Kubernetes — LowEndSpirit

Introducing myself and ElfHosted

funkypenguin — Wed, 22 Nov 2023 10:35:48 +0000

Hi LES, I'm David,

I'm a self-hosting, kubernetes-loving geek, I've been publishing technical guides ("recipes") on Funky Penguin's Geek Cookbook since 2016, one of the most popular being the ever-popular "AutoPirate" stack

In anticipation of applying for a provider tag, and making a Black Friday offer, I'd like to introduce myself and my platform..

How it started:

For the past 2 years, I provided Kubernetes-based architecture consulting to Seedplicity, a seedbox provider who eventually shut down due to increased datacenter costs, and unanticipated capital expenses.

Unwilling to abandon my work , I took the plunge and launched a re-imagined appbox provider, leveraging previous work and my public designs, and building "in public" on Hetzner auction server infrastructure.

The most agonizing part of shutting down Seedplicity was supporting an unhealthy platform long enough for users to migrate TBs of data to their new provider. With ElfHosted, I decided to experiment with a "bring-your-own-storage" model... I'd provide apps using slick automation and all the resiliency I'd architected with Kubernetes, and users could then mount their existing could storage using rclone/samba, and interact directly with that storage from their apps. My thinking was that I wouldn't have to invest in loads of storage, and users wouldn't be locked into my fledgling experiment, and could leave without migration headaches whenever they wanted to.

How it's going

Over the past 5 months (You'll find monthly progress reports at https://elfhosted.com/open), and with the help of lots of initial "elf-a-testers", I've refined the model as follows:

Hetzner Storageboxes are my preferred BYO storage option, due to their proximity to the cluster infrastructure, their price, and their upgrade-ability.
I quickly discovered that I do need to provide storage, so I built a simple, scalable "ElfStorage" product, using a ceph cluster running on spinning rust HDDs (again, Ceph and Kubernetes providing the resiliency)
Torrent clients (and Plex, as of Oct, thanks to the Plex Hetzner ban) require a user to bring their own VPN credentials. A gluetun container is attached to each app pod, and all torrent traffic (and Plex's metadata/phone-home activity) is routed via the VPN.
Users pay per-app, per-day, using "store credits" (apps start at $0.05/day), which they then top up when necessary with real money (avoids payment gateway charges on teeny tiny amounts!)

It's not at all profitable yet, but so far the monthly reports are showing that we're progressing in the right direction!

Importantly, I'm having fun architecting a solution for which I'm the ideal customer (a busy geek, who likes to get right into playing with new apps without a hassle, and have them work reliably and consistently), and I'm enjoying meeting and befriending some of the more technical of our users, in my Discord server.

So that's me - happy to answer any questions about the design / details, and I hope to continue to contribute in future

new cheap managed k8s: Vultr Kubernetes Engine beta

chimichurri — Sat, 07 Aug 2021 09:31:40 +0000

https://www.vultr.com/news/Vultr-Kubernetes-Engine-Beta-is-Available-in-LA-and-NJ/
& https://www.vultr.com/docs/vultr-kubernetes-engine

The Vultr Kubernetes Engine includes the managed control plane free of charge. You pay for the Worker Nodes, Load Balancers, and Block Storage resources you deploy. Worker nodes and Load Balancers run on Vultr cloud server instances of your choice with 2 GB of RAM or more.

At first glance, it seems like a clone of https://docs.digitalocean.com/products/kubernetes/ and https://www.linode.com/docs/products/compute/kubernetes/get-started/ - or perhaps it's more the case that it's not just Kubernetes, but actually all three in general being clones of one another?

Anyway, as far as k8s related differences go, these are the ones I've been able to tell on the spot:
-available regions
-min. block storage (PV) size: 1GB @ DO vs. 10GB @ Vultr/Linode
-load balancers: 3 tiers (from 10k concurent connections @ 10 dollars / mo.) @ DO vs. 1 tier @ Linode (up to 10k concurent connections @ 10 dollars / mo.) and Vultr (10 dollars / mo.; couldn't find the specs)

Based solely on this, out of the three clones (if you will;) , DO seems to be the best one for managed k8s on the cheap. A more detailed comparison including OVH and Scaleway can be found here (warning, post from more than a year ago): https://atodorov.me/2020/06/14/comparing-kubernetes-managed-services-across-digital-ocean-scaleway-ovhcloud-and-linode/

Kubernetes or else on low end hardware?

Robotex — Mon, 23 Dec 2019 00:01:15 +0000

Hello,
I'd like to try setting up a HA cluster for learning purposes, however I'm not sure about the choice of software to use as base. I tried already once, a year ago, to create a Kubernetes cluster using Rancher on top of Debian on a VPS with 2 vCPU and 1GB RAM but the provider shut it off after a few hours for IOPS overload (probably memory swapping?), despite it was only an etcd agent node together with an another VPS as control plane without any service running nor created. For that reason, I had given up never booting it up again and left those VPS idle for the remaining year, believing that Kubernetes had a too high minimum hardware requirements to run.
Now that I got my hands on a VPS with double the resources and a dedicated from kimsufi, I'd like to try again and hopefully run something on it. My idea is to run a static website, a mesh vpn to secure the access, a load balancer and maybe something more serious.
So here are my questions:
1. I read on Google that people are able to create a cluster using only raspberry pis, how is that feasible? Because I own a rPI 3 and AFAIK it has only 1GB of RAM like my previous VPS, so if my previous VPS didn't have enough memory for Kubernetes, how could a few rPI even run anything on it?
2. Is there some way to reduce the resources needed for Kubernetes alone? Because so far, it seems to me that only people with big hardware and €€€ can learn this stuff, since from what I read it requires atleast 3 nodes to set up something that isn't just a toy far from reality.
3. What would be the best method to deploy Kubernetes on a VPS or bare metal server? I know that I can use Ansible to configure and install the required packages, however I heard also about distros like CoreOS, RancherOS, k3OS that are made specifically for running containers, though almost no provider has them available as template. Somehow I was able to mount succesfully a Live ISO of CoreOS (writing a iPXE script, setting boot media, adding ssh key etc) but a lot of stuff, especially configurations, still seems to me like reading arab. My ideal final situation would be that I would just need to press some buttons to install/deploy/upgrade.
4. Does Kubernetes allow to deploy a container only to a specific node? Let's say that I'd like to run a Minecraft server on Kimsufi so that it doesn't overload the lower specs VPS, is that possible?
5. I'm currently giving Portainer a try which uses Docker Swarm and seems to be less eager in RAM usage, so what does Kubernetes give that I cannot do with Portainer?
Thanks in advance to anyone who will be willing to answer my questions

#Kube100 by Civo

AaronSS — Mon, 12 Oct 2020 12:53:05 +0000

Marketed as the world’s first k3s-powered, managed Kubernetes service I started kicking the tires this weekend, and so far I am really impressed. Very easy to use, and fast

Figured some of you might be interested https://www.civo.com/kube100