OpenBSD: Hackers Meet Soldiers

BSDForums writes "OpenBSD has a well-deserved reputation for fanatical security. Why is the U.S. military funding it? What do you get out of it? Cameron Laird and George Peter Staplin investigate and talk to Theo de Raadt, the creator, overseer, and taskmaster of the OpenBSD project!"

Comments

Conspiracy Theories

[Scoria]

Why is the U.S. military funding it? What do you get out of it?

Mulder, is that you?

Re:Conspiracy Theories

[Gortbusters.org]

You'll have to ask the man with the cigs.

Re:Conspiracy Theories

[chef_raekwon]

You'll have to ask the man with the [s]igs.sic

but..but..
everyone has a .sig here....

Re:Conspiracy Theories

[joe_bruin]

from the article (emphasis mine):

So even if a cracker accesses the system, she'll be able to reach only the Apache root

the correct gender-neutral description in english is "he". by saying "she", the author is specifically writing about females (to the exclusion of males), and implying that a female cracker would only be able to reach the apache root (presumably a male hacker would be able to go further?).

Re:Conspiracy Theories

[juan2074]

Why not try using 'whosoever'?

Re:Conspiracy Theories

[cymen]

Where have you been living?
Probably outside of academia. Nobody has time for navel gazing beyond the lunatic fringe.

Re:Conspiracy Theories

[samhalliday]

not america? just a guess. (you small-sighted americano who sees the international internet as 'american reads only'...)

Re:Conspiracy Theories

[slattont]

The prefered gender neutral is she/he/it. Which can be abreviated appropriately.

Re:Conspiracy Theories

[Bluesman]

Yes, but it's better than being called *gasp* insensitive! :-)

Re:Conspiracy Theories

[batura]

Actually, under one of the standard style guides (APA, i think), "she" is considered to be the prefered gender-neutral term.

And they call this bullshit progress?

Re:Conspiracy Theories

[Groganz]

What the fuck is wrong with using "they", after all it used in common speech, and language is created by its speakers.

OpenBSD Secure?

[bluemiracle]

I thought OpenBSD was designed with the idea of providing the most Open ported version of BSD rather than for fanatical security. But now knowing that, I have yet another reason to switch.

BSD is Bad Ass!

Re:OpenBSD Secure?

[LogicFlow]

I think NetBSD falls more into that catagory.
I remember hearing a good explanation of there "roles".
This isn't exact, but close enough.
FreeBSD, a sportscar. Hauls ass.
NetBSD, a hummer (or a jeep). Can go anywhere.
OpenBSD, a tank. I'd feel safe in one.

Anyone know who originally explained it similar to this? I'd like the original quote.

Re:OpenBSD Secure?

[b0r1s]

Hopefully someone from the FreeBSD project will port over propolice from OpenBSD, it'll be nice to see...

Re:OpenBSD Secure?

[guile*fr]

OpenBSD, a tank. I'd feel safe in one.
not quite. if you are in a tank you are more likely to be shooted at. especially if you happen do drive on a freeway.

Re:OpenBSD Secure?

[Anonymous Coward]

I'll take a hummer....please?

I'll give you a pearl necklace in return!

Re:OpenBSD Secure?

[evilviper]

Well, despite the fact that the analogy is very flawed, I should say that OpenBSD is just about as fast as FreeBSD, and works on a good number of platforms as well.

Re:OpenBSD Secure?

[vesamies]

Sure, but when it comes to sport cars a little
is a lot. Compare Free/Open on SMP system,
the fast lane is where speed matters. Sure,
there are "good number" of platforms but
nowhere like in Net. Even OpenBSD realizes it
is a tank. Only they picture it as a blowfish,
a fish that is full of security spikes,
can inflate itself with air, and is very
poisonous too. Of course the tank analogy
is somewhat flaved... where is the cannon
and missile systems! But its really a tank
a tank a tank, I think only, maybe the OpenBSD
guys are not very much war fighting people, but
a more fond of animals (like the object
oriented guys)...

Re:OpenBSD Secure?

[La Temperanza]

Also, OpenBSD still uses our trusty old friend gcc 2.95. While it certainly remains the best choice for secure & stable programs, the performance increases in code built with 3.x are nothing to scoff at.

Re:OpenBSD Secure?

[ebbomega]

Neal Stephenson was the first I heard who used the car-analogy, but it was to describe Windows, MacOS, BeOS and Linux...

Windows was a station wagon, MacOS was a sedan (with the hood welded shut), BeOS was Batmobiles, and Linux was a used lot where a bunch of volunteers were making tanks and lining them up on the side of the street with the keys in the ignition.

Re:OpenBSD Secure?

[dadragon]

Don't bolwfish inflate themselves with water? That is what they live in.

Re:OpenBSD Secure?

[cpeterso]

and here is Linux.

Re:OpenBSD Secure?

[SideshowBob]

Neal Stephenson was most assuredly NOT the first person to apply a car analogy to computer platforms :-)

Its (at least) as old as Usenet's comp.sys.*.advocacy groups.

Re:OpenBSD Secure?

[Anonymous Coward]

Actually propolice was available for FreeBSD (and Linux) first ... it's a GCC extension. OpenBSD just happens to be first to decide it worth integrating in the the OS base (there is a small performance penalty -- perhaps explaining why it hasn't been all that widely accepted in the worlds of FreeBSD and Linux.

Re:OpenBSD Secure?

[evilviper]

the *BSD's are even less functional than Linux.

Well, by that logic, Linux is even less functional than Windows by several orders of magnitude...

Re:OpenBSD Secure?

[evilviper]

NetBSD may support more platforms, but OpenBSD covers the most popular ones.

The vehicle analogy is more than somewhat flawed. You mention weapons, but the truth is that all are stationary systems, that can be attacked by anyone, and can't move out of the way. They do not wear down after tons of successful attacks, but rather are either broken with one, or remain perfectly intact at full strength. I could go on, but there's not much point.

As for the logo, I'm pretty sure the blowfish comes from the widespread use of Blowfish encryption in OpenBSD. The master.passwd uses blowfish by default, OpenSSH uses blowfish as one of the top cipher choices, blowfish is used to encrypt the swap partition (if encryption is chosen), etc.

If the OpenBSD team thought they were making a tank, they just might have used a logo of a FREAKING TANK.

Re:OpenBSD Secure?

[LogicFlow]

How is the analogy flawed? It's an analogy.
Daemon/Daemon/Blowfish wouldn't do very well.
Is FreeBSD making a little red man with a pitchfork? No, there making an OS.
Why the hell would OpenBSD think they were making a tank?
You completly missed the point.
And besides, what exactly does a generally hostile blowfish make you think of? Portability?

Re:OpenBSD Secure?

[evilviper]

How is the analogy flawed?

I already described a couple flaws.

LogicFlow: Why the hell would OpenBSD think they were making a tank?

I have no idea. That is what the parent post said, not me:

vesamies: Even OpenBSD realizes it is a tank.

You completly missed the point.

No, I have to say I think you missed the point.

what exactly does a generally hostile blowfish make you think of? Portability?

No. From the first time I saw their logo, it make me think of blowfish encryption. That was before I knew about their extensive use of blowfish, or even their whole encryption slant.

Re:OpenBSD Secure?

[LogicFlow]

"I already described a couple flaws."
Which missed the point.

"LogicFlow: Why the hell would OpenBSD think they were making a tank?

I have no idea. That is what the parent post said, not me:
vesamies: Even OpenBSD realizes it is a tank."

The point was the logo has nothing to do with the analogy, another one you missed.
"If the OpenBSD team thought they were making a tank, they just might have used a logo of a FREAKING TANK."

"You completly missed the point.
No, I have to say I think you missed the point."
I have to say you still do not understand what an analogy is.

"what exactly does a generally hostile blowfish make you think of? Portability?

No. From the first time I saw their logo, it make me think of blowfish encryption. That was before I knew about their extensive use of blowfish, or even their whole encryption slant."

Missed again. *sigh*. This is getting rather dull. So, you see a logo that reminds you of encryption. What exactly would that tell you about the OS?
Even with your strange ideas of the logos somehow matching an analogy, this still makes sence.

No more with this inane bullshit. Look up what an analogy is, and read it again.
There logos mean shit in this context.
NetBSD _does_ run on the most platforms.
FreeBSD _hauls ass_ especially on SMP boxes.
OpenBSD _is_ the most secure.
And unless you want to spend quite some time hand tweaking your OS this is how it is.
Dont give me shit about securing NT or such either. See above. It will have missed the point.
At that, don't give me anything. You missed the point, and I suspect it was on purpose simply to
initiate a childish pissing contest. Go piss on yourself. I'm done with this thread.

I can see the connection

[Anonymous Coward]

Imagine a Beowulf cluster of fork-touting BSD daemons... they'd call it an army!

Re:I can see the connection

[Blaine Hilton]

If they knew what was good they should be more afraid of that computing power. We can send leaflets to their high-tech sectors with specs!

Smart ships?

[benst]

Why not? They've tried it with Windows nt, which didn't work, so maybe there's more trust in open systems since then.

Re:Smart ships?

[0xB00F]

Reminds me of the South Park movie. But the army in that movie used Windows 98.

0xB00F, the sound of a cream puff exploding in my hand.

Re:Smart ships?

[sql*kitten]

Why not? They've tried it with Windows nt, which didn't work, so maybe there's more trust in open systems since then.

Unless OpenBSD has the magic ability to "do what the programmer meant, not what he wrote" when encountering a divide by zero, the Navy's application would have crashed in exactly the same way on OpenBSD too.

