PaX [comrade/them, they/them]

This has always felt untrue to me. The command line has always been simple parts. However we cannot argue that this applies to all Unix-like systems: The monolithic Linux kernel, Kerberos, httpd, SAMBA, X windowing, heck even OpenSSL. There’s many examples of tooling built on top of Unix systems that don’t follow that philosophy.

I can see why you would come to think that if all you’ve been exposed to is Linux and its orbiting ecosystem. I agree with you that modern Unix has failed to live up to its ideals. Even its creators began to see its limitations in the late 80s and began to develop a whole new system from scratch.

Depends on what you mean. “Everything is a file”? Sure, that metaphor can be put to rest.

That was never true in the first place. Very few things under Unix are actually represented as files (though credit to Linux for pursuing this idea in kernel-space more than most). But Plan 9 shows us this metaphor is worth expanding and exploring in how it can accomplish being a reliable, performant distributed operating system with a fraction of the code required by other systems.

Kubernetes is more complex than a single Unix system. It is less complex than manually configuring multiple systems to give the same benefits of Kubernetes in terms of automatic reconciliation, failure recovery, and declarative configuration. This is because those three are first class citizens in Kubernetes, whereas they’re just afterthoughts in traditional systems. This also makes Kubernetes much more maintainable and secure. Every workload is containerized, every workload has predeclared conditions under which it should run. If it drifts out of those parameters Kubernetes automatically corrects that (when it comes to reconciliation) and/or blocks the undesirable behaviour (security). And Kubernetes keeps an audit trail for its actions, something that again in Unix land is an optional feature.

My point is Kubernetes is a hack (a useful hack!) to synchronize multiple separate, different systems in certain ways. It cannot provide anything close to something like a single system image and it can’t bridge the discrete model of computation that Unix assumes.

This also makes Kubernetes much more maintainable and secure. Every workload is containerized, every workload has predeclared conditions under which it should run. If it drifts out of those parameters Kubernetes automatically corrects that (when it comes to reconciliation) and/or blocks the undesirable behaviour (security). And Kubernetes keeps an audit trail for its actions, something that again in Unix land is an optional feature.

All these features require a lot of code and complexity to maintain (latest info I can find is almost 2 million as of 2018). Ideally, Kubernetes is capable of what you said, in the same way that ideally programs can’t violate Unix filesystem DAC or other user permissions but in practice every line of code is another opportunity for something to go wrong…

Just because something has more security features doesn’t mean it’s actually secure. Or that it’s maintainable without a company with thousands of engineers and tons of money maintaining for you. Keeping you in a dependent relationship.

It also has negligible adoption compared to HTTP. And unless it provides an order of magnitude advantage over HTTP, then it’s going to be unlikely that developers will use it. Consider git vs mercurial. Is the latter better than git? Almost certainly. Is it 10x better? No, and that’s why it finds it hard to gain traction against git.

So? I don’t expect many of these ideas will be adopted in the mainstream under the monopoly-capitalist market system. It’s way more profitable to keep selling support to manage sprawling and complex systems that require armies of software engineers to upkeep. I think if state investment or public research in general becomes relevant again maybe these ideas will be investigated and adopted for their technical merit.

Even an online filesystem does not guarantee high availability. If I want highly available data I still need to have replication, leader election, load balancing, failure detection, traffic routing, and geographic distribution. You don’t do those in the filesystem layer, you do them in the application layer.

“Highly available” is carrying a lot of weight there lol. If we can move some of these qualities into a filesystem layer (which is a userspace application on some systems) and get these benefits for free for all data, why shouldn’t we? The filesystem layer and application layer are not 2 fundamentally separate unrelated parts of a whole.

Nice ad hominem. I guess it’s rules for thee, but not for me.

Lol, stop being condescending and I won’t respond in kind.

So what’s the problem? Didn’t you just say that the Unix way of doing things is outdated?

I think the reason the Unix way of doing things is outdated is cuz it didn’t go far enough!

Dismissal based on flawed anecdote is preconception.

What? lol

It’s not a flawed anecdote or a preconception. They had their own personal experience with a cloud tool and didn’t like it.

You can’t someone into liking something.

