Creating, deploying, updating, and rolling back a Helm chart
As you can see from the previous example, deploying pre-packaged applications with Helm is pretty simple. Deploying your own application is also easy. We start with the following command, which will create a directory in your current directory called myhelmchart and populate it with the relevant files and templates:
$ helm create myhelmchart
By default, the values.yaml file created by this command contains references to a single NGINX pod and creates a ClusterIP service, which is only accessible from within the cluster. The key values from the default file are shown next:
replicaCount: 1 image: repository: nginx service: type: ClusterIP
We can easily deploy this new Helm chart using the following command:
$ helm install example ./myhelmchart --set service.type=NodePort
This will override the service.type value seen in the values.yaml file with a NodePort service, so it is now exposed...