/tech/ - Technology★

>>886267 (OP)

Why would Illumos "of course" be welcome in a thread about BSD?

>>886272

That image is a violation of the FreeBSD Code of Conduct. Report to the Coraline Ada Ehmke Center for Genderfluidity, Feminism, and PoC Appreciation for reeducation and enrichment.

>>886272

Besides the hot FreeBSD mascot, what other incentives do I have to switch from muh comfy Ganoo Loonix to *BullShitD?

>>886283

>Why would Illumos "of course" be welcome in a thread about BSD

Because Solaris has BSD roots afaik. It is also a very niche system and hearing about it from its active users may be of interest to BSD users.

>FreeBSD Code of Conduct

Hello, fellow BSD user.

Good that this is not the mailing list, is it? If you don't know shit about BSDs and have no interest in using them, why are posting in this thread? You should take your outrage about this to their mailing list or reddit, tbh.

>>886283

>Why would Illumos "of course" be welcome in a thread about BSD?

Because it too is under a >>886290

OpenBSD is a meme.

>Filesystem

default FS doesn't even support SSD TRIM, and I don't think OpenBSD supports anything modern like ZFS or BTRFS.

>Security

"Only two remote holes in the default install!!!!!!!"

Yay!

I hope you realize that this literally only applies to a base system install with absolutely no packages added. In other words, not exactly representative or meaningful towards... anything really

>Sustainability

A few years ago, OpenBSD was actually in danger of shutting down because they couldn't keep the fucking lights on. How could anyone see this as a system they could rely on, when it could be in danger of ending at any time?

>Standards-compliance

"B-But OpenBSD is written in strictly standards-compliant C! Clearly that's better than muh GNU virus!"

So you're not allowed to create extensions to the standard? You should only implement the standard and nothing more? Keep in mind that this is nothing like EEE, as the GNU extensions are Free Software, with freely available source code, as opposed to proprietary shite. People should be allowed to innovate and improve things.

If you're gonna be anal about standards-compliance, then why let people make their own implementations anyway? Why not have the standards organizations make one C implementation and force everyone to use it?

>>886311

We already read your uninformed post in another thread, but thanks.

>>886272

What is the difference between UFS and FFS? OBSD was damn slow for me.

>>886312

>uninformed

nice damage control, BSD shill

>>886304

I'll try it out. Maybe I'll install it on my laptop if it's comfy enough. I need to use something with less distractions.

>>886320

If it's a Thinkpad you could try OpenBSD, as most developers use one. Should work well out of the box. Try the IRC channels if you run into problems.

>cuck license

>>886328

Some new TP don't even work with Linux.

>>886330

>communist cuck

>>886331

You could ask in the mailinglist or IRC (#openbsd on Freenode, usually pretty responsive and a lot of developers are in there too) about the specific model

OpenBSD "developers" just take GPL code, remove the license and slap their own on it. All for the freedom to serve their corporate masters.

https://slashdot.org/story/07/04/07/1618239/gpl-code-found-in-openbsd-wireless-driver

>>886339

OpenBSD is full of pretentious idiots.

>>886342

>both you and the guy spamming anti bsd shit are cuckchanners

>>886343

I'm read-only though. Can't into recaptcha.

This was enough for me to stop considering BSD family as a viable choice for production and home environment.

https://www.whonix.org/wiki/FAQ#Why_isn.27t_OpenBSD_Used.3F

https://www.whonix.org/wiki/FAQ#Why_isn.27t_FreeBSD_Used.3F

>>886267 (OP)

>hurr no license talk

>Also: GCC or Clang?

Protip newfag, the only difference between these two is license. Since I'm sure you and other newfags don't understand I will explain.