If you want to criticize NT, fine, go ahead, but you don't have to make stuff up.

Re:Smart ships?

[torpor]

Shh... they don't need that sorta magic in the MS camp... by now they covet the reboot...

Re:Smart ships?

[TheRaven64]

So do all the protected mode versions of Windows, i.e. NT, 2K and XP. Of course this only applies to user space stuff. If you're running any of your code in kernel space then all bets are off, but that's true of any OS.

Re:Smart ships?

[Uwe Barschell]

OpenBSD has "the magic ability" to not let an application crash the OS.

This is actually a very simple thing to do. Any OS designed for minicomputer-class hardware (e.g. VAX, RISC or 386+ CPUs) will include this magic ability (including NT). Such OSes will only crash if there is a bug in the OS itself, or in code that is treated as part of the OS.

One of the flaws with UNIX and NT, as compared to systems like VMS (with four protection modes) and Multics (with up to sixty-four protection rings), is the existence of only two protection modes: User and Kernel. This means that code which requires elevated privileges must be given the same privileges as the kernel itself, since Kernel mode is the only alternative to User mode.

This problem of only two protection modes is deeper than the OS design, however. Most RISC CPUs provide only two modes (the x86 provides four rings; the VAX provided four modes), so in order for an OS to be portable to such architectures, it must be limited to two modes like UNIX. This is probably why NT, which was designed by the architect of the four-mode VMS system, itself only supports two modes (like UNIX).

Remember that UNIX was considered very buggy and unstable in the 1980s, where as VMS (which is a younger system) was seen as rock solid. This reflects the design advantages of VMS, in being tied to the VAX architecture, with its four protection modes and robust instruction set, but that reliance on the VAX architecture was also a major weakness: unlike UNIX, VMS could not be ported to most RISC architectures, or the 386, and so only runs on VAX and Alpha. Both of these architectures support the four modes it requires, but are now niche CPUs with declining user bases. This limited hardware support was the most important reason for the decline of VMS, where as the portability of UNIX and NT were very important factors in their success.

UNIX, BSD, Linux and Windows 2000/XP show that a system with only two protection modes can eventually become stable, through simplified design and/or extensive testing on supported hardware configurations over time, but there is always the risk that new hardware will introduce new device-driver bugs, which automatically become new kernel bugs, thereby reducing any of the OSes to an unstable disaster again. The broader the hardware support is, the likelier it is this will happen.

Re:Smart ships?

[ClosedSource]

A very imformative post. Most people forget the role that hardware plays in OS stability.

If Linux had been created back in the 8088 days, either today's Linux would be incompatible with its legacy apps, or its stability would be comparable to Win9x.

Re:Smart ships?

[Tailhook]

If Linux had been created back in the 8088 days, either today's Linux would be incompatible with its legacy apps, or its stability would be comparable to Win9x.

Linux, as we know it, could never have been created on an 8088. In fact, the minimum x86 processor necessary for Linux is the 386. Linux, like Unix, requires virtual memory, preferably page based, and memory protection.

Linus deliberately set out to create himself an OS that followed the Unix model. He was unhappy with the Unix-like x86 OS implementations of the time and created his own. He clearly had in mind that his system would do as Unix does, not just look like Unix. You can make DOS look like Unix if you install enough GNU utilities, but it is fundamentally not Unix.

In a very real sense the stability of Linux, as derived from Unix, is by design, not simply because the coders are somehow better. By design, Linux proper can not operate on an 8088, and for good reason.

Note: today there are derivatives of Linux that can operate without hardware support of virtual memory. One important example is uCLinux. On systems without memory protection or VM support in hardware, the kernel suffers the same vulnerabilities to failures in user-land code as would DOS. These appeal of these systems is that they provide the POSIX API on very limited VM-less platforms.

Re:Smart ships?

[ClosedSource]

"Linux, as we know it, could never have been created on an 8088."

Yes, that was exactly my point. "Linux, as we know it" requires hardware features that didn't exist in the 8088.

The only detail we may disagree on is who did the bulk of the designing that makes the stability possible. I think most of the credit should go to the 386 implementors (although they weren't the first to accomplish it) rather than to the OS implementors that took advantage of the hardware features.

Re:Smart ships?

[Tailhook]

"I think most of the credit should go to the 386 implementors"

Intel no more invented TLBs than Linus invented fork(). The x86 ISA and the Unix design are both the result of countless prior efforts. They evolved symbiotically over many generations, using many hardware platforms. Attempting to ascribe any particular percentage of credit to one or the other is naive. It took decades of effort to arrive at the contemporary model. The truth is that IBM implemented most of the significant features of multitasking virtual memory systems, both hardware and software, before Linus was born.

Re:Smart ships?

[evilviper]

That's bull. Just because a program was made for an OS without memory protection DOES NOT mean that it will fail to run on an OS with memory protection.

For example, you can still run most DOS apps on NT-based system. If MS had done a better job, they could have preserved the ability of 99% of DOS apps to continue to work.

Re:Smart ships?

[evilviper]

This means that code which requires elevated privileges must be given the same privileges as the kernel itself, since Kernel mode is the only alternative to User mode.

Not really.

If you need to read from a device, a user account can be given access to a device entry. If you need access to a file, a user account can be given access to a file.

Then there is privlidge seperation. You can write a small, secure piece of code that runs as root, and invokes other, more complicated programs, with only the limited privlidges that program needs. Then there is systrace, which is simply a flexible, universal version of priv. sep., but nothing stopped you from doing the same before systrace came about.

There is also cylant, which has a purpose similiar to systrace, but go a bit of a different way about the same goal.

You VMSers really get on my nerves (yes, all 5 of you). Just because the kernel doesn't have the different levels of access built into it, does NOT mean that you are limited to two levels. It's just such a popular method because people are lazy... VMS would be just as vulnerable if admins got lazy and used only 2 of the privlidge modes.

So, your claim that VMS and multics have better security is like saying that MS-DOS can't be networked... Just because it isn't built-in doesn't make it impossible. In fact, I find Unix more secure because of the fact that privlidge seperation *is* seperate from the kernel. Then again, you can still be lazy and not make use of it.

Re:Smart ships?

[Tony-A]

most of the credit should go to the 386 implementors
Yep. The 386 architecture looks more like supporting something more like Multics than Unix, but to my limited knowledge, nothing even remotely takes advantage of it. The main advantage of a segmented address space is that things you shouldn't be messing with are not even addressable.

Re:Smart ships?

[ClosedSource]

I guess you missed the part where I said that the 386 implementors weren't the first to do it. My point was that the credit belonged to Intel more than to OS implementors like Linus or MS.

Re:Smart ships?

[ClosedSource]

Well, I didn't mean to imply that all applications would fail.

On the other hand, it's quite clear that real mode code in Win9x that has no reason to exist except for backword compatibility is a major cause of instability.

It easy to say that MS should have been able to make NT run 99% of legacy apps without any stability problems, but saying is quite different than doing.

Re:Smart ships?

[Uwe Barschell]

This means that code which requires elevated privileges must be given the same privileges as the kernel itself, since Kernel mode is the only alternative to User mode.

Not really.

If you need to read from a device, a user account can be given access to a device entry. If you need access to a file, a user account can be given access to a file.

Do you understand what happens when you access the device entry? Your user-mode code, running under whichever account, makes a request to a kernel-mode device driver, which then controls the hardware. The device driver runs in kernel mode, with the same privileges as the kernel itself. On a system with more than two privilege levels, the device driver could run with more privileges than user code, but fewer than the kernel.

Then there is privlidge seperation. You can write a small, secure piece of code that runs as root, and invokes other, more complicated programs, with only the limited privlidges that program needs. Then there is systrace, which is simply a flexible, universal version of priv. sep., but nothing stopped you from doing the same before systrace came about.

There is also cylant, which has a purpose similiar to systrace, but go a bit of a different way about the same goal.

You are confusing software permissions (e.g. user access rights) with hardware privilege levels.

You VMSers really get on my nerves (yes, all 5 of you). Just because the kernel doesn't have the different levels of access built into it, does NOT mean that you are limited to two levels. It's just such a popular method because people are lazy... VMS would be just as vulnerable if admins got lazy and used only 2 of the privlidge modes.

This is fundamentally wrong. A hardware privilege level is not the same thing as a user account.

So, your claim that VMS and multics have better security

I did not claim this. I claimed that a system which offers more than two privilege levels allows device drivers to run with more limited privileges than the kernel, thereby allowing for a more robust architecture.

is like saying that MS-DOS can't be networked... Just because it isn't built-in doesn't make it impossible. In fact, I find Unix more secure because of the fact that privlidge seperation *is* seperate from the kernel. Then again, you can still be lazy and not make use of it.

Your comments dont make sense, and show that you dont understand difference between hardware privilege levels and user accounts. UNIX relies on hardware privilege levels, just like other OSes.

Re:Smart ships?

[evilviper]

On a system with more than two privilege levels, the device driver could run with more privileges than user code, but fewer than the kernel.

As a matter of fact, it would be possible to give the device driver less privlidges in Unix. Once again, it just isn't done.

Your comments dont make sense, and show that you dont understand difference between hardware privilege levels and user accounts.

Well, actually I do understand the difference, I just made a mistake in interpreting what you were saying... possibly due to lack of sleep... The comments I made were countering the claims I've heard VMSers make dozens of times before. I should have read through your post more carefully.

Re:Smart ships?

