actions-runner-controller VS external-dns

To set-up the self-hosted runner, an Action Runner Controller (ARC) and Runner scale sets application will be installed via helm. This post will be using Azure Kubernetes Service and ARC that is officialy maintained by Github. There is another ARC that is maintained by the community. You can follow the discussion where github adopted the ARC project into a full Github product here

Before this we were using https://github.com/actions/actions-runner-controller but that's running on K8s instead of VMs. So along with common limitations of running CI jos in K8s/container, it cannot have exactly the same environment as the official GitHub runners. Maintaining a K8s cluster was also very difficult.

ARC is great for running GitHub Actions on Kubernetes:

https://github.com/actions/actions-runner-controller

Almost all of our cicd, builds run on GitHub. I'm talking cypress tests, deployments via terraform and helm to over 25 environments, all backend tests, daily test runs etc. Overall we were racking up a cost of almost 20k on GitHub. With the ARC deployed and using spot instances I think our total infrastructure costs went up about 4-5k even though we added more actions. If we switched back to their runners we'd probably be around 25k at this point.

What else needs to be moved to my artifactory (charts - https://github.com/actions/actions-runner-controller/tree/master/charts ) - if so tar or entire folder or anything else ? ) What should the above steps correspond to?

You need to use the steps in the repo instead of the steps on the docs if you're using enterprise server.

Honestly not a fan of Github docs.....I feel like the ones in the repo are much clearer and easier to understand/read.

Its pretty easy to set up honestly. Deploy this on your k8s cluster https://github.com/actions/actions-runner-controller and a runnerDeployment and youre good to go.

Trying to use the Actions Runner Controller (https://github.com/actions/actions-runner-controller) to utilize self-hosted runners. I keep getting this error on the controller.

I'm convinced one of (or a combination) of things is happening here in regards to authentication. This GH enterprise account is configured with SAML. I feel like that is a valid data point. I'm using https://github.com/actions/actions-runner-controller as a reference guide for what I should be doing. I suspect whoever is Owner of this organization has modified what I can do as a user. The steps in the doc where I can actually Install the Application isn't available to me. When configuring the GitHub App I'm given two options. I select the option for "this account only" knowing the documentation says it is possible to use this Github App with a repo in the Organization as long as I have Admin privileges or I'm the owner.

The second one is a combination of tools: External DNS, cert-manager, and NGINX ingress. Using these as a stack, you can quickly deploy an application, making it available through a DNS with a TLS without much effort via simple annotations. When I first discovered External DNS, I was amazed at its quality.

One of the reasons why I chose Hetzner was that it WAS supported by the ExternalDNS project. I didn't quite understand why the Hetzner provider was pulled, but I saw that an attempt of re-adding it was refused, on the ground that the upcoming webhook architecture would have allowed to better maintain providers.

Write a custom controller for the external DNS controller, or setup some form of ArgoCD app / appset templating.

so I am looking at having an automated way for new routers registered in Traefik to also have the corresponding DNS entry added to my Pihole instance similar to external-dns but obviously, this is exclusive to ingress on k8s environments. my current setup is traefik in a container on unraid.

You technically don’t need to run a load balancer or have a virtual IP for your control plane. If you control your dns, you can add an A record pointing to all IPs for your control plane nodes. It won’t load balance your traffic, but combined with something like External DNS it gives you HA for the control plane.

I normally deploy external-dns, which automatically updates DNS with the ingress controller's external IP address.

Background: Having a look I can see this https://github.com/kubernetes-sigs/external-dns

3) Use the Kubernetes External-DNS. I've never used this, but this is assuming it can update DNS for each pods/app to point to the correct Node (it'd need to update my Homelab DNS running on Windows Server)

So here’s the solution I have taken for our current stack. EKS and its dependencies are created through terraform using the eks module as well as provision a route53 subdomain and a wildcard cert. Once we have that created, I have installed this deployment into the cluster via the helm module: https://kubernetes-sigs.github.io/aws-load-balancer-controller/v2.4/. This allows me to use kuberentes resources (load balancers or ingress objects) and it will handle all the provisioning of load balancers and security groups for me, based on my application yaml and annotations. We also use https://github.com/kubernetes-sigs/external-dns to manage all of our specific host names for the applications through annotations. So to generally put, terraform manages out Kubernetes clusters, and Kubernetes manages the deployment of anything needed for the application including volumes, load balancers, hostnames though Kubernetes system deployments

Metallb for your load balancer (layer2 mode will do) NginX-ingress, will be spot on for internal home apps External-dns to publish your dns record to your Dns server at home, https://github.com/kubernetes-sigs/external-dns