I’d rather hire an open-mined junior than a gray-bearded Unix wizard that dismisses anything unfamilar.

I’m not a gray-bearded Unix wizard and I’m not dismissing these tools because they’re unfamiliar. I have technical criticism of them and their approach. I think the OP feels the same way.

The assumption among certain computer touchers is that you can’t use Kubernetes or “cloud” tools and not come away loving them. So if someone doesn’t like them they must not really understand them!

It’s hard to not take that as bad faith.

They probably could’ve said it nicer. It’s still no excuse to dismiss criticism because you didn’t like the tone.

I think Kubernetes has its uses, for now. But it’s still a fundamentally limited and harmful (because of its monopolistic maintainers/creators) way to do a kind of distributed computing. I don’t think anyone is coming for you to take your Kubernetes though…

That’s a great way of putting it, thanks. I’m actually only 30 years old (lol).

Yeahh, and I saw someone compare you to the “old man yelling at cloud” lol. Even though there are good reasons to yell at the cloud hehe

Sometimes I feel there’s so few people who’ve ever used or written software at this level in the part of the industry I find myself in. It seems more common to throw money at Amazon, Microsoft, and more staff.

I’ve replaced big Java systems with small Go programs and rescued stalled projects trying to adopt Kubernetes. My fave was a failed attempt to adopt k8s for fault tolerance when all that was going on was an inability to code around TCP resets (concurrent programming helped here). That team wasn’t “unskilled”; they were just normal people being crushed by complexity. I could help because they just weren’t familiar with the kind of problem solving I was, nor what tooling is available without installing extra stuff and dependencies.

I haven’t had the “privilege” of working for a wage in the industry (and I still don’t know if I want to) but I think I know what you mean. I’ve seen this kind of tendency even in my friends who do work in it. There is less and less of a focus on a whole-system kind of understanding of this technology in favor of an increased division of labor to make workers more interchangeable. Capitalists don’t want people with particular approaches capable of complex problem-solving and elegant solutions to problems; they want easily-replaceable code monkeys who can churn out products. Perhaps there is a parallel here with what happened to small-scale artisan producers of commodities in early capitalism as they were wiped out and absorbed into manufactories and forced to do ever-increasingly small and repetitive tasks as part of the manufacture of something they once produced from scratch to final product in a whole process. Especially concerning is the increasing use of AI by employed programmers. Well, usually their companies forcing them to use AI to try to automate their work.

And like you gave an example of, this has real bad effects on the quality of the product and the team that develops it. From the universities to the workplace, workers in this industry are educated in the virtues of object-oriented programming, encapsulation, tooling provided by the big tech monopolies, etc. All methods of trying to separate programmers from each other’s work and the systems they work on as a whole and make them dependent on frameworks sold or open-sourced™ by tech monopolies at the expense of creative and free problem-solving.

Glad at least you were able to unstall some of the projects you’ve been involved in!

Thanks for your understanding :)

Glad we could share ideas :3

You and other people in the thread gave me a lot to think about. Hope this comment made some sense lol.

Yeahh, you have a good point lol. Bash and the GNU ecosystem have developed their own sprawling problems.

I’m glad I at least got closer to understanding your criticism than they did.

Don’t let anyone tell you you’re old or naive or “stuck in the past” for thinking these things! There is a real crisis in the operating systems world that your criticism is reflecting. It takes an army of software engineers and billions of dollars to keep this ecosystem and these systems going and they still struggle with reliability and security. The reason it’s like this is an issue of economic organization.

We can’t go back to the old way of doing things but we can’t keep maintaining these fundamentally flawed systems either. You may find something inspiring in this brief presentation by Rob Pike: http://doc.cat-v.org/bell_labs/utah2000/

True, but a man page is a different thing from a tool’s built-in usage information.

You just said that this software was much more complex than Unix tools.

That’s the problem. The reason Unix became so popular is because it has a highly integrated design and a few very reused abstractions. A lot of simple parts build up in predictable ways to accomplish big things. The complexity is spread out and minimized. The traditional Unix way of doing things is definitely very outdated though. A modern Unix system is like a 100 story skyscraper with the bottom 20 floors nearly abandoned.

