GitLab GitOps, 7 Nov 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: GitOps Series - The Infrastructure

Description

GitOps is the use of a git repository as a single source of truth for all infrastructure and application deployment code. In this video we will walk through the process of creating application infrastructure using Terraform and Terraform Cloud. We will build a kubernetes cluster, add a service account, and register the cluster with GitLab for future application deployments.

See the code at: https://gitlab.com/gitops-demo

Learn more about GitOps: https://about.gitlab.com/topics/gitops/

A

Hello today we're going to talk about get ops specifically how gitlab enables get-ups and how you can use get ops to deploy infrastructure as code. So what is get ups get ups is the use of a git repository, it's the single source of truth for all infrastructure and application deployment code. As you can see from my example here, we've got various different teams that contribute to code that will live in a git repository and then be through through automation, using terraform to deploy out to different public cloud services.

A

In this example, I'm gonna walk you through how we create three different kubernetes clusters in three different public clouds. All using a common process collaborating with my team, all within get lab, so first thing I'm going to do is go to get lab and log into my group called get ops demo, looking at the readme section here and I just want to show off kind of the structure of the group. So at the top we have a top-level group called get-ups demo.

A

We've got a couple of various different individual projects and two sub groups, one for applications and one for infrastructure. Today, we're going to look at the infrastructure.

A

I'm now in my infrastructure group, I've got a repository for each cloud that I'd like to work on and one for templates. If I click on my AWS repository as an example, I can see all the files associated here. These are all written in terraform and I've got a git lab CI mo file. This provides the instructions for the automation.

A

Let's have a quick look. Around I've got a couple of different files here, a back-end file.

A

That dip that calls out that we're going to store our terraform state files in terraform cloud in an organization called get-ups demo and a works place called AWS. This keeps a running state in the cloud provider, so anybody can access this at any time. Well, anybody on my team, at least.

A

I've got another farm file here for the kubernetes cluster itself, leveraging the eks module I, give it a few parameters like what kind of subnets how many nodes I want. What size and some tags.

A

I need to create a gitlab admin user on the kubernetes cluster I want that done automatically as code and managed by terraform, so I leverage the kubernetes provider. To do this.

A

And lastly, I want to register this kubernetes cluster with get lap so I'm going to use the git lab provider to create a group cluster called AWS cluster.

A

Give it a domain name and an environment scope and pass in the credentials so after this runs, all of this will be deployed. Micheal Esther will be created in AWS and it will be automatically registered to my get ops demo, slash apps group. Let's take a quick look at that.

A

Here I have my applications here: I have my eks cluster.

A

All the information is provided automatically to get lab through terraform.

A

The URL that was out, we created the cluster name, the CA certificate and the token now to deploy applications to this cluster. All I need to do is run my CI jobs in my various different application groups.

A

Sample of this I.

A

Can see here that the build test.

A

I can see that my build stage all of my testing stage, including my security scans, and then my deploy into production was successful.

A

If I look at the logs here, I can see the last time this one was deployed. It was actually deployed to the AKS cluster.

A

Let's take a look at the CI code that deploys these kubernetes clusters with terraform. Go back to my infrastructure group. Look at my templates and look at my terraform template.

A

What I've got here is a CI file that has a few different stages.

A

Validate plan apply and destroy it's going to use hashey corpse, terraform base image container and we're going to run a few different tasks. We're gonna, initialize terraform, we're gonna, validate that everything's correct we're gonna, follow good practices, get ops practices by creating a merge request that merge request will define what's going to happen. The plan file will then be uploaded to the merge request.

A

We'll do a final plan before deploying after deploying to master and then we'll apply. This we've even got a destroy stage. If we wanted to delete all of this this code, that's it. This is how you use terraform to deploy. Infrastructure is code with git lab using git ops processes. Thank you.