GCC and Clang are both the exact same gigantic tumors of c++ cancer. They are both on a race to the bottom in terms of generated binary safety and quality. They have little to no care about doing things the right way they only care about who benchmarks faster. Since this a BSD thread we will use OpenBSD as an example. Lets navigate to the source tree (/usr/src), now lets look and see what folder contains GCC and Clang. Thats right! Its the gnu folder. What does gnu stand for in OpenBSD land? Gigantic Nasty Unavoidable! If you don't agree with this try timing the build time of clang and modern gcc. Then compile gcc 2,3, or 4.2.4, and witness that it is multiple orders of magnitude faster than clang and modern gcc. Next you might question why OpenBSD deems them unavoidable. The only thing in OpenBSD base that requires a c++ compiler is in fact the compilers themselves no? Well this is true unfortunately c++ has weaseled itself in almost everywhere else. If OpenBSD wasn't to include it then many people would have to install one immediately afterwords and as far as I am aware these modern compilers can't even bootstrap themselves from a purely C compiler. Bjarne Stroustrup was a mistake

>>886272

>OpenBSD

>security is governing consideration for everything

This is incorrect. Simplicity and Correctness are the governing features for OpenBSD. As it turns out security is tangential to these things, however it is not the main focus of OBSD.

>linux-compatiblity-layer unavailable since 6.0 due to security

See above, it was removed because it was too complex. Security could have been fixed.

Otherwise I think the rest of the post is not bad.

>>886358

>tangential

parallel*

>>886358

>Simplicity and Correctness are the governing features for OpenBSD. As it turns out security is parallel to these things

Noted. Probably made other mistakes along the way - the comparison isn't really complete. I asked for input on those points, but there was only so much response

>>886358

Simpler and less code makes it also more secure. Nobody was using Linux compat feature, so it got removed. It's the same reason code for DOS and other ancient platforms got removed from LibreSSL. Useless clutter is a distraction (waste of time) at best. Besides, nowadays it's more common to run stuff in a VM than depend on these flakey compat layers that nobody is interested in maintaining. Or did you submit any patches?

>>886272

I see a lot of FreeBSD vs OpenBSD posting, but never much for NetBSD. I know it is primarily of interest for old hardware and the like, but has anyone here tried it on some things remotely modern (or at all really). How is it? Why would you use it over one of the other two?

>>886368

You just re-worded and expanded upon what I said.

>Or did you submit any patches?

Obviously not, I wasn't using it either. I stated it was removed because it was too complex, which along with being irrelevant is exactly why it was removed. If it was relevant then any security holes could have been fixed. You mentioned a prime example of this with libressl.

>>886378

My main issue with NetBSD is that some of their dev's don't realize when reported security vulnerabilities are in fact vulnerabilities. See the Illja Sprundel talk that gets posted for examples of this. I do love seeing all the classic hardware that the Japanese NetBSD users rice out with weebshit. I personally really want a luna68k or luna88k, but would likely use the openbsd port instead.

I installed OpenBSD on my laptop the other day.

Things are more or less the same as in Linux, but different enough to be confusing. Good mans help a lot though.

Had some problems with wifi it was access point's fault (I configured it for packets to be too short), though no other device complained about it.

Now I struggle to make jabber work. My server only supports SSL and uses self-signed certificate, but mcabber shits its pants when I turn TLS off, and doesn't really tell me what's wrong.

Maybe I should use this as an opportunity to try other clients, though there aren't many.

Outside of that, for a system commited to having sane defaults, it ships with some ugly fvwm config.

I want sucklessBSD now.

>>886353

>"Who are those many security professionals and how thoroughly is the code reviewed?".

6 to 12 at any time

https://www.openbsd.org/security.html

>As an example, see security vulnerability - NTP not authenticated.

This is definitely a problem for whonix, but isn't a problem for most people, overall this is a valid criticism.

>In addition, previously the OpenBSD website was not reachable over SSL.

OpenBSD website is HTTPS, all websites were previously HTTP...

>>886353

>muh https

Libtard/commie logic. Whoever wrote that page, and (you) should gas yourselves if you actually think a static site needs https. In current_year all their release sets are signed, and previously you could verify your set based on the pictures of the cds that were all always posted all over the mailing list. Of course you always could have cloned the source from cvs and built it yourself. Not that I should take anything they say seriously when they unironically claim that there are only 17 people using the OS.