[Uwe Barschell]

As a matter of fact, it would be possible to give the device driver less privlidges in Unix. Once again, it just isn't done.

Device driver code running in kernel mode has the same privileges as the kernel, and the kernel therefore cannot restrict it. This is why hardware architectures with more than two modes have been designed.

OpenBSD Installation

[djtrippin]

I've been wanting to set up an OpenBSD installation on one of my boxes for a few months now but the task of installation looks rather daunting according to their FAQ. Can anyone here who runs openBSD or who has installed it a few times and has a grip on the process email me with some thoughts about exactly how rough it would be for a BSD n00b, like myself, to do his first installation.

Re:OpenBSD Installation

[LogicFlow]

Just get a grip on diskslices. You'll laugh at everything else. The installation is beautyfull.

Re:OpenBSD Installation

[Anonymous Coward]

I've installed it quite a few times on DEC Alpha systems (back when it was kind of-sort of supported) , Sparc and x86 systems. It wasn't exactly rocket science. Just try it, its not like your computer will explode if something goes wrong.

Re:OpenBSD Installation

[(startx)]

I know you'll probably get a thousand responses saying this, but OpenBSD is the easiest install I've ever done. So just jump in there an do it!

Note: you will install it twice, because everyone fucks up there first install. everyone. period.

Re:OpenBSD Installation

[jazman_777]

I've been wanting to set up an OpenBSD installation on one of my boxes for a few months now but the task of installation looks rather daunting according to their FAQ.

OpenBSD is way easier than any Linux. Pf bodyslams iptables for ease. Package management (ports) rocks, Gentoo sees the advantage. Plus, development is not random (wink wink over to the SCO thing).

Re:OpenBSD Installation

[Tet]

Note: you will install it twice, because everyone fucks up there first install. everyone. period.

No, not everyone. My first OpenBSD install is still up and running as my home firewall, and has been fine since the day of install. Now my second OpenBSD install, that one went wrong[1] :-)

[1] For some reason, I couldn't get pf working the way I wanted it to (I tried to duplicate my earlier ipf rules, but it didn't want to know)

Re:OpenBSD Installation

[Edball]

Like many people have said, its a really easy installation, well doccumented in the faq. The most intimidating part to a newbie would be partitioning the disk.

The recommended method is creating individual partitions for /, swap, /usr, /home, /tmp, and /var. Deciding the appropriate sizes for each of these partitions when you have no experience is probably the hardest part - but there's plenty of recommendations online. Personally, I'd recommend 80MB for /, 300MB for swap, 500MB for /tmp, 1GB for /var and split the rest between /usr and /home (/home is where most of your personal files will be stored and /usr is where most packages are installed).

All of the comands are well doccumented during the install if you type 'help'. The only other thing that could cause some confusion to somebody new is that by default all drive input sizes are by hd sectors - Not Bytes. The simple way to avoid calculating everything is just append all partition sizes with a 'M'. This lets the system know that your number is in Megs, not sectors.

Hope that helps you out some.

Re:OpenBSD Installation

[Tony-A]

OK, I'd classify myself as still somewhat of a Linux newbie, but I have installed and run OpenBSD on a few boxes. Those were where it was easier to install OpenBSD than mess with some not-quite-cooperative hardware. The documentation looks intimidating but it is extremely complete and accurate. Probably the main hurdle is that if you are very familiar with DOS partition tables, OpenBSD does not think that way. Partition does not mean what you think it means. That said, its not that hard to bull your way through it. Just make sure you are using disks that you can afford to lose ALL content on. Currently I've got an older OpenBSD running on an older Toshiba laptop. No idea why other than at the time it was probably easier to get OpenBSD up and running that any of several versions of RedHat. Oh, you probably want the bash shell ;-)

Re:OpenBSD Installation

[evilviper]

When it asks you if you want to use the entire disk, say YES! As long as you use the entire hard disk, the rest of the installation is incredibly simple.

In fact, with that one exception (ie. if you say NO to using the entire disk), it is far less complex than even installing Linux.

Re:OpenBSD Installation

[djtrippin]

Thats the problem. I want to install to a partition on the drive. I dont need BSD raping my drive to pieces...

Re:OpenBSD Installation

[evilviper]

So, you are going to trust that you can successfully play with your partition tables... using an OS and an FDISK you've never even seen before, let alone never having used it.

I suggest you invest $5 in a hard drive. It will probably be the best investment you've ever made. Multilpe OSes on the same hard drive is just not smart, even if you are an expert on how to make sure they don't eat each other, there are many problems.

Re:OpenBSD Installation

[djtrippin]

No, I've pretty much decided I'm not going to install OpenBSD until the next format. As for Multiple OS's on the same hard drive. I have four on this hard drive and not a single problem in the year they have been sharing it for.

Re:OpenBSD Installation

[Shanep]

No, I've pretty much decided I'm not going to install OpenBSD until the next format.

Ahh, "the next format", you must be a Microsoft user! I remember the days where the biggest performance boost you could give a Wintel box was to reformat and reinstall once every 6-12 months.

I have four on this hard drive and not a single problem in the year they have been sharing it for.

The problem of one OS overwriting disk space of another on the same disk is possible, but a user doing this out of merely being human is likely.

I've booted between 7 different OS across two disks, trying to learn the differences between various OS and not had any trouble outside of the fact that some OS are finicky about only booting off the first disk and others finicky about multi-booting at all. But just because I had no baddly behaving OS doesn't mean it cannot happen.

But then... a few months ago, while installing my favorite OS (OpenBSD since 2.5) I managed to wipe out my Win2kPro installation. It came down to a mistake I made with OpenBSD's "fdisk". I think that is where the real danger is.

I still love OpenBSD.

Re:OpenBSD Installation

[raddan]

OpenBSD is a lot of fun, but unless you really enjoy digging around in a BSD system, you might want to try an easier-to-use BSD. I find that OpenBSD's post-install is the most complex part of a setup. I say cut your teeth on FreeBSD, which is a snap to install, especially if you have a bootable CD drive, and come back to OpenBSD when you are comfortable with a BSD.

Re:OpenBSD Installation

[djtrippin]

:-) The most dangerous factor in working with computers is the interface between the chair and the keyboard. I've learned this many a time.

Re:OpenBSD Installation

[Shanep]

And just think, within 24 hours, this factor will be playing out in a much scarier scenario, but it will be the interface between chair and control column.

Then again, I guess Nukes receive target info via keyboard...

Crickey.

I'm convinced

[solarrhino]

I never knew any reason to consider OpenBSD (versus FreeBSD or one of the Linux distros). But security is a pretty good reason. I think I'll put OpenBSD on one of my servers and see how it looks. Thanks, Slashdot!

Re:I'm convinced

[solarrhino]

Thanks for the insight. I still think I'll spin a copy up when I get some time - performance sometimes isn't the most important driver, in my experience - but I'll keep an eye open for limitations that you describe.

As for simplicity - I'm all for it! As Albert said, "as simple as possible -- but no simpler."

Re:I'm convinced

[TheRaven64]

You might also want to put it on a machine with nothing else on the hard disk. The OpenBSD install doesn't play nice with DOS partition tables. You can still use it on a machine with non-OpenBSD partitions, but you have to make absolutely sure you type in the partition boundries correctly. I've made enough typos without realising not to trust my ability to do this without trashing my disk. Oh, and FreeBSD's post-install configuration tools are (in my subjective opinion) much nicer.

Re:I'm convinced

[shadowbearer]

I've encountered conflicts between windows/dos partition tables and other flavors of *nix in addition to BSD. I suspect it's *DOS* that doesn't play nice with partition tables (particularly extended partitions - isn't there a warning in linux fdisk about this? ).

SB

Answer to your question ...

[SuperDuG]

Why is the US military paying ideology-driven foreign hackers?

In a nutshell, not everyone in the "government" is a complete idiot ... *gasp* ... and sometimes ... just sometimes these "agencies" come up with supporting something that is actually useful to them and what they're trying to do.

OpenBSD is designed with security in mind. The article goes into great lengths about OpenBSD and what they've managed to acheive.

Anyone who has read my comments knows that I'm pretty much a BSD cheerleader because when I start to work with servers I will always pick a BSD solution wherever possible.

For many reasons there is a level of obscurity (try explaining to a "1337 h4x0r" what a "wheel" is ...) which also goes along with that there is some differences in the file structure as well (slackware doesn't count).

Plus theres the stability, I know linunx is stable, but the BSD stability is tested for stability and there isn't any "new exciting" features plugged in and not tested (okay at least in OpenBSD ... NetBSD does NOT count for this argument *grin*)

And my absolute favorite NO MORE THAN YOU NEED is installed!!! Something that I have also been arguing over in the SuSE disucssion ...

So what do we have, Simple, Stable, and Secure ... KISSS!!

Go DARPA, I've got tuition to pay so I can't buy an OpenBSD CD Set this semester :-(, but I did pay income taxes (so I guess I did kinda fund OpenBSD!!!)

Re:Answer to your question ...

[imnoteddy]

In a nutshell, not everyone in the "government" is a complete idiot ... *gasp* ... and sometimes ... just sometimes these "agencies"
come up with supporting something that is actually useful to them and what they're trying to do.

Like the Internet.

Re:Answer to your question ...

[ratamacue]

OpenBSD was created through voluntary association. People supported the OpenBSD project because they wanted to support it, not because they had to. Why have we gone and ruined a perfect example of what liberty can accomplish? Why force consumers to support a product instead of leaving the choice with the consumer? If OpenBSD is truly a superior product then it will continue to thrive without the "help" (coercion) of government. If OpenBSD loses support, it is not because of a lack of coercion!

