Red Hat OpenShift GitOps | OpenShift, 14 May 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Helm Deployments with Argo CD

Description

In this video, Christian Hernandez will go over a brief demonstration on how you can deploy your Helm charts using Argo CD.

A

Hi, my name is christian hernandez from the cloud platforms business unit over at red hat in this video. I'm going to be going over a brief demonstration on how to deploy helm, charts using argo, cd.

A

Before we begin, I installed the red hat openshift get ops operator beforehand. You can install this operator at the operator hub. Once I have the red hat openshift get ops operator installed, I will have access to argo cd.

A

There are two main ways to deploy a helm chart using argo. Cd first is to use the native integration within argo cd to deploy the helm chart. You do this by clicking new app and give your application a name.

A

Then specifying your sync policy.

A

Next, instead of deploying a git repo, you will select helm here. You will put in your helm, repo, url and the chart you want to deploy.

A

Next, the destination cluster which you want to deploy and the name space you want to deploy to.

A

Here inside argo cd, you can specify your helm values here that you want to override.

A

In this instance, I'm not going to do a build because I already pre-built the image beforehand, so I will select, builds, enabled equals false and give it the specific image that I want to use for this chart.

A

Once I click create you'll notice that you'll have a helm logo specifying that this is a helm deployment.

A

Now I can interact with this application using the argo cd cli using the argo cd cli. I can do argo cd app list to list my applications.

A

Now I interact with this application and set further override values. For example, if I want to set the replica count equal to two.

A

Once I set this argo cd sees the change and will sync the application for me, giving me two pods now. This is a perfectly valid way to use argo, cd and helm, but it's not necessarily git ops, friendly in a git ops way of doing things. You need the manifest to be reflected in a git repo, not just inside of argo cd.

A

To do this, you can do what we call a sub chart method going over to my ammo and looking at the subchart resource you um application url.

A

You can see that I have a repo url of an actual git repo, the path that I want to specify where my yaml manifest lives and also things like the server and the destination name space.

A

Going over to my git repo and my quarkus subchart directory, you can see that I set up what is known as an empty helm. Chart this empty helm chart has a dependencies of the actual chart that I want to use.

A

Also inside this subchart directory is I have the values.yaml file of the values I want to overwrite for my home chart.

A

Next, I'm going to apply this manifest, creating the argo city, application.

A

Once I create it, it'll create another application.

A

This time note that the application will have a get logo instead of a helm, logo.

A

Notice how I have three pods now, if I want to make a change, I don't interact with the argo cd cli. Instead, I will use a git ops workflow by changing the get repo upstream, for instance, I want to set up this to have one replica instead of three.

A

Once I set these values I'll go ahead and do a hard.

A

Sync and show now that argo cd sees that I want one replica instead of.

A

A

I hope you enjoyed this video and invite you to try out deploying your helm, charts using argo city on openshift. Thank you.

A