Kubernetes and its users would probably be happier if it was used to manage a completely different operating system. In the end, Kubernetes is trying to impose a semi-distributed model of computation on a very NOT distributed operating system to the detriment of system complexity, maintainability, and security.

Until you need authentication, out of the box libraries, observability instrumentation, interoperability… which can be done much more easily with a mature communication protocol like HTTP.

I agree that universal protocols capable of handling these things are definitely useful. This is why the authors of Unix moved away from communication and protocols that only function on a single system when they were developing Plan 9 and developed the Plan 9 Filesystem Protocol as the universal system “bus” protocol capable of working over networks and on the same physical system. I don’t bring this up to be an evangelist. I just want to emphasize that there are alternative ways of doing things. 9P is much simpler and more elegant than HTTP. Also, many of the people who worked on Plan 9 ended up working for Google and having some influence over the design of things there.

They’re not, and I’m disappointed that you think they are. Any individual filesystem is a single point of failure. High availability lets me take down an entire system with zero service disruption because there’s redundancy, load balancing, disaster recovery…

A filesystem does not exclusively mean an on-disk representation of a tree of files with a single physical point of origin. A filesystem can be just as “highly available” and distributed as any other way of representing resources of a system if not more so because of its abstractness. Also, you’re “disappointed” in me? Lmao

They can, and they still do… Inside the container.

And how do you manage containers? With bespoke tools and infrastructure removed from the file abstraction. Which is another way Kubernetes is removed from the Unix way of doing things. Unless I’m mistaken, it’s been a long time since I touched Kubernetes.

because rejecting a way of doing things based on preconception is a lack of flexibility

It’s not a preconception. They engaged with your way of doing things and didn’t like it.

in cloud ecosystems that translates to a lack of skill.

By what standard? The standard of you and your employer? In general, you seem to be under the impression that the conventional hegemonic corporate “cloud” way of doing things is the only correct way and that everyone else is unskilled and not flexible.

I’m not saying that this approach doesn’t have merits, just that you should be more open-minded and not judge everyone else seeking a different path to the conventional model of cloud/distributed computing as naive, unskilled people making “bad-faith arguments”.

Running grep without parameters is also pretty fucking useless.

The difference is grep is a simple tool that can take in text, transform it, and output it to a console. It operates in a powerful and easy to understand way by default (take in text and print lines in the text containing the search parameters). This vmalert tool is just an interface to another, even more complicated piece of software.

Claims to have a Unix background, doesn’t RTFM.

Since when do Unix tools output 3,000 word long usage info? Even GNU tools don’t even come close…

Translation: Author does not understand APIs.

The point is that these abstractions do not mesh with the rest of the system. HTTP and REST are very strange ways to accomplish IPC or networked communication on Unix when someone would normally accomplish the same thing with signals, POSIX IPC, a simpler protocol over TCP with BSD sockets, or any other thing already in the base system. It does make sense to develop things this way, though, if you’re a corpo web company trying to manage ad-hoc grids of Linux systems for your own profit rather than trying to further the development of the base system.

Ok. Now give me high availability

I would hope the filesystems you use are “high availability” lol

atomic writes to sets of keys

You’re right, that would be nice. Someone should put together a Plan 9 fileserver that can do that or something.

caching, access control

Plan 9 is capable of handling distributed access controls and caching (even of remote fileservers!). There’s probably some Linux filesystems that can do that too.

In the end, it’s not so much about specific tools that can accomplish this but that there are alternatives to the dominant way of doing things and that the humble file metaphor can still represent these concepts in a simpler and more robust way.

This reads as “I applied to the jobs and got rejected. There’s nothing wrong with me, so the jobs must be broken”.

This is the maybe the worst way of interpreting what they said. They can come and correct me if I’m wrong but I read that as: they have a particular ideological objection to this “cloud” ecosystem and the way it does things. It’s not a lack of skill as your comment implies but rather a rejection of this way of doing things.

I’ve struggled with many of the same tools. What we need is real distributed operating systems, like Plan 9, not increasingly complicated hacks and kludges to keep old-world operating systems relevant in a networked world.

In addition to the other answers, you might consider using getline() if you’re on a POSIX-compliant system. It can automatically allocate a buffer for you.