>>886443

>(you) should gas yourselves if you actually think a static site needs https

If you do not use https on your site, you are making your users susceptible to MiTM attacks when they visit it. This means an attacker can add advertisements, add username / password fields, add donate buttons, or plain redirect to another compromised server.

>>886457

While I agree with this theoretical issue, https in its current state is not the solution. Navigate to openbsd.org and tell me who issued the certificate. It certianly wasn't theo or any member of the openbsd foundation. How can you expect anyone to trust an unrelated third party anymore than you can trust a static http site. In fact this is exactly the same issue you are pointing out with serving an unencrypted site.

sage because off topic.

>>886459

I agree with you that the whole certificate system is inherently flawed and can easily be abused due to the human element it involves.

While you could pull off the same attack by being able to sign certs. The attack for http is going to be much more common than the one for https, so at this current point in time you should be using https at all times instead of http.

>>886394

Why is it that NetBSD is more popular in Japan compared to other the ratio NetBSD has to the other BSD's elsewhere? Or do I just have overexposure to that side of it?

>>886534

This is just me guessing, but there was a time when their commits for FreeBSD were being rejected. Another reason could be that they like the support for exotic architectures.

>>886534

The Japanese are Brand-fags, in that they'll prefer a Commercial offering over a public-domain/folk product. In the context of computers, that meant they were preferring Solaris/IBM-Linux for their IT servers, so FreeBSD wasn't a thing. FreeBSD didn't put much effort into their non-English markets, either.

Japan did however have an established Academic 4.x BSD scene. Sony even had a successful line of 4.3 BSD UNIX Workstations (https://en.wikipedia.org/wiki/Sony_NEWS) that borrowed from their homegrown talent (and went on as the core technology of the Playstation 1, 2, etc.) The Academics, who by then were big into IPv6 research, made NetBSD their platform of choice, and the disproportionally big NetBSD hacker community is due to the community those guys fostered.

>>886466

What is the point of the whole certificate system anyways? I never understood the need for them.

>>886574

Part of the fun is finding one. Its more of one of those things that I can look off in the future and think about finding. For something more attainable I would like to get a sparcstation. There are probably a lot more useful things I can do with one of them instead of a luna. Plus I already have an m68k Sega Genesis.

>>886576

>>It's much smaller, cleaner and more portable than OpenBSD

I'm not so sure about that. In some places they definitely are(non-binary true/false) but in others they have loads of legacy cruft from trying to support every single architecture ever thought up.

>>886534

I found the episode of bsdnow where they talk about it. Hopefully your gookenglish is up to par: https://youtu.be/uvplmGGAPvU?t=37m

>>886637

The point is to be able to pair a person or company that exists with a pubic key they have. This is unlike the way ssh works and more like the way code signing works in the Windows world.

When you first make a connection over ssh, you'll get a message saying what the server's public key is and whether you trust it. If you accept it, your client will remember it for further connections. If an attacker were to MiTM the first connection to your ssh server they could give you the public key of a reverse proxy which they log what you type. In this scheme, cryptography ensures that the person you are communicating to is the only person to be able to send or receive cleartext messages, but not to verify who the person you are talking to is.

Code signing on the other hand, like certificates for HTTPS are given to people / companies. This allows you to theoretically verify who you are talking to and not just ensuring that the person you are talking to is the only person you are communicating to.

>>886642

Thanks for that (NetBSD in JP); it was interesting.

>>886272

You forgot the superior DragonFly BSD

>>886443

lack of TLS allows false signatures to be delivered

>>890198

>Some articles on their Wiki are old, laptop page doesn't seem very updated, but the Wiki itself and the guide seem good, all in all. IRC channel is friendly and ready to help out.

They are also very accepting when it comes to people helping out with documentation. Submit a documentation or article update to them and if its good they will publish it.I have found they are not as cliquey as other projects and any developer can update the wiki.

>>890243

Not how it works lad. You are expected to trust your initial release. It contains the keys for the next release. After that you have a chain established and no longer need to worry about the website. Though the mirrors containing the releases are mostly https so I'm not sure how that relates to the main site being unencrypted.

