I made this decision at a startup (albeit when the eng team was ~30 people, and we had a monolith with ~10 supporting services). I wouldn’t do it again, even for the reasons stated in the article. The uniformity is nice, we were moving from apps running directly ec2 instances provisioned with ansible. Each time we spun up a new service it was a process to get the ec2 instances provisioned just so. But k8s is such a p…
Pretty sure if there was a simple alternative, people would hate it. Everyone initially wants thing A. But then they want to customize it to do all permutations and combinations n of A, B, C. They want it to be extensible. They want redundancy. They want orchestration. They want integration. It’s why practically every config file format eventually becomes its own scripting language. Even HTML started off simple — now…
What job interviews taught me about Kubernetes
91–100 of 233 posts
Re: What job interviews taught me about Kubernetes
#92One year ago I might agree that Kubernetes is an overkill but today? Ask your favorite GPT to generate manifests, get primary app into cluster with telepresence or execute straight from container and switch contexts and clusters like it's 90s again. One reason I dislike Docker Compose and Docker is lack of isolation. Yes sure if you put your arm deep enough you can get it, but on local k8s I can spin cluster per work…
Finally just bought a piece of my own hardware and got LLM to deploy k3s cluster on it.
I think diy homelab/hosting is more accessible than ever.
Cut costs on cloud spend and invest into AI spend.
For a solo dev on a budget, I think it just makes sense.
Re: What job interviews taught me about Kubernetes
#93Earlier quoted context omitted.
I use NixOS with nginx + acme / caddy, coredns and no docker anywhere. It's extremely homogeneous, easy to scale out (add another flake output, deploy to a new server, update DNS records). You could easily automate some of that with more nix, but I don't bother because that's already only like 50 lines of config. I have a strong preference for renting bare metal and it has served me extremely well.
I totally believe this works for you. But in your case, isn't NixOS just another declarative orchestration system like Kubernetes? Similarly I can just run a standalone nginx, caddy with acme, and a coredns pod in a bare minimum k8s cluster. Personally, I think the complexity is on the same level.
NixOS has no "runtime" or controlplane to maintain.
Re: What job interviews taught me about Kubernetes
#94Earlier quoted context omitted.
I don't think you can provide all the features of Kubernetes while reducing the complexity. What is possible is to support a subset of the features of Kubernetes while making it easy to use. https://github.com/openrundev/openrun is a project I am building. It supports declarative deployments, on a single-node with Docker or onto Kubernetes. The target use cases is limited to standalone web app, like internal tools. N…
I look forward to the evolution of your project into a less standardized Kubernetes as end users request more and more features of your project.
Targeting a specific use case (internal tools) should hopefully help avoid feature creep. Also, the goal is that an OpenRun config should work on a single-node with Docker and with Kubernetes. That limits the types of features which can be implemented (for example no Docker Compose support, no Helm support).
Re: What job interviews taught me about Kubernetes
#95Earlier quoted context omitted.
Strong agree, if there's one thing LLMs are excellent at, it's writing Terraform and Kubernetes deployments (and/or helm charts). What used to be half a day of research, trial and error, is now 20 seconds of AI churn and 98% of the time it nails it on the first try. And then point it at grafana and tell it to write you a dashboard for the new service/s. Easy peasy lemon squeezy. What used to require a team of 4 devop…
LLMs are pretty bad at writing those things in my experience. They will invent HCL syntax that doesn't exist, generate absurdly overwrought Helm charts, put in assumptions that don't make any sense, and so on. It's faster, and better quality, to write the stuff myself.
Which LLMs?
Re: What job interviews taught me about Kubernetes
#96The grass is not always greener. I would change the Service Fabric crap I have to deal with any day to K8s.