Now that OpenBSD has been subsidised by the US government, what can we expect? Strings attached. Government money is always accompanied by government control. I, for one, will never endorse a software product which the US government has assumed control over.

Re:Answer to your question ...

[ryanvm]

Out of curiosity, have you ever tried Debian? I'd be pretty surprised if it also didn't fit your operating system requirements of "Simple, Stable, and Secure".

Of course, it's not source based like the *BSDs, but that's why I'm playing with Gentoo now.

Re:Answer to your question ...

[Bishop]

I run Debian, OpenBSD, and FreeBSD. Debian is not even close to the simplicity of OpenBSD. In particular Debian suffers from a lack of a true default (complete) install. OpenBSD's default install is "everything you want in a *nix and nothing more." FreeBSD has a few more nice to have features. By comparision Debian is quite complex. Debian's philosophy of choice, choice, and more choice has its place (and is the reason I use it). However a side effect is that the Debian team won't come out and package up a default system that is well integrated and contains the basic unix tools and servers. OpenBSD has a smaller footprint then Debian for the same basic functionality because OpenBSD does not have all the support files Debian needs to seemlessly work with any of the 56 mailservers packaged in debs.

All three systems are equally easy to administer due to ports and apt-get. I do find that Debian is easier to keep current as apt-get/dpkg does a better job of upgradeing from one version to the next. Some admin basic admin tasks are easier under OpenBSD/FreeBSD as there is just less cruff to deal with. I prefer *BSD on my servers as I can "install and forget." I prefer Debian/testing on my workstations as it combines a good mix of current software and stability.

If you like Gentoo (which I also use) switch to FreeBSD. The packages in FreeBSD ports are better then Gentoo's ebuilds. FreeBSD aims for stability. Gentoo has a tendancy to apply too many bleeding edge patches. FreeBSD does have a completely different way of manageing the source, so it will take some getting used to.

Re:Answer to your question ...

[cgenman]

Why have we gone and ruined a perfect example of what liberty can accomplish? Why force consumers to support a product instead of leaving the choice with the consumer?

The government is not "forcing" anyone to use anything, they are subsidizing the development of something because it is A: beneficial to society and B: cheaper for the government to spur this development than to buy and attempt to secure copies of windows.

If OpenBSD is truly a superior product then it will continue to thrive without the "help" (coercion) of government.

Sorry to feed the libertarians (I'm sort of one), but in case you failed econ 101, 0 income - n expenditure = failure in the marketplace. If you need it / want it, you support it. That's what the government has been doing with software over the years, to our benefit.

If OpenBSD loses support, it is not because of a lack of coercion!

What?

Now that OpenBSD has been subsidised by the US government, what can we expect? Strings attached. Government money is always accompanied by government control. I, for one, will never endorse a software product which the US government has assumed control over.

I assume you have no idea why BSD is called the Berkeley Software Distribution? Or where Google got it's roots. Or the Arpanet for that matter.

Government money is not always accompanied by government control. Having friends in the civil service, I can confidently say the US government hasn't a clue where the domestic money goes, nor does it seem to care. Entire departments of useless people with useless jobs linger on for years without the slightest hint of going away, continuing to... say... monitor rural electrification in New York. You'd never get this level of freedom / incompetence in a private company. If anything, US funding is just about the cleanest dollars you can get, domestically. They don't even care if you finish. This isn't true internationally where I know we can be right bastards, but trust me, if the US government sets up a yearly 50k check to OpenBSD's headquarters, that check will be sent for 30 years after the project shuts down before anyone notices.

Re:Answer to your question ...

[ak3ldama]

I am in a similar situation with sound support. I am currently running Red Hat 8 Pro, since FreeBSD 7 wouldn't load on my machine (an ide driver problem), and the onboard AC97 isn't supported until FreeBSD 7 or -Current. Though RedHat is a decent OS once you throw out the compute-intensive (and unnecessary) stuff and switch to the Motif Window Manager. *BSD has always been my OS of choice, the simplicity of the file system tree and the kernel config file are quite nice.

Re:Answer to your question ...

[Tony-A]

LOL
Now that OpenBSD has been subsidised by the US government, what can we expect? Strings attached.
That does not sound like Theo.

Re:Answer to your question ...

[kcelery]

Terrorist also use telephone, so phone company should also be bombed, some terrorist went to MacDonald for a burger, ok, bomb the MacDonald. Terrorist drinks from the tap, ok, we poison the reservoir... uumm.. what's next??

Re:Answer to your question ...

[Bishop]

The organization I current work for and its 1000+ employee use more then just 4 OSs. Off the top of my head there is: WindowsNT, Windows2k, the Solaris servers, the IBM AIX CAD workstations, the SGI IRIX tape library, the squid proxy cluster running some linux, the old IBM OS/390 document server, the procurement and budget software runs on IBM MVS/AS400 something, Cisco PIX, various Cisco IOS, and whatever the firewalls run on. There is probably some older machines that has been long forgotten by all but the two people who use them. I have done work for several companies of this size and larger. All have had a large mix of operating systmes. Admittedly these guys are an older company so this example is a little extreme. They have had more time to accumulate systems.

It is too bad that you trolled as an AC. I would have liked to know how you acquired your vast knowledge of "work" that deemed 4 OSs to be too many. I suspect I already know the answer. And no /. and your high school computer lab don't count as real experience.

Least there shouldn't be any back door

[rf0]

I mean the US military is funding it. Commercial software I might be a bit wary of. Least with Open Source other people can vet the code to make sure there isn't any backdoors. You get the best of both worlds so all in all I'm up for this

Rus

Re:Least there shouldn't be any back door

[debrain]

Least with Open Source other people can vet the code to make sure there isn't any backdoors.

Not strictly true - the single point of failure is the compiler. If the Intel or gcc compiler has a smart 'exploit', (1) they can backdoor specific or general programs without an exploit in source, and (2) this exploit can self-propagate in the compiler, as the backdoor compiler compiles the new compiler, so once written the source for the self-propagating compiler exploit can be deleted. Donald Knuth did this with gcc(?), iirc.

Just FYI. ;)

Re:Least there shouldn't be any back door

[coyul]

Way offtopic here now, but it was Ken Thompson, not Donald Knuth. Here's the discussion in question: Reflections on Trusting Trust.

Also a summary entry in the Jargon File, for those who don't want to read the paper: http://www.catb.org/~esr/jargon/html/entry/back-do or.html

Re:Least there shouldn't be any back door

[kwoo]

Actually, you're thinking of Ken, one of the fathers of Unix -- and it wasn't gcc. :)

Re:Least there shouldn't be any back door

[ax_42]

Now if only gcc was OSS, then we could vet that as well......

Oh, wait :)

BTW the compiler backdoor you mention was described by Ken Thompson. Read about it here here among other places.

Answers

[ErikRed1488]

Why is the U.S. military funding it?
My guess would be that the military will either take OpenBSD, combine it with some code from the NSA, and make a really secure OS, or take some code from it and add it to an OS they already use.

What do you get out of it?
It's Free Software so we get to see the source code that's being developed as part of the project. We get to tweak that code, make it better, port it to another system, etc.

I think it's pretty cool the US Gov. is partially funding OpenBSD. I guess it's no different that government grants to universities for medical research and such.

Re:Answers

[MnO-Raphael]

I don't know anything about how/if NSA secure Linux development was stoppped, but it nevertheless looks pretty real to me...

Re:Answers

[xtermz]

My guess would be that the military will either take OpenBSD, combine it with some code from the NSA, and make a really secure OS, or take some code from it and add it to an OS they already use.

And my guess is that they will simply use OpenBSD out of the box, thus incorporating whatever developments are made by the gov't funded OpenBSD programmers.

I need to choose my words wisely here, but the govt isnt the big spender it used to be, at least in terms of developing their own solutions. Especially in the military, there has been a big move to use COTS based components whenever applicable. Command and control systems are running off of NT systems nowadays.

Back when they were running on Legacy systems, yea, alot of applications were made "in house". But you would be surprised if you were to walk through a modern military facitlity or a government agency the amount of x86 based hardware laying around.

Security?

[gillbates]

Adobe and Network Security Technologies, Inc., use OpenBSD, although many of them keep their choice private for security reasons.

Kind of like how Microsoft keeps its code private for security reasons too....

If BSD really is as secure as it has been touted, why keep your choice private "for security reasons"? Sorry, I don't mean to flame, but this statement has done more to hurt BSD than help it.

Re:Security?

[sacherjj]

Regardless of how secure your OS is, why help those trying to hack in? Attacks are going to be different for Windows boxes, OSx, Linux, BSD, etc. The more work you have to do to start getting in, the more trail you will leave be and the harder it will be in general to accomplish.

Re:Security?

[Tumbleweed]

>If BSD really is as secure as it has been touted, why keep your choice private "for security reasons"?

Security through obscurity should never be one's ONLY line of defense, but as anyone truly into security knows, it IS a good idea to have it as a PART of one's defense. There's absolutely NO reason, other than OS evangelism, to advertise what kind of security you have. It's not the business of businesses to worry about helping advertise their choice of OS or security technology.

Re:Security?

[Anonymous Coward]

no. This just shows that companies that are security conscious, take EVERY possible measure to protect their network. So - if these security conscious companies are taking the steps to protect themselves at this length by not disclosing their OS, or blanking banner messages to not identify OS type, software type or version information - then MAYBE that just goes to show that the people that take this whole security aspect seriously - also run oBSD.

