Hello, brand new user to OpenSUSE, Rancher, and K3S I have successfully installed and configured the preliminary settings of K3S on two Raspberry Pis,

kubectl get nodes

returns the master and slave correctly Now, I just finished installing rancher-desktop into an OpenSUSE VM On first run, I get the welcome screen, but I am unsure what exactly the differences are between the options "containerd - use with nerdctl" and "dockerd (moby) - use with Docker CLI and k3d" and have been unable to find information a newbie like me can readily understand. can someone tell me which to choose, and a short description of why?

So at an extremely high level it's asking "what app do you want to use to manage your containers". Docker is the original "container runtime application" and is still widely used by developers today. Its engine (or core of the app) is called dockerd. docker is run by a private for-profit company and the CNCF (cloud native computing foundation) which is the "owner" of kubernetes has the aim of making kubernetes accessible and not locked to a specific vendor. dockerd (the core of docker) is open source so they made a fork (an alternative version based on) of dockerd. Kubernetes has mostly migrated the container runtime app from docker to containerd. So it's definitely more compatible to use containerd but a couple companies (like SUSE Rancher) support using docker if your team still needs it. So if you don't have a need for docker containerd would probably be the best choice. To make things more confusing dockerd is the engine of the app "docker" which you can manage with the docker cli application. Containerd is also just an engine and still needs a tool to interact with it and since it's focus is being as neutral as possible there are many apps to "control" containerd such as crictl and nerdctl. Hopefully this help clear some confusion

perfectly, thank you so much

FYI Rancher Desktop is not Rancher. I say this because you say you have K3s running on two Pis. Whilst you could use Rancher Desktop to host Rancher it's the latter you need not RD

😅 back to square one

Also, I want to note that since you are saying you have 2 pis: K3s supports multiple datastores and if you use an external database. Such as MySQL this won't be a problem. But if you use the inbuild database for kubernetes called etcd you want an odd number of master servers: https://docs.k3s.io/installation/ha-embedded#new-cluster

To run K3s in this mode, you must have an odd number of server nodes. We recommend starting with three nodes

This is because etcd does a process called

leader-election

so the servers need a

quorum

. You can see the HA matrix and more info from the etcd docs: https://etcd.io/docs/v3.5/faq/#why-an-odd-number-of-cluster-members This was a large area of confusion for me when I started out so hopefully this will save you some trouble.

Could this be done using rancher in vm on my laptop and then the 2 Pis? Or I should get a 3rd Pi And if this would work with the first option, will it still work when I turn off my laptop? If I should get the 3rd Pi, will this work when I use rancher from VM on my laptop? :face_with_spiral_eyes: :face_with_spiral_eyes: :face_with_spiral_eyes:

No need to apologize, it can get confusing and there is a lot of confusing information online. But at its core Rancher is an app you install on top of a kubernetes cluster using a helm chart. Rancher is an app that manages other kubernetes clusters. There are a lot of prepackaged all-in-one rancher install solutions like the Docker install but we do not typically recommend this set up for more than a quick demo/testing due to the lack of flexibility of this approach. Normally we recommend installing Rancher using the Helm Method on a new Kubernetes cluster.

So for non-HA, I don't have to work about the even/odd database thing for now but when I expand the system, I still have to add 2 each time to keep it odd from then on? I still follow the Helm Method?

So for non-HA, I don't have to work about the even/odd database thing for now

correct! 1 is an odd number 🙂

but when I expand the system, I still have to add 2 each time to keep it odd from then on?

yes but at a certain point its not worth adding more master modes, you shouldn't need more than 3 master nodes for most setups. cons outweigh the pros at 5 or 7 because the additional network traffic from all the syncing can really bring things too a crawl on a typical network. I want to not that this odd node requirement is ONLY for the master nodes since etcd typically only runs on master nodes. if you add worker nodes - you can have any number of those as you want.