Personally I already have a trusted install with a cloned src repo so I can cut any release I need for other machines myself and never have to go through the mirrors signed by unknown trusted (((third-parties))) ever again. If you are a newfag I have no idea how you are supposed to obtain a trusted release, maybe from a personal friend, but we don't care about new users so figure it out yourself.

More info on release signing can be found here: https://www.openbsd.org/papers/bsdcan-signify.html

I used to use FBSD and OBSD variably last year. I liked it, but I needed Linux for what I do. How easy is it for me to cross compile packages, though? The software I'm thinking of is some client side stuff that interfaces with a server. Can't be too hard, right?

>>890287

>You are expected to trust your initial release

we both understand this is the problem

yet you do not provide a sufficient solution

thanks for trying

What text editor does Theo use? I'm curious.

>>890477

sublime

>>890431

You order an official OpenBSD CD.

You go to a hackathon or other event where you can meet developers or other well-known/trusted people, and get a copy from them.

At some point you have to trust something or someone, unless you're going to go straight to Theo's house and download code from his master server that's kept offline.

If you don't want to go through all that trouble or even pay for a CD, you can download different releases at different times and places, and cross-check. Get 2, 3, 5, a dozen this way, or whatever it takes. If one of them doesn't match, then something is up. But that's also a lot of work and trouble to go through. If you're just lazy, you'll just give up and state that it's impossible.

>>890654

>>890654

That's pretty funny.

>>886837

Can DF even be considered a mainline BSD?

>>890673

Compared to Mir, Hardened or Liberty, yes.

Free or Open for general desktop bullshit?

>>892370

Open, the developers actually use it as their desktop OS' so desktop programs are much more likely to work. Don't know if you are on a laptop, but on my chinkpad I get much better battery life with 'apmd -A' than I ever did with whatever was recommended on lunix and Free.

i always wondered, what does BSD offer over linux, disregarding security features? cleaner code? tiny install size?

>>892390

>cleaner code?

OpenBSD does

>tiny install size?

both open and net offer this.

Otherwise the main difference is a consistent base system, or rather a base system at all, and usable documentation. There is nothing better than being able to type man "x" and get useful and relevant information. I use both linux and openbsd regularly and holy shit linux man pages are fucking awful. Of course having the base system developed in the same place and by the same people as the kernel allows the software to actually use all of the features the kernel has to offer but this isn't something a new user is going to notice, unlike the man pages.

how the hell do you mount udf-formatted flash drives on *bsd?

keeps saying invalid argument everytime i try to mount /dev/da0

>>892370

Free has lots of binary packages and the ports system is probably the best of the bunch. If you absolutely need some specific software, you could look beforehand if both offer them. FreeBSD is more likely to support the latest hardware stuff, but it may depend. OpenBSD seems to also generally be a good choice for laptops - Thinkpad support is good as many devs use it, but I used an old HP and it was fine too.

OpenBSD has openports.se for searching, NetBSD has something similiar - Free probably too, you just have to look for it.

Also consider the "feature list", as all of them bring something different to the table.

>>892385

The way the system is laid out and configured also immediately makes sense. It's hard not to notice if you try it for the first time. This and the documentation cause you to solve problems on your own rather than consulting google everytime. /var/log is actually helpful and you don't need to wrestle binary logs (only had the misfortune to need to do that once, but it was enough). If you try it in a VM you'll probably see what I mean.

>>892443

Can't be of much help here, but in case you didn't try: the handbook/FAQ/guide may cover that. NetBSD has a wiki that answers a lot of spevific questions like this.

>>892443

You probably have to tell it the partition you want to mount. /dev/da0 is the buffered device name, and /dev/rda0 should correspond to the raw device. Use disklabel to display the partition table, and then pass the correct partition name to the mount program.