This is NOTHING alike MS keeping code private. It's like going into a fight and telling your opponent that your left ribs are sore ... WHY detail your possible weakness, just so your opponent can take advantage of a known vulnerability??

makes sense? yaw? Whoever +4'd your comment should be shot. "hurts bsd" ... if anything, your comment weeds out non thinkers like yourself, from a very good product.

StatiK76

Re:Security?

[jazman_777]

Security through obscurity should never be one's ONLY line of defense, but as anyone truly into security knows, it IS a good idea to have it as a PART of one's defense. There's absolutely NO reason, other than OS evangelism, to advertise what kind of security you have. It's not the business of businesses to worry about helping advertise their choice of OS or security technology.

I'd add that obscurity only helps when _all other pieces_ of security are in place. That is, it's a bit of icing.

Re:Security?

[241comp]

Because "Security Through Obscurity" is the only method of security asside from a mathematically proven source which has a possibility of 100% success. I'm not saying it makes a good defense alone but coupled with the correct software it is excellent. Take this example:

1. You have OpenBSD which one can assume has a finite number of attacks which could be used against it as it is a finite system. Therefore, if there is a security flaw it WILL be found in finite time.

2. You have OpenBSD but no-one knows you do. They have to assume that you could have 1 of any number (a number which is constantly growing) of systems. These systems have a constantly growing number of possible attacks. In order for one to attempt every possible attack they have to attack faster than operating systems (all of them) are being developed. This is not really feasible which means that if there is a security flaw it MAY be found in finite time (it may NOT).

Now, would you rather have a system in which a flaw WILL be found or one in which a flaw MAY be found?

Re:Security?

[BiAthlon]

You want to keep it private so that other people (read non US friendly) don't get your source code and make computers that you can't crack. I doubt it has anything to do with securing the box.

Re:Security?

[Tony-A]

Kind of like how Microsoft keeps its code private for security reasons too....

Nothing at all like it, unless you're counting cases where Microsoft choses not to run their own code, but that's keeping the knowledge of which code Microsoft is running private rather than keeping the code itself private.

It's more like banks not heavily advertising what brand of safe they use.

Methinks it helps BSD more when those "in the know" quietly use one of them instead of blandishing their choice.

Re:Security?

[reinard]

Heh.

I guess it's no longer an issue now that they decided to "keep their choice [of OpenBSD] private for security reasons".

Btw, there is a difference between not making your OS very easily detectable and not letting anybody see the source so they could check it.

Any security expert will tell you that obscurity is not a good model for security, BUT it is a helpful first barrier. Just look how well it's working for MS. There are probably hundreds of bad bugs in their code, but very very few people will be able to exploit them (THE UNKNOWN ONES), because you can't just download the code look through it and find a bug and attack. I'm definitely not agreeing with this model, but something like making your machines unable to be OS-fingerprinted IS a useful security measure that will at least make things more difficult for the attacker, even though it could be considered as 'ADDITIONAL security through obscurity'.

Re:Security?

[myov]

Part of security is making it difficult for an attacker to comprimize a system. So why advertise your OS? You're just making it easier for the attacker, as they immediately know what exploits to run.

Installing OpenBSD is extremely easy.

[Anonymous Coward]

Ok, before I get started, let me say that OpenBSD developers and users have no patience for people who aren't willing to read documentation, since they pride themselves on it (even man pages) being extremely well written and kept up to date. If it's not, file a bug, and _IT WILL GET FIXED_ (of course if it's just _your_ problem, then don't file a bug regarding the project, file one in your own bugtracking system as something you need to work on).

That said, the following FAQ explains the installation process far better than anyone writing you email ever will be able to, including a complete install process in grey, which has example responses in bold [for the most part]. If you can't get it from this, then you aren't reading, and it doesn't matter if someone writes you an email message with the same thing (written more poorly no doubt). If you can't read and follow instructions, then OpenBSD is not for you, and honestly - you shouldn't bother.

Most people don't have this problem, but there are always some feeble minded folks who think that life is easier if they're spoonfed on IRC and the like. To such people: you aren't welcome. The answer to this attitude has already been given: don't ask questions that already have explicit, clear answers publically available.

If you have a problem with the instructions (not enough detail supplied, typos, etc.) then please let the OpenBSD developers know about them in order that they may be corrected. If _you_ have a problem, in that you can't understand them, well... maybe it's _JUST YOUR PROBLEM_. It might be something that you need to work on. Of course, there is an opportunity for things to be unclear, and in such cases - again, submit a bug: "such and such statement regarding fdisk is unclear, suggest more detail on partitioning so that xyz is unabiguous"

Now, if you -want- to install OpenBSD, go read:

http://www.openbsd.org/faq/faq4.html

Re:Installing OpenBSD is extremely easy.

[Anonymous Coward]

Let's hope that the OpenBSD folks respond to questions they don't want to answer by just ignoring them rather than wasting everyone's time by returning insults.

Perhaps they'd be better off developing a closed version so they wouldn't have to deal with the outside world of the feeble minded at all. Oh, but I forgot about that ego thing. I wonder, If there's a group of geeks in the forest and they have an OS but there's no one there to use it, are they still superior?

Re:Installing OpenBSD is extremely easy.

[Anonymous Coward]

This has nothing to do with OpenBSD folks responding (or ignoring) questions they don't want to answer.

This has to do with common questions, being answer _already_ and people not taking the time to read the answers.

Does it make sense to repeat oneself over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over and over, just because the question was asked again? No, it does not - this is why FAQ's came about, LONG AGO - in the time before OpenBSD (or Linux) there were still FAQ's.

The idea was, you could tell someone: read the FAQ here, and they would, and you wouldn't have to reanswer the same goddamn question 1000 times. OpenBSD has an excellent FAQ, and excellent man pages. They provide answers to commonly asked questions, any OpenBSD person will tell you to go read those sources and try them out. If they don't work for you, submit a bug report explaining why they don't work (install dies on xyz hardware for xyz apparent reason - help!). Someone will try to help you.

Asking people to email you a personal answer to a question, already answered is not welcome, because the answer is already provided, and people have likely told you where to look and to try that out first.

This has nothing to do with superiority or even OpenBSD itself, this is extremely basic learning skills, and if you don't have them - go away. Any OpenBSD 'folks' (well, developers) will likely tell you that as well, you are wasting their time.

Section 4 of the FAQ as linked in the original response is extremely clear and well written. It has a walkthrough of examples showing exactly what you can type to install a machine. Suggesting that someone go read that first, I think saves a lot of time for everyone.

Re:Installing OpenBSD is extremely easy.

[rabidcow]

because you can BUY a CD, boot off it, and have it automatically probe your system.

The only thing you have to do manually is partition the hard drive.

Other than that the install is a piece of cake.

Re:Installing OpenBSD is extremely easy.

[Arandir]

Monday - Mailing list FAQ is automatically sent out
Tuesday - Question covered in the FAQ is asked
Wednesday - 50 answers, ten of which are complete and detailed are sent
Thursday - The same damn fscking question is asked again

I will be more than happy to tell someone how to access the handbook, how to use man pages, where online documentation resides, etc. But if you don't even try to help yourself, don't expect me to be patient with you.

Unless you pay me $35 an hour. If you pay me $35 an hour I will patiently look up the information you need in a manual and recite it to you, just so you don't have to bother knowing what a manual is. For $55 dollars and hour I will even do it cheefully.

Re:Installing OpenBSD is extremely easy.

[Groganz]

Well, the answer is quite easy really isn't it: don't answer. If someone wants to answer let them. I am as pissed off by by idiots who repeatly do not use documentation as I am by people who jump on newbies with 'RTFM' without any link to the 'M'. There is probably someone who is quite willing to help out a newbie with the answer to a question. If you don't want to answer a question, then DON'T.

RedHat DII COE [ot]

[llouver]

For those of you interested in this topic, you should also be aware of RedHat's DII COE (Common Operating Environment) kernel available at DISA. The kernel is available at http://diicoe.disa.mil/coe/kpc/linuxpc.html

The creation of DII COE kernel for RedHat implies that there may be some pressure to accept GNOME as a valid component of the Joint Technical Architecture (JTA).

In other words, the military bureaucracy is beginning to accept the fact that linux is part of the modern computing landscape. (Watching the wheels of military technology turn is like watching grass grow)

Stop stealing idea's from SCO

[Billly Gates]

There is no possible way OpenBSD can be that secure and stable without stolen key Sco OpenServer source code.

No and ifs or buts. Its not like this technology is well known or taught.

After all, everyone knows that sco is the most stable, secure, and scalable unix ever made. All the great unix's borrow code from sco. There is no way Sun could of made solaris scalable without the ultra secure and scalable Xenix code. Just ask David Bois. Shesh.

Re:Stop stealing idea's from SCO

[jc42]

Hey, you need a smiley. The folks here are often humor-impaired, and would't recognize irony if it whumped them upside the head. You're most likely to just get a "troll" rating unless you make it obvious that you're writing with tongue in cheek.

Re:Stop stealing idea's from SCO

[irc.goatse.cx troll]

Moderation is for comments; the zoo system is for users.
-1, Troll is for posts that are a troll, not for trolls. If a user is a troll and you dislike trolls (How could you? we're cute and have long colored hair with a little diamond on our stomachs), add them to your foes list and set the foes modifier to -5.

Even trolls post useful/funny comments, discriminating against them is pointless since it will probably be hit with M2, or even other mods fighting you.

