cluster-api VS demos

In this blog post, we will demonstrate how easy and fast it is to deploy Sveltos on an RKE2 cluster with the help of ArgoCD, register two RKE2 Cluster API (CAPI) clusters and create a ClusterProfile to deploy Prometheus and Grafana Helm charts down the managed CAPI clusters.

4. Having moved to a container orchestrator, all of my nodes are immutable. Hardware and VM instances _can_ be born magically into existence. Nearly all infra providers support [cluster-api](https://cluster-api.sigs.k8s.io/). Network infrastructure can now be managed with TF, so I go that route.

## Linux curl -L https://github.com/kubernetes-sigs/cluster-api/releases/download/v1.4.4/clusterctl-linux-amd64 -o clusterctl sudo install -o root -g root -m 0755 clusterctl /usr/local/bin/clusterctl ## Mac brew install clusterctl

Did you ever try CAPI? https://github.com/kubernetes-sigs/cluster-api

You might find interesting the capi-rancher-import k8s operator we use in Sylva, it would adopt in Rancher server the Cluster API created k8s clusters (with bootstrap provider kubeadm or even rke2 - you can lookup CAPBR for the latter). I understand your clusters are not created by Cluster API, so if you could move the workloads/resources to new clusters created by Cluster API, this can come handy. (Adoption of non-CAPI clusters into CAPI is not yet a standard practice, more in https://github.com/kubernetes-sigs/cluster-api/issues/7776)

Most of the comments have mentioned older tools like kubespray, Ansible, Rancher etc. I would suggest the cloud native way using ClusterAPI or use a tool that relies on ClusterAPI in the backend called Talos

Cluster API

Sveltos is a powerful open source project that makes managing Kubernetes add-ons a breeze. It automatically discovers ClusterAPI powered clusters and allows you to easily register any other cluster (like GKE). Then, it seamlessly manages Kubernetes add-ons across all your clusters.

can be found here.

The ConfigMap referenced by EventBasedAddOn instance contains all resources that will be deployed in each cluster where a Pod in crashing state is found. In this case:

All YAMLs used in this example can be found here


https://github.com/projectsveltos/sveltosctl#display-outcome...

But I was still not happy. Main reason, users still had to go and manage cluster labels. I wanted clusters' labels to change as cluster runtime state was changing. So I can express my intent and then forget about it.

So I introduced a second concept in Sveltos. Classifier CRD (https://github.com/projectsveltos/demos/blob/main/classifier...) which allows users to classify clusters based on cluster runtime state (currently kubernetes version and/or resources deployed, but I am working on adding more).

Doing so I can easily now say:

1) I wanted a controller running in the management cluster, so I can express my intent and as clusters come and go or change their runtime state, controller can make sure my intent is applied. For instance, combing Classifier and ClusterProfile I can say things like, in any cluster running kubernetes version v1.24.x install calico v3.23 and in any cluster running kubernetes version v1.25.x install calico v3.24. as clusters are created/upgraded, sveltos makes sure my intent is deployed (https://github.com/projectsveltos/demos/blob/main/classifier/classifier.gif)

So a second concept was introduced in Sveltos: Classifier CRD (https://github.com/projectsveltos/demos/blob/main/classifier/classifier.gif) which allows users to classify clusters based on cluster runtime state (currently Kubernetes version and/or resources deployed, but I am working on adding more).