>>892390

openbsd taught me patience

Bug on FreeBSD's Berkeley Packet Filter used to exploit the PS4.

https://github.com/Cryptogenic/Exploit-Writeups/blob/master/FreeBSD/PS4%204.55%20BPF%20Race%20Condition%20Kernel%20Exploit%20Writeup.md

OpenBSD is hard-forcing IPv4 out of their next release.

What a fucking systemd-level decision.

>>892551

What's wrong with tossing obsolete stuff?

>>892551

Sauce? I seem to have missed that.

Either way one of the reasons I use openbsd is because they kill off old bloat.

>>892560

https://marc.info/?l=openbsd-cvs&m=152256582629837&w=2

100% of BSD UNIX is Old Bloat.

Find me one person on /tech/ who knows how to configure a home IPv6 network.

It's over-complicated shit that can be on as many iRetard Smartphones it wants, and would work well there-- as the phoneowners are paying for someone with 10x their IQ to manage their consumer media. But IPv4 persists in non-pro use because it just works, simple, and has actually uploads the ideals of what makes Unix.

>>892401

>base system

It's all well and good while you stay in standard Unix tools territory, but as soon as you step outside you either get programs you aren't going to use or you don't get programs that you really need.

Same exact problem as with Desktop environment, and why some users like installing minimal distros and selecting their own windows manager and other bits and pieces manually.

And speaking of windows managers, OpenBSD comes with 3 of them preinstalled, but I just want to run dwm. Dwm package comes configured to use xterm, so I rebuild it from source with my own changes. But doing it properly through ports is too bothersome, so I just "make install" but at this point the whole concept of base system goes off the rails.

>installing OpenBSD on an ME- and Spectre-ridden Intel platform

Isn't that kinda like trying to build a fortress on moving sands?

>>892585

>Find me one person on /tech/ who knows how to configure a home IPv6 network.

May ISPs have been pushing IPv6 down users' throats for quite some time. After 2020 the phaseout of IPv4 will begin in earnest and by 2030 it will probably be a legacy feature hardly or barely supported anywhere.

>>892585

>>892551

>>892560

>Changes by: benno@cvs.openbsd.org 2018/04/01 02:09:01

>04/01

Fucking April fools

>>892585

>Find me one person on /tech/ who knows how to configure a home IPv6 network.

I do. We're the last of the wise ones. Where did everyone else go? I hate this place now.

>>892592

Increasing your OS security still makes sense. In fact, it becomes even more critical.

Moving to other platform is better, but a lot of people already paid for Intel botnet and don't want to just throw the computer away yet.

>>892604

Kek, I was going to say I just upgraded from last weeks current to now and didn't notice anything.

>>892588

My point wasn't that you will never have the desire to stray from base; rather that having the kernel developed by the same people developing the 'coreutils' allows for those tools to properly use the features provided by the kernel. When comparing to linux this is a major difference since gnu coreutils is in no way integrated with linux (at least as far as linux exclusive features go).

>minimalism

I don't use those window mangers either but I don't really mind the extra disk space they take up since I have literally never used more than 15% of my OS disk. I will say that as hard as I try to lower my ram usage on gentoo (even as far as using musl and replacing most @system programs) I can not get it to use as little as openbsd.

>dwm configuration

You are supposed to cp -r /usr/ports/x11/dwm /usr/ports/myports/x11/dwm and edit it there. But I do agree that its annoying. I can never remember the make update-patches||upgrade-patches||update_patches etc command. I tend to run dwm-HEAD so I don't use ports for it. Also I use pkg for most things but I use ports for st so that I can apply my own config.h and its frustrating that it gets replaced every time I do pkg_add -u.

>>892617

>Kek, I was going to say I just upgraded from last weeks current to now and didn't notice anything.

The premise of throwing IPv4 out of the window appeared absurd straght away (and the timestamp only confirmed that). Theo is no idiot not to realize that IPv6 will possibly never completely supplant IPv4, and even if then not earlier then in ten or even twenty more years.

>>892615

Shouldn't getting rid of obvious botnet that sits underneath everything else be top priority though?

Does OpenBSD have a MEI driver (and if yes, it it enabled by default)?

