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

Will Kubernetes Collapse Under the Weight of Its Complexity?

influxdata.com

Post reply on HN
Print view View original

181–190 of 213 posts

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#181

Post by nvarsj »

erikb wrote: post #150
nimish wrote: post #9

kubernetes is a classic case of a tool designed for consultants and companies to sell consulting services (Including cloud services, which is why every cloud provider leapt onto it). In like 90% of the cases when someone used Kubernetes, Docker Swarm would have easily sufficed.

Why is this so heavily downvoted? A clear, legit opinion to have. In my experience it's even like this: I have never seen any working kubernetes app. But I deployed my own hello world docker swarm app in 2 days. (And that is a fact, not an opinion. The opinion part is up to you.)

Because it's not remotely true, and is very cynical and biased without any evidence. k8s success isn't because of some consultancy marketing, it's because a lot of people are using it very successfully in production, myself included.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#182

Post by hedora »

pas wrote: post #156
discodave wrote: post #129

Earlier quoted context omitted.

Compare to setting up a Lambda application with a DynamoDB or Aurora database. Very simple, has limitations but you get a scalable distributed system for (almost) no time investment.

You can buy hosted k8s, one click, and your cluster is ready. https://cloud.google.com/kubernetes-engine/

But that’s expensive.

Plugging in numbers for my $1000 synology, which runs a half dozen dockers, a vm, and is off-site backed up (for $120/yr), it tells me that I’ll pay $135/month to run comparable kubernetes at google.

Even if you pretend my $75/month broadband connection is only used by the synology, and include power costs, the synology still wins by $10’s/month.

(I included 3TB of storage for kubernetes, and 6TB usable for synology, with 3TB used, and sized for the high-memory 4 core machine, since dram is the synology bottleneck, even with a low wattage cpu. The synology is intentionally over-sized and is basically idle, so “but scalability!” isn’t a valid complaint.)

Also, the 9’s I’ve observed were trouncing amazon for a while, though they may be catching up, thanks to comcast...

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#183

Post by itomato »

manigandham wrote: post #51

Microsoft Word is also incredibly complex software with decades of development and features, and yet it's just a word processor. Everyone uses a small subset of the actual functionality which is why the entire system can be complex and simple at the same time, depending on your needs. It's exactly the same with Kubernetes. It's just clustering software that ties multiple servers together to give you an PaaS-like work…

When things go wrong in Word, there's no need (or potential) to dig into the underlying code for analysis.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#184

Post by erikb »

jrs95 wrote: post #166
erikb wrote: post #153

Earlier quoted context omitted.

> The complexity of Kubernetes largely reflects the complexity of the problem. I can solve most of these problems with a simple Linux + systemd in a much more readable way and am finished much quicker, even now that I am using k8s full time for 2 years. So, no. What it represents is the complexity of the community having multiple sources of requirements, and the complexity of the landscape with everybody trying to ma…

How is systemd going to help you with service discovery, load balancing, autoscaling, automated rolling deployments, etc? That being said, if you don’t need any of those things you probably don’t need Kubernetes either.

> service discovery

only used internally and not in a scaling fashion

> load balancing

External component that is hooked into k8s in some way

> autoscaling

External component that is hooked into k8s in some way

> automated rolling deployments

I doubt that many people are at a point yet where they can reliably do that. There are basic assumptions baked into that feature like that the k8s cluster survives until the next update comes out.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#185

Post by erikb »

nvarsj wrote: post #181
erikb wrote: post #150

Earlier quoted context omitted.

Why is this so heavily downvoted? A clear, legit opinion to have. In my experience it's even like this: I have never seen any working kubernetes app. But I deployed my own hello world docker swarm app in 2 days. (And that is a fact, not an opinion. The opinion part is up to you.)

Because it's not remotely true, and is very cynical and biased without any evidence. k8s success isn't because of some consultancy marketing, it's because a lot of people are using it very successfully in production, myself included.

Oh didn't know we are in the comedy channel here.

A penguin walks into a bar, goes to the counter, and asks the bartender, "Have you seen my brother?"

The bartender says, "I don't know. What does he look like?"

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#186

Post by erikb »

kazen44 wrote: post #152
erikb wrote: post #142

Earlier quoted context omitted.

You can make sense of all the back and forth in k8s networking? That needs more than package capturing I suppose. How do you translate a million packages into something useful?

most of k8s networking is actually not part of k8s? They don't really have a integrated overlay network except for kube-proxy. (which is lackluster). If you are running k8s, you should run a proper overlay network first.

