Useful Kubernetes Tools for 2020 | Hacker News Reader

Useful Kubernetes Tools for 2020

caylent.com

26 pointssharjeelsayed18 comments

I've been in contact with Kubernetes (either following development or actively managing clusters) since the 1.0 days and all I can think when I read a list like this is: wake me up in a year or two when consolidation is over because it's simply too much for a single person.

It's just a bad list. It includes hosted Kubernetes solutions, application unrelated to Kubernetes (which may run on top, think Postgres), deployment options, built-in metrics, and lots of other random things. It's as if someone googled Kubernetes and copy & pasted and first 20 pages without any understanding or theme. This looks like some PR person hired a a low-cost freelance writer to "write something about Kubernetes because it's hot and we need traffic, deliver it tomorrow" - It's like someone writing "50+ useful Python packages" and just listing 50 completely random packages.

I've been operating Kubernetes clusters for years now, and most of these are completely irrelevant to me, dead, or I've never heard of them. Don't equate bad lists with complexity.

Thanks for the insight. Do you happen to be able to recommend a better resource?

Would you consider sharing some of the tooling you have found useful for your context?

Although it says that this list is "for 2020" it seems pretty outdated at a glance through. They've got items like Twistlock which got bought by Palo Alto in July 2019, and another one in the security section they mention hasn't been updated since October 2018.

This article seems to be copied word for word from this article from February with a few other things added.

https://medium.com/faun/10-useful-kubernetes-tools-ddffa6208...

This is too much, but I'd love to hear any specific recommendations from HN'ers.

What makes your lives better?

Me, I'm looking for something to help optimize pod resource allocations, and visualize utilization of both pods and hosting instances at the same time.

> visualize utilization of both pods and hosting instances at the same time

Kubernetes already exposes all the metrics, so if you run Prometheus + Grafana you get that for free. You can also find pre-built Grafana dashbaords for that by searching a bit on Github (the official Grafana dashoard sharing page sucks).

Low effort outdated rubbish.

A lot of them are redundant, like there's minikube, kubeadm, but no k3s; and tail tools too. Also, people running k8s with service mesh, security, but without backup/restore?

https://k9scli.io/ is missing

Yep, was also wondering. One of my most used tools.

Before I even went to the article I thought to myself: If this list is missing k9s and kubectx/kubens, then the whole article is basically a fluff piece.

lo and behold... It's a fluff piece. And stolen from another source.

Low-effort clickbait. Flagged.

I love that it is "50+ tools".

For, you know, your streamlined and simplified devops ...

The lack of reasonable defaults and reliance on countless competing third party tools leads to staggering complexity and highly individual, company specific "devops" setups.

Some operations guys died and became garbage men. The clever ones adapted by creating incomprehensible systems, calling themselves devops and becoming unfireable once again.

It came out 6 years ago... I doubt the situation will improve.

Heroku and its competitors are the best we have, because they lack all this crazy customization. But as long as Kubernetes offers as much freedom as it does, it will never become simpler than it is now. All we get is more and more abstractions.

Clickbait

Reply on news.ycombinator.com