>>892621

The

>"put it in right after unlock" deraadt@

had me going for a second though since I knew 6.3 was about to be released.

>>892624

*BSD loves proprietary though, they all load firmware and microcode blobs.

>>892585

>Find me one person on /tech/ who knows how to configure a home IPv6 network.

Been there, done that. IPv6 in networks with low security requirements (such as home networks) is a piece of cake. Campus networks with thousands of untrusted devices however...

>>892679

LibreBSD when?

>>892679

Every OS that supports "modern" hardware does this. If you want no firmware at all, you have to go back to early 90's or prior, but you'll still have chips with code in ROM doing effectively the same thing.

>>892958

False. Debian works on TALOS II. What matters is whether it can be updated, firmware in ROM can not and can therefore be treated like hardware. Microcode and the firmware in linux-firmware can and does get updated and therefore is software, which ought to be free like the rest of the system.

>>893007

>What matters is whether it can be updated, firmware in ROM can not and can therefore be treated like hardware.

Stallman is being full retard on this point.

First, putting Windows on a ROM chip doesn't make it less of a botnet.

Second, it can be updated, you just either need some extra hardware or replace chip manually (And back in the 80-ies some computers shipped software updates this way, just by sending you a new ROM chip with new firmware)

>>893007

>Debian works on TALOS II

TALOS II is the only open thing that exists, you have a dataset of one.

>>892679

>man fw_update: Since firmware with an acceptable license is already present in OpenBSD, fw_update exists purely to deal with firmware that may not be freely distributed with OpenBSD.

>type fw_update

>fw gets updated, including proprietary

truly the work of evil

Does OpenBSD always force its bootloader into the MBR? There doesn't appear to be a question where to put it during the installation process.

>>893014

>putting Windows on a ROM chip

Won't work you can try if you don't believe me.

It will still crash as soon as there's no more space, which hasn't been written too or deleted from, left.

It may take a while but it would happen quickly under normal usage.

>>893014

Who says hardware should not be free? Anything on a ROM is an unchanging part of the hardware, unless you physically replace (parts of) it.

>>893111

You trust hardware vendors once, software vendors (which includes updateable firmware) in perpetuity.

>>893007

What's the percentage of Debian users running it on TALOS as opposed to Intel/AMD and ARM? Not only that, but how many current such users are going to upgrade to a $5K machine designed for corporate use?

In the end, the firmware is still going to be a common problem, until we get stuff like RISC-V commonplace and affordable. And that's assuming it's done correctly. IIRC, TALOS has at least one component that requires firmware blob.

Also this whole business of "it only matters if it can be updated" is bogus, because you can just as well have botnet hardware. The only thing you gain from no updates is that "good" hardware can't be easily subverted.

>>893014

I updated my Amiga 500 to KS 2 and WB 2.04 this way. ROMs came with my GPV SCSI controller+HDD. Wish I still had that computer. ;_;

>gcc or clang

Both are terrible

OpenBSD, because it's the only distro that makes any sense nowadays, they aren't scared to completely rewrite ssh if it sucks

>>892679

They don't love proprietary, they just want people to run safe and proper software, if you're a company please don't use some shitty insecure implementation because of licensing issues, rather use this implementation, because in the end, if a company get's hacked and peoples data gets leaked because they outsourced their ssh client to India, who really get's fucked?

>>893173

You benefit from proprietary software. Stallman benefits from proprietary software.

>>893173

Not being able to use free software in proprietary software isn't True Freedoms, though.

>>893175

Until it turns on you and say "All your data are belong to us!". Haven't you been paying attention? Grindr, the non-free (((app))) for faggots and bug chasers sent HIV-status on an unencrypted channel, and you really should have heard about the Cambridge Analytica scandal by now, not to mention the exercise (((app))) MyFitnessPal that was breached and the crackers extracted dox on 150 million people. So it would seem that proprietary software does more harm than good, and should thus be shunned and not advertised. However the worst of them all is Windows 10. It collects way too much information (anything is too much, but still), and if that's not going to dissuade you, it constantly crashes, hanging you out to dry and in some cases is stuck in an eternal reboot cyclus. A GNU/Linux distro with command line only is more convenient than WIndows 10 bullshit.