Re: What job interviews taught me about Kubernetes
#97I made this decision at a startup (albeit when the eng team was ~30 people, and we had a monolith with ~10 supporting services). I wouldn’t do it again, even for the reasons stated in the article. The uniformity is nice, we were moving from apps running directly ec2 instances provisioned with ansible. Each time we spun up a new service it was a process to get the ec2 instances provisioned just so. But k8s is such a p…
Nomad, Consul and Vault all running on VMs that you manage with Terraform. The problem is that when you run this long enough you want K8s features anyway.
And your starter “production” deployment of the Nomad/Consul/Vault stack is literally 12 VMs, comprising three independent Raft clusters. There is no decent way to do zero-downtime instance replacement without building your own orchestration layer, but also they’ve had a years-long track record of shipping bad upgrades and following up with only manual remediations or workarounds instead of a fix.
As someone who has productionized and maintained truly hundreds of those clusters across several jobs, it is hard at this point for me to recommend Consul, Nomad, or Vault to anyone serious about building reliable applications. Too many broken upgrades and manual click-ops tasks just to keep them online. (…and I’ve said nothing of the actual product!)
Re: What job interviews taught me about Kubernetes
#98I made this decision at a startup (albeit when the eng team was ~30 people, and we had a monolith with ~10 supporting services). I wouldn’t do it again, even for the reasons stated in the article. The uniformity is nice, we were moving from apps running directly ec2 instances provisioned with ansible. Each time we spun up a new service it was a process to get the ec2 instances provisioned just so. But k8s is such a p…
>I think there is a real hole in the market for a simple solution that lets you deploy some containers to some instances in a declarative fashion without all of that complexity
That's how I see it as well but it's really tough to go against the grain. I have a small enthusiastic community of users around Uncloud (https://github.com/psviderski/uncloud) who went full circle - fed up with k8s and came back to simple, boring declarative Compose deployments across a handful of interconnected hosts.
Uncloud is essentially a cluster version of Docker Compose without a control plane and cluster management overhead.
Re: What job interviews taught me about Kubernetes
#99I made this decision at a startup (albeit when the eng team was ~30 people, and we had a monolith with ~10 supporting services). I wouldn’t do it again, even for the reasons stated in the article. The uniformity is nice, we were moving from apps running directly ec2 instances provisioned with ansible. Each time we spun up a new service it was a process to get the ec2 instances provisioned just so. But k8s is such a p…
I really wish there was an 80% kubernetes. I think you could get there with some changes:
1. No overlay networks. 1 IP per machine. pods use dynamically allocated ports, and the kubelet enforces pods listen only on their assigned ports using seccomp.
2. No kube-proxy or equivalent Layer-4 "load-balancer". It's not good, but it's often used. You should use some kind of Layer-7 load balancing instead. Also you need to look up the port number from (1). This also greatly lessens the need for DNS.
3. A better config language. YAML and helm templates are terrible. kustomize is built into kubectl, but it's frustratingly limiting and also still very complicated. Something like nix would have been great. This can make it easier to upgrade third party configs since you can have more logic to validate and merge your settings with upstream defaults or templates.
4. Maybe an EBF-like for the api server? If the built-in k8s objects don't have a setting for something, then you need to write an operator or control loop yourself and then run that too, which is a big lift. Over time, k8s just keeps adding more and more built-in things and then revising them, which creates a ton of churn. If you could easily script simple operations, then they wouldn't have to build in every permutation ahead of time. E.g. the HorizontalPodAutoscaler has 24 config object types with several fields each, but all it does is set replicas based on data read from the api-server, so it could be replaced by some kind of flexible script that runs in the control plane.
Re: What job interviews taught me about Kubernetes
#100Earlier quoted context omitted.
Dear friend, you have built a Kubernetes https://www.macchaffee.com/blog/2024/you-have-built-a-kubern...
There was once a time when we could deploy software without spinning up 3 etcd databases, multiple controller processes constantly running event loops, and a virtual networking layer, before you even get off the ground. Perhaps those days are behind us.
No way. In the future people will be able to vibe code entire businesses in optimized assembler