README.md

Ink! Kubernetes charts

Here are two charts for running production ready ink! network.

Ink-network

In the ink-network folder to deploy a network use sh deploy.sh

or follow the steps bellow.

Templates

Templates consist of backend and frontend statefullsets (which represent server and plugin dockerfiles) and service which exposes ports and networking inside cluster. There are also ingress-rules for these services that are used on Ingress for network routing and as a rules for load balancer.

Cert

For generating certs run generate-certs.sh if you want to generate certificate with let's encrypt on Digital Ocean DNS.

Reserving static ip address can be computed using reserve-static-ip.sh.

Flow of the process

Prerequisites

• Minikube or your GKE/AKS/EKS cluster
• Helm (https://helm.sh/docs/intro/install/)
• Tiller

Configure default values

Update values in values.yaml file to your requirements (can be used as is)

Running in minikube

Start minikube with default values.

minkube start

Once it's finished you can proceed to next step.

Running in any kubernetes provider

Run helm and install the chart

helm install --namespace ink --name ink .

Once it finishes you can open dashboard on your cloud provider or with minikube use:

minikube dashboard &

Upgrade

helm upgrade ink .

Delete

If you want to delete just run

helm del --purge ink

Running in Kubernetes

For running in single cluster go to single_cluster_deployment.sh

For running multicluster deployments setup mesh and then install helm chart for both clusters (production.sh)

Troubleshooting

If you are missing Tiller just run:

helm init

Hacky solution for tiller problems:

kubectl --namespace kube-system create serviceaccount tiller

kubectl create clusterrolebinding tiller-cluster-rule \
 --clusterrole=cluster-admin --serviceaccount=kube-system:tiller

kubectl --namespace kube-system patch deploy tiller-deploy \
 -p '{"spec":{"template":{"spec":{"serviceAccount":"tiller"}}}}'

helm/helm#3130

If you see that some pods are not running, check if your clusters have enough RAM. If not, you can set "--num-nodes 2" when creating cluster or manually expand him.

Installing Ingress

helm install stable/nginx-ingress --namespace ink --name ink-ingress --set controller.replicaCount=2 --set rbac.create=true

Apply the rules

kubectl apply -f ingress-rules.yaml