>>893176

There is a stark difference between not offering proprietary software on a silver platter and making it impossible to run said software. I don't know of any free and libre operating system that makes it illegal to distribute proprietary that runs on such a system. The Linux kernel has a syscall exception[2], the GNU C Library is under the Lesser General Public License, and the GCC uses the Runtime Library Exception[3] so that even when GCC inserts some of itself into the software, proprietary software makers are still not bound by the GPL. The Bison parser also has similar permissions. So please stop spreading FUD: you can distribute proprietary applications for free operating systems.

(I know you are both trolling, and successfully baited me, but I think it's important to counter these assertions with facts.)

[1] https://itvision.altervista.org/why-windows-10-sucks.html#Windows10

[2] https://sfconservancy.org/blog/2017/oct/20/additional-permissions/

[3] https://www.gnu.org/licenses/gcc-exception-faq.html

>>893173

Because GPL is communist. Proprietary is national socialist.

>>893189

>writes a bunch of dumb shit that has no point

You benefit from proprietary software. Stallman benefits from proprietary software.

>>893210

>having no reading comprehesion whatsoever

>speaking like a malfunctioning robot

>providing no sources for controversial statements

>>893217

Sources? You provided anecdotal information as absolute truth. I provide an observation based on logical process. Have you ever looked in a mirror before?

>>893199

>>893210

>>893219

Has Google's neural net escaped, and we weren't notified of this breach?

>>893222

waste of trips

>>893199 (good dubs)

not me.

>>893223

>I'm a chatbot that thinks like a 12 year old and care about dubs, trips and quads as if they grant some mystical kekistani meme magic

>not saging an irrelevant conversation

>>893225

Just because you disagree with the facts does not mean the conversation is irrelevant. It is now apparent that you've never looked into the mirror before.

>tfw still can't mount the ntfs external hd I used to store media

I only keep Gentoo around for when I need to mount it, then I transfer the data to a usb, and then it ends up on my OpenBSD box. It's a shame.

>>893267

It should work. NTFS has been enabled in GENERIC kernel since a long time ago. Are you passing "-t ntfs" to mount, along with the right partition from disklabel?

If you have interesting details, report it as a bug, but it will probably be low priority. Most people use FAT32 for external HD, since that's more portable, and resilient to sudden disconnection (no need for FS check).

>>893389

Yes, I've tried the following (there are only two partitions, c and i):

doas mount -t ntfs /dev/sd1i /mnt

doas mount -t ntfs /dev/sd1c /mnt

It still fails to mount -- but the exact same command except with '-t msdos' mounts my usb just fine. In all honestly, I feel like it may be too captious an issue to send a bug report for.

>>893440

That sounds like FAT32 filesystem. They're usually on the 'i' partition.

BTW the 'c' partition represents the entire disk, rather than an actual partition. You might use it to image a disk with dd or stuff like that, but it's rarely used.

And the 'b' partition is reserved for swap.

>>893067

Anyone? How to control bootloader code install during installation?

>>893458

>BTW the 'c' partition represents the entire disk, rather than an actual partition.

Why? In DOS/Windows the first partition is C: because A and B used to be reserved for floppy drives, but why is c reserved for the whole disk in BSD?

If you had to list 5-10 most significant differences between a Linux distro and and a *BSD system and tell a Linux user who is new to *BSD, what would those be?

>>892585

> it just works, simple, and has actually uploads the ideals of what makes Unix.

