine. See here for instructions. The last command in the instructions will print the token you need to access the Kubernetes dashboard. The IP address and port number for the dashboard can be found in the kubernetes-dashboard service.

   sudo snap install microk8s --classic
   sudo apt update && apt install -y ufw
   sudo ufw allow in on cni0 && sudo ufw allow out on cni0
   sudo ufw default allow routed

2. Add your user to the microk8s group.

   sudo usermod -a -G microk8s $USER
   sudo chown -f -R $USER ~/.kube
   newgrp microk8s

3. Enable the following MicroK8s addons:

   • dns
   • dashboard
   • storage

   microk8s enable dns dashboard storage

4. See the status of the MicroK8s cluster and optionally get the token for the Kubernetes dashboard.

   microk8s kubectl get all --all-namespaces
   token=$(microk8s kubectl -n kube-system get secret | grep default-token | cut -d " " -f1)
   microk8s kubectl -n kube-system describe secret $token

5. Follow the instructions in the how to deploy Postgres on kubernetes article to deploy a Postgres instance. You can use the files starting with postgres- in this repo as a starting point.

   microk8s kubectl apply -f postgres-config.yaml
   microk8s kubectl apply -f postgres-pvc-pv.yaml
   microk8s kubectl apply -f postgres-deployment.yaml
   microk8s kubectl apply -f postgres-service.yaml

6. Install GatewayD on your machine.

   microk8s kubectl apply -f gatewayd-config.yaml
   microk8s kubectl apply -f gatewayd-deployment.yaml
   microk8s kubectl apply -f gatewayd-service.yaml

7. Scale GatewayD to any number of replicas you want.

   microk8s kubectl scale deployment gatewayd --replicas=3

8. Use the following command to find the exposed port for the GatewayD service.

   microk8s kubectl get all

9. Use the following command to test the GatewayD service. Replace the port number with the one you found in the previous step.

   psql postgres://postgres:postgres@localhost:31221/postgres