This article is disappointing.

[Anonymous Coward]

Not to harp on more publicity for OpenBSD, but this piece was a real letdown. One quote from Theo in the whole thing?! (note, I do not consider quoting terms such as "unix semantics" or "setuid program" to be substantive -real- quotes).

Maybe this will be useful to those who have never heard of OpenBSD, or are unfamiliar with its improvements for the past two years (only propolice incorporation is something more recent) - but for anyone with more than a cursory knowledge of the project, this is just not good journalism. Here you have an opportunity to have Theo answer your questions, and really get down to the meet behind the scenes, how the DARPA funding came about - how they approached him, whether there were any conditions to the work, if OpenBSD could use more of this funding, etc. But no, nothing, one quote - no new insight.

This might serve OK as an advocacy piece, and hopefully it will. But if you have two people "talk[ing] to Theo de Raadt" you would hope that they would have some more to talk about.

I find that reading interviews are far more enlightening than summary tripe such as this, because you're not just presented with a set of facts, but you get to hear information that goes beyond just the answers to questions. Often times, you then learn about things beyond the scope of the story, upcoming developments, sore spots. Say even a mention of how unfathomable it is that Sun has been holding back documentation to OpenBSD, given how many other private, public and governmental organizations (e.g. DARPA) that make no pretenses about support the opensource community are providing support to OpenBSD, whereas Sun is totally going against their own doctrine and ignoring OpenBSD developer requests (not even _offering_ an NDA as Linux et al have been presented with).

If this were a paper for a class or a personal site, fine no problem, what can a student or hobbiest do? But if you are in a position to provide journalism, it's really sad to see that power completely wasted in such a way.

Oh well, at least it can be added to the "OpenBSD is secure, free and neat, you should buy a CD" article pile, oh, I forgot to mention - continually overlooked. I guess there can never be too many of those, but it's sure starting to feel that way.

And -TWO- people wrote this article. Goddamn, two people, no brain.

OpenBSD isn't the only one ...

[jstockdale]

Although OpenBSD has recently gotten a reputition for being ubersecure, and thus this article about how it has been getting funds from DARPA, it is by no means unique. It seems that this perception of OBSD has come from its ability to do encrypted swap, and encryption in most faculties; however, it blatently neglects disk based security.

I'd like to point out that DARPA is also funding the FreeBSD project, specifically enabling the development of FBSD 5.0's geom/gbde functions, which enable a fully modular disk access system, and transparent drive encryption. Really cool features, and it looks like once the code gets a stronger review from the crypto community it should really open up the possibilites for securing FBSD.

Re:OpenBSD isn't the only one ...

[kryps]

And DARPA is also funding Reiserfs v4 development.

-- kryps

Re:OpenBSD isn't the only one ...

[edhall]

DARPA funding for BSD goes way back -- long before OpenBSD, FreeBSD, and NetBSD existed. One of the most important instances was DARPA's funding of the development of BSD's TCP/IP network stack back in the mid-1980's. This made BSD the first system in wide deployment that supported TCP/IP. It's hard to overestimate the affect that this has had on Unix and the Internet since then.

-Ed

Re:OpenBSD isn't the only one ...

[evilviper]

it blatently neglects disk based security.

I think not. Sure, Theo et al. haven't written a filesystem encryption program to be included in the base system, but so what? There are at least half a dozen filesystem encryption programs that function on OpenBSD.

You might as well say that OpenBSD "blatently neglects" GNOME, because it doesn't include GNOME in the base install.

OS Money

[L7_]

It would seem to me, that for 'enterprise' level government work (i.e. defense related software) that stability would be more of a requirement than speed, portability or feature-set.

You don't want your tank software blue-screening in the middle of a fight. "Hold on guys, don't fire at me for a second, I need to reboot my tank."

Alot of UNIX vendors have realized this, and they know that if they make products that the gov'ment likes that contains the features that they need, then they will continue to sell products. Lots of products. Its not so much a niche market, more like market share.

Re:OS Money

[dsplat]

You don't want your tank software blue-screening in the middle of a fight. "Hold on guys, don't fire at me for a second, I need to reboot my tank."

I think a second is an understatement here. I'm not trying to criticize boot times for various OSes either. Military computer hardware is spec'ed to be tough, but not always fast. Yes, they want fast computers for some things. But for battlefield stuff, hardening it against the conditions it is likely to meet is more important than raw speed. So I don't expect that it would reboot in a second if it bluescreened. By the time you have to reboot it, the situation is already pretty bad.

Re:OS Money

[ddama]

Actually, real time OS used by the military typically have reboot times expressed in fractions of a second. Standard, and even RT variants, of general purpose OS like *nix or Windows just don't cut it in those applications.

Hybrid vigor

[dsplat]

Fortunately, it's open source. We can learn from it and take the lessons with us to other code. While there are a lot of people getting mileage out of the amount of malware out there that attacks Windows, one of the reasons there is so much of it is that it is absolutely no challenge to find Windows machines on the net because of their sheer number. And many of them are poorly secured because Windows is the OS that is shipped on machines that are sold to people who have neither the knowledge to secure a computer nor the time to learn how.

There are several efforts to improve the security of Linux and *BSD. In the end, I think they'll benefit us all. Bruce Schneier talks about the window of exposure in his book Secrets and Lies. Efforts to improve the security of open source OSs have several benefits in reducing that window.

Some bugs will be fixed before they are ever exploited. A security vulnerability is still a vulnerability. But the damage is much less in this case.

Some bugs will be fixed faster after they are first exploited. Again, this reduces the damage that is done.

But in the long run, a greater benefit is the number of people who acquire some knowledge of how to analyze and test for security vulnerabilities and how to fix them. That is going to be greatest in open source. It provides the opportunity for competent programmers to wear the white hats.

Inform SCO

[gokulpod]

Quick, someone call up that SCO lawyer. Tell them that OpenBSD has got recognition from DARPA for security. I am sure they will file a claim of $1 billion against them too. The next day, the U.S. army will "accidently" test a MOAB on SCO hearquarters.

We will no longer need to worry about the lawsuit they filed against IBM.

Yeah, but GPL would be better

[argoff]

Contributions to BSD don't really help us as much because they can just be forked off into proprietary OS'es like Microsoft - which they will promptly use to put the reams to us with custom extensions. It would be much nicer if they went all GPL and nothing else.

I think the real problem is this attitude that free software is morally and intellectually equivalent to "owned" software. IMHO, this is an intellectual fraud, it screwed SCO, it will screw Sun, and it will screw us too until we finally get it.

Re:Yeah, but GPL would be better

[jslag]

Contributions to BSD don't really help us as much. . .

Speak for yourself - those of us who run BSD on our production servers find contributions useful.

If you pay a little attention to what the OpenBSD core team says and does, you'd realize that there is little-to-no danger that government funding will take the project in any directions but those stated in the project goals.

Re:Yeah, but GPL would be better

[jtdubs]

I love this kind of logic.

"The BSD license let's people do too many things, some of which I don't like. Therefore, the BSD license is TOO free."

"The GPL however, has just the right amount of freedom. It's still mostly free, without crossing the line of 'TOO free'. People can do what they want with it, as long as 'what they want' != 'what the FSF doesn't want'."

I have no moral problem with the GPL. I just wish people would stop calling it "free", unless they are going to put a (TM) or something after it. If you wanted your software to be truly free, you wouldn't be putting a copyright on it that contains words like "except" and "however."

Justin Dubs

Re:Yeah, but GPL would be better

[Uwe Barschell]

How is it that Apache and XFree86 have not been forked off into proprietary products and promptly used to put the reams to you with custom extensions? Apache is perhaps the most successful open source project, and XFree86 is perhaps second. This is in part because they do not use the GPL, and are therefore free from its restrictions.

Re:Yeah, but GPL would be better

[raytracer]

Contributions to BSD don't really help us as much because they can just be forked off into proprietary OS'es like Microsoft - which they will promptly use to put the reams to us with custom extensions. It would be much nicer if they went all GPL and nothing else.

This is just the wooly-headed mindless nonsense that seems so prevalent amongst the stylish Slashdot crowd. Perhaps you could explain BSD licensed software is less useful than GPL'ed software. If you don't like using BSD licensed software, don't. If you don't want to use BSD-derived software with lots of custom extensions, don't. But it's hard to imagine how software with fewer restrictions on its use and distribution could somehow be less useful than software with morerestrictions.

Re:Yeah, but GPL would be better

[Guido von Guido]

How did this get marked "insightful?" Both licenses have their places, and successful projects have used both. How has Microsoft forked off apache? Sheesh.

Re:Yeah, but GPL would be better

[Anonymous Coward]

Code is not free under the GPL. Free means no restrictions; the GPL imposes restrictions.

According to most people with a clue and dictionary.com, free means "Not controlled by obligation or the will of another" and "Not subject to external restraint."

Re:Yeah, but GPL would be better

[fferreres]

It's free for anyone that plays by the rules. You can use the software, but not abuse it.

Re:Yeah, but GPL would be better

[Arandir]

It's free for anyone that plays by the rules.

I just don't get this new liberalism. Why, when I was a kid, it was only free for anyone that obeyed the king's dictates.

Re:Yeah, but GPL would be better

[adri]

Right. Tell that to all of the BSDs.

The nicest thing that I find about working on/with BSD software is that you _can_ keep it proprietary - or could come to the realisation that if you release bits and pieces of it to the public you'll end up having them improved.

If everyone kept their BSD improvements to themselves then we wouldn't be advancing anywhere nearly as fast.