CIDR is an ugly hack, just like BSD Unix

IPv6 is like beautiful Inferno

>>892710

it's already run by a spic on reddit

>>892624

> obvious botnet that sits underneath

hence coreboot

>MEI

why would it

>>892679

>loves proprietary

>ignoring Theo's wifi rants

>>893474

init system

package management

monolithic structure (BSD is more monolithic and integrated than Linux)

community attitude (newfags vs oldfags, router cucks vs webdev poofs)

corporate allegiance

abstraction preference (everything is a bitstream, BSDickheads like their bitstreams to be muh holy plaintext, Linux users prefer GUIs and blobby binary blunders)

>>893474

>cuck license

>>893470

>c is entire disk

You probably noticed that during installation too, unless you just resized the default partitions and didn't see it. I just accepted it at that time, but didn't look into the reasons tbh

>>893558

Also

>manpages and guide/handbook/faq are good, conf files have good manpages

>/var/log is useful for solving problems

>/etc contains what it should (generally things are where you expect them to be)

>>893554

>CIDR is an ugly hack

You'd rather go back a quarter century to classful addressing?

>>893555

>coreboot

Does it fully disable ME (or otherwise render it inoperable)? If so, why still the need for things like me_cleaner?

>>893555

>MEI

>why would it

Linux has had /dev/mei since many years.

>>893683

>CP/M doesn't support subdirectories

That's also why DOS makers thought "if we don't have directories then we can use the / symbol for something else" and used it to signal a command line switch. Later version of FAT of course had directories, but for compatibility reasons it was too late to go back and reassign the meaning of /, so they used \ instead as the directory symbol. What were they thinking? Were they seriously believing that if CP/M had no directories and DOS 1.x had none then they could get away with not having directories forever?

>>893683

>the raw, unbuffered device (aka character device)

How does doing I/O one byte at a time (rather than in blocks of arbitrary length) imply those bytes represent characters?

>>893914

It probably refers to the char datatype in C.

>>893922

It's some .Xresources tweaks someone posted on openbsd-misc some years ago, so I just copied it. That also means man pages in the text console remain black and white, unless you can think of another way to change them.

*VT100.Foreground:      white
*VT100.Background:      black
*VT100*colorULMode:     on
*VT100*underLine:       off
*VT100*colorBDMode:     on
*VT100*colorUL:         red
*VT100*colorBD:         magenta
*VT100*faceName:        DejaVu Sans Mono
*VT100*faceSize:        15

>>893469

>>893067

So there is not a single person here who could say something about how bootloader installation works in OpenBSD (specifically how can it be controlled not to overwrite the MBR)?

Also, has anyone successfully installed OpenBSD (or other BSD variants) to a logical drive rather than a primary partition?

>>894110

I dual-booted Windows XP and OpenBSD like 10 years ago (it was not my laptop so I didn't want to erase Windows completely). All I did was make a second partition, then copy the OpenBSD 512-byte boot block into a file on the Windows C:\ root directory, and add a one or two-line entry in the Windows boot config file. It was very easy, and that was enough to present an OS selection menu after the BIOS screen.

I guess the instructions here were probably modified for Windows 10, but it looks like the same basic idea, and they also give GRUB example:

https://www.openbsd.org/faq/faq4.html#Multibooting

You can probably just copy /usr/mdec/mbr (that's the 512-byte boot record) to whatever partition and directory your boot manager looks for such things. If it doesn't want a file but instead expects to find the boot block at the start of a disk partition, then you can just copy it there with 'dd'. Keep in mind that dd expects the raw device (like /dev/rsdXY, not /dev/sdXY).

>>894217

Did the OpenBSD installation overwrite XP's MBR (and you had to restore it)? Or does OpenBSD's MBR code do pretty much the same what the DOS/Windows MBR does (do one thing and do it well - find the active partition and load its VBR), so XP's NTLDR was loaded if the XP partition was active?

>>894300

Can't remember. I probably saved the original boot block to a file with 'dd' before doing OpenBSD install, just to be safe. But I don't remember what OpenBSD's installboot does when you don't select "whole disk" in the install script.

>>894491

it makes the openbsd area bootable but that's it, doesn't mess with any existing bootloader (grub, lilo etc)

>>894514

Ok, so if you don't use the "whole disk" option but install it to an 0xA6 type partition then it will put some boot code into the superblock of that partition, but will leave the MBR alone (and then whatever main bootloader you use will be able to chainload the OpenBSD partition boot code), is that correct?