Configuring Nexus as a Helm repo

What we will do:

– create a private (hosted) repository for our own package

– create a proxy repository pointing to stable helm repo server (https://kubernetes-charts.storage.googleapis.com) for 'kubernetes'or other private repo.

I suggest you create a new blob store for each new repo you want to create. That way, the data for every repo will be in a different folder in /nexus-data.

But this is not mandatory for it to work.

private repo

A repository for Helm repo that your team creates.

Create a new Helm (hosted) repository and configure it like:

proxy repo

A repository that proxies everything you download from the official repository, Kubeapps Hub. Next time you download the same dependency, it will be cached in your Nexus.

Create a new Helm (proxy) repository and configure it like:

Configuring Helm for private repository

Configuring Helm to use Nexus Repository is fairly easy! Once you have Helm up and running you'll want to run a command similar to the following:

​$ helm repo add nexusrepo http://localhost:8081/repository/helm-proxy/

Replace nexusrepo with what you'd like the repo to be called, and the url with what the full url to your proxy repository is. From that point you can install helm charts using a command similar to the following:

​$ helm install nexusrepo/mongodb

If everything went smoothly, the command above will install the latest version of mongodb.

Create a package

What is a helm chart? It is basically a set of templates and a file containing variables used to fill these templates. Let’s have a look at an example. I assume that you already have Helm installed and configured at this point.

To start working on a chart, Helm uses a simple command create:

​$ helm create my-app

After that Helm creates a directory with the following layout:​

my-app/
├── charts
├── Chart.yaml
├── templates
│     ├── deployment.yaml
│     ├── _helpers.tpl
│     ├── ingress.yaml
│     ├── NOTES.txt
│     └── service.yaml
└── values.yaml
2 directories, 7 files

It has charts directory with chart dependencies, but we don’t need it at the moment. Next comes Chart.yaml containing global variables for the chart such as version and description:

​$ cat my-app/Chart.yaml
apiVersion: v1
appVersion: "1.0"
description: A Helm chart for Kubernetes
name: my-app
version: 0.1.0

Then comes templates directory – there you put all the *.yaml files for Kubernetes. Helm uses Go template markup language to customize these files. Helm creates three default file types: deployment, service, and ingress. All the files in this directory are ‘skeletons’ which are filled with the variables from values.yaml when you deploy your Helm chart. File _helpers.tpl contains your custom helper functions for variable calculation. By default, helm creates an nginx deployment. Let’s customize it a bit. Add new ConfigMap to the templates directory:

​$ cat << HELM &gt; my-app/templates/cm.yaml
apiVersion: v1
data:
nginx.conf: |
  events {
    worker_connections 1024;
  }
  http {
    server {
      listen 80;
      location / {
        return 200 "===============================\n\n &nbsp;&nbsp;This is your helm deploy! \n\n===============================\n";
      }
    }
  }
kind: ConfigMap
metadata:
name: nginx-config
HELM

Point our nginx Deployment to that ConfigMap. Add the following lines to the deployment.yaml:

volumes:
- name: config
  configMap:
    name: nginx-config

and:

volumeMounts:
- name: config
  mountPath: /etc/nginx/nginx.conf
  subPath: nginx.conf

That’s it! Let’s check if we are doing the right thing: helm template my-app

This will generate all templates with variables and show the output. Now that we know everything is OK, we can deploy the chart:

helm install my-app --name=my-app-name

Then check that Service and Deploy have been created and curl our Service:

$ kubectl get svc

NAME                          TYPE CLUSTER-IP EXTERNAL-IP   PORT(S) AGE

my-app-name-my-app            ClusterIP 10.100.56.254 <none>        80/TCP 44m

$ curl 10.100.56.254

===================

 This is your helm deploy!  

===================

Congratulations! We have created and deployed our first Helm chart.

Additionally, you can create a package:

$ helm package my-app

This command creates an archive like my-app-0.1.0.tgz — now you can share your chart with others. For instance, you can upload this file to Helm repository, which we are going to do now.