Case in point - there's a netgraph l2tp module. You can combine them with other netgraph modules to have a complete in-kernel l2tp->ppp->ip encaps and decaps path. Very fast. However, there's no userland implementation that exists. This leads me to think that someone _wrote_ an l2tpd, released the ng_l2tp module into the kernel and hoped that people would improve on it. The community benefits from having an l2tp module written, the company benefits from any improvements to it.

Its called "mutual benefit". I love it.

Re:Yeah, but GPL would be better

[Groganz]

True libertarians are fuckwits without 1/2 a clue, there is no such thing as total freedom, and there are rules in the BSD license too. You must include the license with a binary or source redistribution. Also, common copyright law means you can't slap a GPL license on it because you felt like it. The freest code is Public Domain.

Re:Yeah, but GPL would be better

[AME]

You must include the license with a binary or source redistribution.

Attaching a licence file to a distribution is not very onerous, especially where the succinct BSD licence is concerned.

Also, common copyright law means you can't slap a GPL license on it because you felt like it.

Who says? I see nothing in the BSD licence that keeps you from doing what you want with the software, even if what you want is to impose a more restrictive GPL license on your distribution.

Incidentally, I've always appreciated the wording of the MIT license, as it makes explicit that you may do as you wish, including sub-licensing. In legal terms, however, it is no different than BSD on this point.

The freest code is Public Domain.

Perhaps true. But unlike MIT, BSD, and the like, Public Domain affords no real protection from losing your shirt in court because you wrote some software. And, like I said, the retention of Copyrights is not very onerous. (Remember, by asserting my Copyright, I am saying that, "I retain the right to copy my stuff." The license makes explicit that the recipient has that same right, so there's no real loss of freedom there.)

Bluntly

[Tony-A]

A world with OpenBSD is much safer than a world OpenBSD.
This holds even more if you do not use OpenBSD.
(Like cars are much safer in a world with crash dummies;)

OpenBSD security on Debian

[PyTHON71]

I have the desire (but not the skill) to port as many security features from OpenBSD to Debian as possible without massive license violation. Anyone know if such a project is in the works? (And if not, why not?)

Daniel.

Re:OpenBSD security on Debian

[Anonymous Coward]

Search usenet archives - there has been an effort at some point, and since BSD is higher in the foodchain of licenses than GPL, you don't need to worry about that (BSD can become GPL, whereas GPL cannot become BSD). Just don't pull a MicroBSD copyright screw over (i.e. don't search and replace, actually append your changes, don't change other people's and you'll be fine).

The -point- for doing something like that, instead of simply improving OpenBSD with its own license, is completely beyond me. Does there need to be a GPL'd debian released OpenBSD? Answer that question first. I see absolutely no reason to give something that is already active and has an open source license, a simple copy, with a more restrictive license (GPL is more restrictive than BSD, MIT or PD licensing).

If you just want to do porting efforts, a lot has been done already - their ftpd has been ported, systrace (google for niels provos) is being ported to some linux platform [it's already on several others since OpenBSD], propolice is currently not integrated into other projects to the same level that it is in OpenBSD, but OpenWall Linux (www.openwall.com) has some similar protections, though not quite as full blown. I don't know about pf being ported anywhere, but it's a best-of-class product right now (only thing currently lacking is a non-kludgey [or at least well documented] way of doing stateful failover).

Port all you want - they strive to keep their code as FREE as possible.

Re:OpenBSD security on Debian

[RazzleDazzle]

If you are interested in OpenBSD's security, than use OpenBSD. OpenBSD didn't get as good as it is by copying someone elses code and features, they developed their code from the ground up with security (code correctness) in mind from the start. Of course they did not develop all of the code in the system, but they did create a lot of cool security practices (create really isn't it either, more like 'strictly adhered to'). With things like ProPolice, systrace, chrooting more, good clean code, PF (great program), suid reduction, strong documentation, etc, and last I checked, OpenBSD supported more platforms than Debian, why not try OpenBSD?

Thin Gruel Indeed

[divide overflow]

BSDForums writes "OpenBSD has a well-deserved reputation for fanatical security. Why is the U.S. military funding it? What do you get out of it? Cameron Laird and George Peter Staplin investigate and talk to Theo de Raadt, the creator, overseer, and taskmaster of the OpenBSD project!"

OpenBSD has a reputation for very good security. I wouldn't consider the quest for strong security "fanatical" any more than I would consider the quest for a bug-free operating system "fanatical."

Why is the U.S. military funding it? What do you get out of it?

The U.S. military is funding it because it makes sense to do so. Anyone who looks at OpenBSD's record for security and stability, the fact that it is free to use and modify in any way you desire, and doesn't consider it as a potentially cheap and useful platform for security applications...well, they aren't thinking clearly.

What do you get out of it?

I find it makes a great platform for firewalls and terminal servers, among other things. Ones that are reliable, very secure, with no software cost and lot of online support information.

Cameron Laird and George Peter Staplin investigate and talk to Theo de Raadt, the creator, overseer, and taskmaster of the OpenBSD project!"

They may have talked to Theo, but they sure didn't *quote* him much. The article was very thin on information. In my opinion it hardly merited a /. posting.

Re:Thin Gruel Indeed

[divide overflow]

This isn't a referendum on Theo, so take your gratuitous ad hominem attack somewhere else. There are plenty of other more appropriate discussions where you can rag on Theo.

OpenBSD In USA?

[debrain]

I seem to recall that OpenBSD was developed exclusively outside the USA because of export restrictions on crypto. Now it is being funded by DARPA? I am little confused on the matter, but thought that it was an interesting enough point to post.

Re:OpenBSD In USA?

[mslinux]

This is a damn good question. Mod parent up! Can anyone in the know provide an answer?

What I do with my OS

[vcbumg2]

I use FreeBSD for Internet Services OpenBSD VPN/Router/IDS NetBSD EOF Exotic Hardware running modern O/S e.g http://www.spectechnologies.net/projects/ehardware /index.html Solaris Database/CAD WS Win2000 Games/office Linux cluster/3d viz http://www.spectechnologies.net

I want to try OpenBSD but...

[sukotto]

I recently wiped Win98 off an old deskpro 4000 with the idea that I would make a firewall/router/whatever box (for fun and for my own education).

I'm a newcomer to bsd/linux and don't really know anything. I've heard that OpenBSD is really good and wanted to try it. Ok great, I'll go to OpenBSD.org, download, burn and install.

wtf? how to I download the iso?
After searching around I saw the entry in the faq that I have to buy it... or I can try to download it myself and figure out what bits go where on the CD. Oh, and the layout (or something) is copywrited so I can't grab it off the net someplace without breaking the law.

Well, I'm sorry guys. I don't know how to do that and the documentation I've seen doesn't tell me enough. I want to learn... but I'd like to do it incrementally, not all up front. So I'll be giving OpenBSD a pass and using something else.

Yeah, I'm sure I'll get at least one flaming response about how the team doesn't have to provide the iso, how they need money to continue development, how it's only $40 for God's sake.

I know all that already.

But I have a choice too. And I'm going to choose a distro I can try before I buy without having to figure out where stuff is supposed to go on the CD and whatever else I have to learn just to install the thing.

It's too bad. I've heard a lot of nice things about OBSD and I want to try it. But I'm going to go with someone else... and if I like FreeBSD, RedHat, or whatever, I'm going to end up sending that company money. And OBSD is going to miss out on that little bit of income.

If I'm not too firmly entrenched in the future, or if I actually learn enough to install OBSD myself then maybe, just maybe I'll give that distro a try.

Too bad :-(

Sukotto
(hmm... that's a little more whiny than I intended)

Re:I want to try OpenBSD but...

[Stormgren]

Wow, your reading comprehension sucks.

If you can't figure out how to download a floppy image and do a FTP install, or download the base dir for your arch and burn that to a CD and install, you shouldn't even try.

Hell, you shouldn't even try using anything other than windows, if you want your hand held.

Ooooh, no .isos, it must suck. Get your head out of your ass and reread the docs.

Re:I want to try OpenBSD but...

[Cirvam]

Stolen from a slashdot post I saw long ago:

If you want x86, then just download it from the OpenBSD ftp site.

wget -r ftp://ftp.openbsd.org/pub/OpenBSD/3.0/i386/ Makes it easy.

Once thats done...

cd ftp.openbsd.org/pub/OpenBSD

then...

mkisofs -v -r -l -L -T -J -V "OpenBSD-3.0" -A "OpenBSD v3.0-Release, Custom ISO, 17-03-2002." -b 3.0/i386/cdrom30.fs -c boot.catalog -o openbsd-i386-3.0.iso -x openbsd-i386-3.0.iso .

Burn that ISO!

Re:I want to try OpenBSD but...

[sukotto]

I gave up too soon. As several people pointed out careful re-reading of the documentation and the use of a d-link router allowed me to do an FTP install.

One reason I thought I couldn't do a FTP install before was that I'm on DSL with PPPoE. My solution was to hook my box into a dlink router that was already set up to handle the special connection.

So now I have it running on my box and am having fun setting it up as a gateway/firewall. If anyone else is trying the same I suggest this how-to as a good addition to the standard obsd faq: http://real.ath.cx/BSDinstall.html

Thanks all, for the info and kicks in the ass :-)

Sukotto

How about let Americans work on it, gov't?

[frankjr]

The government won't let us distribute our own crypto freely, but they fund foreigners to make cryptography, to distribute to the whole world?

