Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

microk8s.io

Post reply on HN
Print view View original

31–40 of 138 posts

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#31

Post by tux3 »

mradmin wrote: post #27
kalev wrote: post #18

Having a RaspberryPi laying idle in a drawer somewhere, I wonder if others have installed microk8s or k3s on it. What kind of workloads are you running on it? Whoever has OpenFaas installed on your rpi, what type of functions are you running?

Can you run microk8s or k3s on a single server/device? Even if you can, seems like the wrong tool for the job with unnecessary complexity...

I run k3s on a single node. Used to be two, but I consolidated because this isn't a production usecase for me. Availability isn't the point for me. If I have to turn the thing off for an hour ever year or two to fix something, sure, fine.

The real value I get is Infra as Code, and having access to the same powerful tools that are standard everywhere else. I can drop my current (dedicated) hardware and go publish all my services on a new box in under an hour, which I did recently.

From my point of view, I already pay the complexity by virtue of k8s being The Standard. The two costs of complexity are 1) Learning the Damn Thing 2) Picking up the pieces when it breaks under its own weight. 1) I already pay regardless, and 2) I'm happy to pay for the flexibility I get in return.

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#32

Post by daitangio »

Fiahil wrote: post #23
m1ghtym0 wrote: post #2

How does it compare against minikube? Seems to be based on Snaps vs minikube's VM-based approach. Any other major pros/cons?

I thought minikibe would not cluster multiple machine ?

Yes, it seems not possibile. You can EMULATE a multi mode cluster but inside a big host machine.

Minikube is very effective on my huble opinion to STUDY K8s becauese it is always strong aligned with K8s releases.

I do not think minikube is a good choice for production but hey, I could be wrong... someone want to share any experience?

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#33

Post by tannhaeuser »

kissgyorgy wrote: post #6

If anybody seriously believe that Kubernetes is good on Edge and IoT, our industry is in deep trouble. I usually like Canonical, but this is next level bullshit.

Don't worry, as evidenced by recent posts eg [1] k8s has already entered the trough of disillusionment in its hype cycle for some time now, though Stockholm's is particularly strong on this one given its labyrinthic complexity.

[1]: https://news.ycombinator.com/item?id=27903720

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#34

Post by detaro »

polskibus wrote: post #25
detaro wrote: post #8

Earlier quoted context omitted.

Do you mean connectivity to the outside, or inside the cluster? The examples of Kubernetes and similar things in such scenarios I've seen usually had stable connectivity between nodes. E.g. an edge scenario would be one tiny well-connected cluster per location, remote-controlled over the (bad) external link through the API.

I meant intra-cluster communication between nodes, when some nodes are on the Edge, some are inside the datacenter. The Edge may have pretty good overall connection to DC, but have to work with intermittent connectivity problems like dropping packets for several minutes, etc., without going crazy.

Yeah, spanning a cluster from DC to Edge is probably not a good idea, but also generally not what I've seen suggested.

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#36

Post by gnfargbl »

Plain k8s has a fearsome reputation as being complex to deploy, which I don't think is quite deserved. It isn't totally straightforward, but the documentation does tend to make it sound a bit worse than it actually is.

I run a couple of small clusters and my Ansible script for installing them is pretty much:

  * Set up the base system. Set up firewall. Add k8s repo. Keep back kubelet & kubeadm.
  * Install and configure docker.
  * On one node, run kubeadm init. Capture the output.
  * Install flannel networking.
  * On the other nodes, run the join command that is printed out by kubeadm init.
Running in a multi-master setup requires an extra argument to kubeadm init. There are a couple of other bits of faffing about to get metrics working, but the documentation covers that pretty clearly.

I'm definitely not knocking k3s/microk8s, they're a great and quick way to experiment with Kubernetes (and so is GKE).

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#38

Post by antonvs »

kissgyorgy wrote: post #6

If anybody seriously believe that Kubernetes is good on Edge and IoT, our industry is in deep trouble. I usually like Canonical, but this is next level bullshit.

Canonical is just one of several players in this area. There's also k3s and k0s, for example.

I suppose it might depend on what you count as "edge", but we're using kubernetes to distribute a complex product to customers onprem. The product has multiple databases, services, transient processes, scheduled jobs, and machine learning. It needs to be able to run on a single machine or a cluster depending on customer requirements. It needs to support whatever Linux variant the customer allows. Using Kubernetes solves a lot of problems for us.

Re: MicroK8s – Low-ops, minimal Kubernetes, for cloud, clusters, Edge and IoT

#40

Post by ofrzeta »

cyberpunk wrote: post #14
ofrzeta wrote: post #13

Earlier quoted context omitted.

I had a good time with Kubespray. Essentially you just need to edit the Ansible inventory and assign the appropriate roles.

Sure, if it works, upgrades are somewhat fraught though (I mean, upgrading a 20 node cluster is an hour long ansible run, or it was when we were using it) We switched to rke, it’s much better.

An hour to upgrade a 20 node cluster doesn't seem unreasonable for me - when you are doing a graceful upgrade that includes moving workloads between nodes. I don't know anything about rke. Might be interesting but it seems different enough from upstream Kubernetes that you have to learn new things. Seems to me a bit similar to Openshift 4 where the cluster manages the underlying machines.
Post reply on HN