"Running k8s" means also running a CNI of course, otherwise k8s doesn't work.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#187

Post by erikb »

pas wrote: post #154
erikb wrote: post #128

Exactly what I'm always saying. It's also nearly impossible in an Enterprise IT environment to get Kubernetes working on your laptop. Minikube and Docker Edge both seem to fail way too often. As a developer one wants to spin up a system to work on, then work on it, then push results to some repo. And this loop simply isn't possible (yet?). Also what the author didn't mention is that even the vanilla k8s stuff is alre…

> It's also nearly impossible in an Enterprise IT environment to get Kubernetes working on your laptop. huh? What does that mean? Maybe that means the environment is broken, not the software you want to use/work on. > For instance, why should a developer even know what an ingress is and that it might be something he needs? Then probably that dev shouldn't work with k8s. At all. If the dev wants/needs a hello world on…

> > It's also nearly impossible in an Enterprise IT environment to get Kubernetes working on your laptop.

> huh? What does that mean? Maybe that means the environment is broken, not the software you want to use/work on.

That is what the word "Enterprise IT" means. If you ever work in a company that makes more than a million USD per year, you will find an imperfect network in a nearly unknown state, with proxies and firewalls making your life hard, and automatic reconfiguration tools/scripts/antivirus-software resetting everything to "not working" the moment you take your eyes of the config files.

People and Software who really want to make money in Enterprise need to be able to handle that somehow. If you develop software on your macbook in an environment with the complexity of a Starbucks Wifi nobody can actually use your software in Enterprise.

Btw. did I mention that Windows+Outlook+Lync is the high standard of Enterprise laptops? Forget Enterprise users if you are not developing ON Windows.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#188

Post by MBCook »

barrkel wrote: post #133
MBCook wrote: post #71

Earlier quoted context omitted.

There’s a big difference between a toolbox and a box with those same tools jumbled up inside. One is organized and sensible. The other is a mess. Just because they are both technically capable of the same things doesn’t mean we shouldn’t hold ourselves to the higher standard. (Haven’t used Kubernetes, just going with the analogy)

Don't pile in on k8s criticism without familiarizing yourself, please. IMO it's a bit like Java for the cloud, the write once run anywhere bit (same caveats, but the win is still there). It intermediates the differences between cloud platforms. That alone makes it strategically worthwhile even if it was overcomplicated. But I don't think it is. Its core control loop is simple in principle, and the concepts and extens…

I explicitly said I hadn’t used k8.

I was simply commenting on the “it’s a complex thing so it had to be complex” line of reasoning. It’s faulty, you can make complex thing easier with good design.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#189

Post by erikb »

tuananh wrote: post #130
erikb wrote: post #128

Exactly what I'm always saying. It's also nearly impossible in an Enterprise IT environment to get Kubernetes working on your laptop. Minikube and Docker Edge both seem to fail way too often. As a developer one wants to spin up a system to work on, then work on it, then push results to some repo. And this loop simply isn't possible (yet?). Also what the author didn't mention is that even the vanilla k8s stuff is alre…

it's been extremely easy with Docker for desktop. Just 1 click a way from "enable kubernetes"... in Docker preferences.

That worked for me for 2 days. Then a manager was breathing down my neck to debug his cluster. So due to stress I overwrote the kubeconfig file with my managers kubeconfig.

Now I can't find a way to get the docker kubeconfig back, resetting the cluster, reinstalling kubernetes etc does nothing, and now that I have uninstalled and reinstalled docker, kubernetes doesn't even come up anymore. I'm also not allowed to share debug information with the Docker company. No real help from Docker developers without debug info. Because they can't tell me how docker generates a kubeconfig or where it writes its logs without seeing my diagnostics output...

Not the vanilla perfect world use case is what makes your software but how well your software manages edge cases.

Re: Will Kubernetes Collapse Under the Weight of Its Complexity?

#190

Post by core-questions »

GiorgioG wrote: post #15
outside1234 wrote: post #3

> kubectl scaffold mysql --generate This exists, its called helm, which in fact delivers the productivity gains the author is looking for.

I have Kubernetes in Action on my desk and I haven’t cracked it open yet because Kubernetes seems monstrously complex. Sure helm gets you up and running. When something goes wrong in prod at 3am, what do you do?

This is my issue as well. I'm not comfortable running magic commands that create vast realms of infrastructure that I am responsible for but do not understand. It's one thing if it's all going to run on someone else's Kube, but even still, I need to be able to troubleshoot the apps on top, which are my responsibility.
Post reply on HN