Re:How about let Americans work on it, gov't?

[evilviper]

Call me crazy, but as I see it, OpenBSD is doing an incredibly small ammount of crypto. For the most part, they depend on OpenSSL for the crypto.

People on ship said WinNT not problem

[AHumbleOpinion]

Why not? They've tried it with Windows nt [gcn.com], which didn't work, so maybe there's more trust in open systems since then.

The news agency that originaly broke the story you cite later distanced themselves from it by calling it early speculation. My understanding is that a naive server app corrupted it's own database and naive client apps (the infamous "LAN consoles" that crashed) needed that database to function properly and to operate equipment. Rather than rely on the early speculation of *NIX advocates why not rely on someone who was on the ship and someone who wrote the software:

http://www.sciam.com/1998/1198issue/1198techbus2.h tml

"Others insist that NT was not the culprit. According to Lieutenant Commander Roderick Fraser, who was the chief engineer on board the ship at the time of the incident, the fault was with certain applications that were developed by CAE Electronics in Leesburg, Va. As Harvey McKelvey, former director of navy programs for CAE, admits, 'If you want to put a stick in anybody's eye, it should be in ours.' But McKelvey adds that the crash would not have happened if the navy had been using a production version of the CAE software, which he asserts has safeguards to prevent the type of failure that occurred."

I know you're trolling, but.....

[styrotech]

Way to go Theo. I hope you realize you're indirectly assisting the U.S. military in perpetuatating American hegemony around the globe while killing thousands of innocents. Oh, but you live in Canada, I guess you don't have to worry about that...

Way to go DARPA, I hope you realise that you are funding foreigners to indirectly assist Terrorists by making their systems harder to crack by US intelligence agencies.

Sound ridiculous? I hope so.

Or: Way to go Theo, I hope you realise that you are indirectly assisting civil rights and human rights groups by making their systems harder to crack by corrupt dictatorships.

Re:one thing

[Tony-A]

Openbsd is about qualtiy. It has les bugs, which equal less possible exploits, but security is not their objective. Hell, they only recently got a basic acl and added stack protection, stuff that has been available for *ages*

Oh, and theo's stubborn incorrect opinion that users don't need security models. This is wrong, as we need stuff like rsbac or grsecurity to bring *nix security up to a powerfull level.

With OpenBSD not implementing such a basic ideaology, They might suceed as a hobbiest OS, but never as a *secure* os.

Partially correct, but my impression is that if you want Multics, then use Multics.

Regarding OpenBSD and it security models or lack thereof. Theo's opinion matters. Yours does not. Mine does not. They are responsible to themselves for their own definition of what OpenBSD should be. ONLY. They happen to be nice enough to share the fruits of their labors, but that is their decision not our right.

as a hobbiest OS
Yep, but that's one hell of a hobby. It strikes me as what paranoid professionals use on their own private systems when they like to sleep peacefully at night.

Isn't it ironic

[Black Copter Control]

(hey.. Alanis is Canadian too).

If I remember correctly, OpenBSD development was based in Canada (in part) because encryption code was considered a munition and thus the US government refused to allow it's export (while it was allowed from Canada).

Now the military (who were probably the source of these rules) are paying for the continued development of a technology that the forced out of the country on security grounds.

Convoluted enough for you???

Re:Isn't it ironic

[Tony-A]

To add to the convolutions, it seems like the Department of State decides what is a munition.

Re:Isn't it ironic

[two_socks]

They've done it before. The weapon that went on to become the M240 was designed by a US company as a replacement for the M60. It got shot down. The design was sold to Fabrique Nationale. Following that, it was decided that it really is a great weapon. The US Army is now using both vehicle mounted and man-carried versions of what has become a belgian weapon.

Unless they're fighting a war

[Tailhook]

(Watching the wheels of military technology turn is like watching grass grow)

A couple of recent rapid developments serve to disprove that particular bit of common wisdom. The military, when pressed, kicks ass like no other organization in existence.

Re:Unless they're fighting a war

[nr]

Thanks for the link. It was a interesting read.

Re:one thing

[rjh]

Openbsd is about qualtiy. It has les bugs, which equal less possible exploits, but security is not their objective.

Spoken like someone whose knowledge of security comes from a Web page they read once.

Underline this one and write it in bold: high-quality software is secure software. Software insecurity arises when software either does something it's not supposed to do, or does something it's supposed to do in a way it's not supposed to do it.

When software works exactly as intended, even in the face of an adversary who is deliberately attempting to subvert it into malfunctioning, that software is secure, for practically any realistic definition of "secure". Software which works exactly as designed, which is fault-tolerant and has good failure modes in the event of intolerable faults, is a cracker's worst nightmare.

Re:Who's teh dumbass now?!!

[slashLART]

If you don't have a hdd, how do you expect to -download- and burn an ISO to begin with? Oh, with your knoppix CD? How did you burn that to begin with? Your bootloading methodology is flawed, as is your argument. If instead you mean, is there some OpenBSD install that doesn't touch the hdd - well, you can use bsd.rd [rd being RAMDISK] or look at related projects such as www.opensoekris.org which are designed more for that application, the thread was about _installation_ and for that a file in a .iso format is not necessary. And for those who are used to burning boot disks with tools like Nero or mkisofs, then the cdrom32.fs file provided by OpenBSD is more than enough to let folks burn an install CD.

That's nice...

[ebbomega]

I never said he was...

Hacker Soldiers

[two_socks]

Why is the U.S. military funding it? What do you get out of it?

Because they want the most secure operating system available. I may get my ass shot at a lot less. Or, maybe, terrorist hackers won't be able to figure out when my flight home is leaving Kuwait City International Airport.

I'm in the Army National Guard. It used to be my full time job. Now I'm a "weekend warrior".

I used to administer NT boxes for the Army among other job duties. It gave me the heebie-jeebies! I am a helluva lot more comfortable with military secrets residing somewhere else.

Before someone trots out the "you're just a weekend warrior" pony - after I left the guard full time, I was deployed to Kuwait for six months of middle-east summertime bliss. I was there for September 11. And, yes, I really did fly home out of KCIA, and I was damned glad the time we flew out was kept secret, even from us. And if the only computer that info ever lives on is an OpenBSD box, I'll sleep better at night. And so will my wife, parents, etc.

When Hackers meet Soldiers...

[No. 24601]

geeks once again get their asses kicked by the jocks.

Re:about me ...

[jwdeff]

First off DO NOT go for a CS major, unless you don't want a job.

Most of the "*BSD is Dying" posts are cut and paste troll comments. Check out bsd.slashdot.org and set the threshold to -1.

I don't think too many people really think, *BSD is dying. Just flamers flaming.

You are confusing privilege with access modes

[anonymous cupboard]

VMS has a lot of privileges, the user or an image may be assigned particular privileges which they can then subset when they don't need it.

Access modes is how the system is structured. There are two main modes, each with two sub modes. In the system space, which is common to all processes, there is kernel mode where the OS runs and exec mode where RMS (Record Management System) and databases run. They use the common nature of exec mode for global buffer management between processes. In per-process space we have executive mode and user mode. Exec mode is where the shell runs, and use mode is where most normal programs run.

Normal users do not write stuff that runs at elevated access modes. They require privileges to enter elevated access modes which they normally do not have. However, it is possible to enter an elevated mode through a declared entry point (think call gates in the x86). Arguments to these calls are checked for address violations in the space from which the system service was called. For example if you want to read something into a buffer, and you call from user mode then the buffer must be user mode writeable. Areas of system space are only accessible to a user if the system sets the protection accordingly.

The main benefit is that it is extremely difficult to get out of user mode except through defined entry points. However, if you particularly want to do things inside exec mode or the kernel, you can extend the system API with your own loadable service routines or you can enter the system with a change-mode-to kernel API call, where you stay in your program but now run with full access to the kerenl address space until you return. Naturally such a call is protected with its own privilege (CMKRNL). To write some thing that runs in exec or kernel mode requires more skills because although the call list will have been address checked, the references will not.

This means VMS is tight and with excellent availability. The various checks as you cross address spaces means that on any given system, Unix will always run faster, but with lower availability.

Theo's Hardware

[theolein]

In the article there is a link to Theo's personal site. He lists his hardware there, and the amazing thing is that he doesn't have a single machine capable of more than 200MHz.

I find it amazing in these days of 3.6GHz machines needed to run bleeding edge games and gimmicky OS's and everyone and their mothers going gooey over the latest GHz jump in analy embedded mobile devices that OpenBSD's chief developer uses computers that actually fit his needs. It is comforting to know that the SECURE processing and dissemination of digital information can be done efficiently without the large, bright, rounded, colourful buttons and Windows found in most other OS's.

Re:Theo's Hardware

[evilviper]

I believe that would be due to the page being out-of-date.

Although, I've said it before, and I'll say it again... If it wasn't for video (playback and encoding), I would be using a 100MHz system right now.

Re:OpenBSD isn't the only one ... RE: fs-crypto

[jstockdale]

There are at least half a dozen filesystem encryption programs that function on OpenBSD

Well, yes and no.

CFS - Weak (DES) encryption

TCFS - Slow (3DES) encryption

cryptfs - Blowfish (good) encryption, but the system relies on mounting loopback / stacked devices, which although being the best option available, is still slower than crypto integrated straight into the disk structure itself.

In fact all the fs-crypto mechanisms that I know of that work on OpenBSD either are slow due to the loopback method of mounting the drive, are based on weak encryption, or are just made as proof-of-concept rather than DOD-standard encryption implementations.