Slashdot Mirror
OpenBSD Clashes with Adaptec In Quest for Docs
TrumpetPower! writes "OpenBSD developers have been asking for documentation from Adaptec for over four months. Adaptec's response has been to deliberately misunderstand what is being asked of them. A former Adaptec employee admits that the hardware is buggy and tricky to get right. So, as a result, OpenBSD 3.7 will ship without Adaptec RAID support. Personally, I'm glad that Theo isn't resting on his laurels."
I connected to adaptec's mail server, and it told me it would accept mail for the account in question. I guess they had a config problem, and these guys got snippy when they got a 500 error...
It would be nice if more of the Linux big names would jump on the bangwagon and lobby with companies to get open source drivers for hardware.
In soviet russia the old OpenBSD kills the dead Adaptec
There's an old saying, which I think fits well here.
"Never ascribe to malice that which is adequately explained by incompetence." - Napolean
When the hardware vendors will release all the specifications of their hardware to the OpenSource teams? It's so difficult to do so?
"I'll not release my documentation because others business can get all of my secrets and my bugged harware."
http://www.michel.eti.br
You've just gotta admire the guy.
I know that when I'm buying hardware, I first make sure that there's at least a reasonable chance that it will work in my operating system (Linux, by choice). So, in this case, if I was choosing a RAID card, and my system was BSD-based, then Adaptec would be down a few quid.
Absolutely. Open source drivers would be a beautiful gift, in this case it's actually more than what is being asked for. Adaptec is asked to release specs on their raid controllers, they chose not to.
They are under an obligation to provide usefulness on legit architectures, but they aren't doing that. Adaptec should get over their shame of bugs, and allow the driver people at OpenBSD a chance at making things work.
There is no general fix for this problem, often specs are released way too late. On the other hand, releasing open source drivers will open specs for the same device. These specs aren't just trade secrets, they're actually necessary for building drivers.
"And we have seen and do testify that the Father sent the Son to be the Savior of the World"
1 John 4:14
There's a very simple solution for this: Don't buy anything from Adaptec, ever. They'll be out of business; problem solved.
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
read the fucking discussion. they're not bitching about driver support but about lack of _documentation_
Just a note; the "former Adaptec employee" is Scott Long of the FreeBSD project.
I have not been using OpenBSD sice 1999, but hardware support was never its strong point... though what it supported was,like all the BSD's, supported extremely well.
It's a good call, in spirit of BSD. Scott's drivers are exellent and they just need to port those.
That's not the point - if it was easy to get specs for hardware, then /all/ operating systems would benefit - not just the well-used ones.
OpenBSD confirms it... Adaptec is dying!
Try to hack my 31337 firewall!
Your opinion is both wrong, stupid, and has no real world bearing. OpenBSD isn't a depreciated version of FreeBSD; both projects have a completely different focus, and depending on your needs one may be more suitable than the other.
-If God wanted people to be better than me, he would have made them that way.
Which make you think: "Why is OpenBSD doing this and not FreeBSD". I think it's sad that the FreeBSD developers don't seem to care that much about having free drivers (the AAC is free I believe, but the management interface is not).
Of cause what really annoys me is that the Linux developers seem to care even less. Why is it that the developers of free software can't stand togther and demand documentation? And why is it that it's the smallest team that must make these demands?
It did not start with BSD4.4-lite, go to 386BSD, move to NetBSD, then OpenBSD, then DragonFlyBSD and then FreeBSD. Each are their own system which split at one time or another from the same tree.
All four of those systems are maintained today and therefore it is not like Windows 9x complaining about hardware support. Windows does not maintain new versions of Windows 95.
OpenBSD is the extremely secure and extremely open of the BSDs and Unix-likes. OpenBSD refuses to have anything that isn't as Free and Open as their goals describe into their system. Linux and FreeBSD are more into the functionality over ideals idea. NetBSD I cannot speak for though as I don't really follow them.
I'm sick of following my dreams - I'm just going to ask them where they're going and hook up with them later.
FreeBSD is more common as a desktop OS and webserver, but OpenBSD is more common as a firewall. And it often goes unnoticed by people because it just sits there firewalling things. Remember that companies rarely announce the details of their security arrangements. Needless to say, these firewalls are mission critical and need RAID.
Theo is a belligerent prick so he gets noticed more than the others, but every open source OS has identical problems with driver support. Why do you think Theo got that award when he and Stallman don't exactly see eye-to-eye?
I rarely criticize things I don't care about.
it seems as though their website is IE only, as i couldn't send them feedback using opera.
It allows so many companies to sell you a leash and handcuffs. Yes go ahead and wear them, the great advantage is that you'll never go where you shouldn't and that you'll never hurt anyone.
The amazing thing about this whole afair is that Adaptec itself is also a leashed and cuffed company. But after some thinking I realized Nvidia is just such a company. Even if they wanted to release the _specifications_ of their hardware they couldn't.
All in all this forces people to stick to one OS. That's why it is so important people step up for free specifications of their hardware. Because without them you are bound to be tied to a monopolist.
Theo didn't get that FSF award for nothing.
did someone say theo?
It's all about making sure the big shareholders know that the company's policies are costing them sales.
People say that Theo should stop being so annoying, but the only way shareholders find out is when it gets massively publicised like this.
It worked for the 802.11 drivers. It's worth a shot here.
I rarely criticize things I don't care about.
Like their old AAA ide raid controllers which was nothing more than IDE paddle boards with software raid logic..marketed as true hardware raid.. Documentation exposes the magic behind the illusion..(sometimes)
Reminds me of Promise's definition of "Linux support" for a card I bought.
In the case of the SX-150 SATA raid card (which has a hardware XOR engine and whatnot), that meant "we have binary drivers for distributions which are several years old".
There is some source. Well, it's a 'wrapped' binary driver, and it's only available from "some guy" in Germany who begged Promise support long enough they gave it to him. You a)cannot compile it into the kernel b)cannot compile it for 2.6 because it simply isn't compatible. I sent numerous emails to Promise asking when a 2.6 driver would be available or if there was any updated source code. None were ever answered.
Same story with the tools- unless you're running Redhat 9.0 or some ancient version of Suse, forget ANY on-line monitoring.
Not that the customers are much better- one page I found about the card suggested that "software raid is faster anyway", which is an absurd proposition by itself. Regardless, why would you spend $100-200 more on a hardware-raid card complete with cache memory, and then just use the 2.6 SATA driver which only drives the SATA interfaces?
From what I understand, 3ware has better support for Linux, but that means I have to migrate a large amount of data off the old array..
Please help metamoderate.
Just use IBM serveRAID controllers...
Linux fought the battle for obscure cards in the 1990s. They lead the fight regarding video cards (there is even less video oriented stuff on the BSDs). They also conduct the fight for things like video chat software which doesn't interest the BSD community as much.
All free OSes combined don't really add up to that much market share for many of these hardware devices. The fact is we don't have enough pull to demand most of the time (RAID controllers might be an exception).
Because BSD is not Linux and it is definently not GPL. They have very different views on free software. While Linux and GPL support everything to be opensourced (even drivers) BSD people have a different approach - everything they develop is completely free software (unlike semi-free GPL) and if a company wants to put proprietary drivers in it, then so be it.
The whole MIT/BSD culture can be described with one sentance;
Being able to read other people's source code is a nice thing, not a fundamental freedom.
It seems to me that this will either result in people not upgrading or Adaptec will just release a binary driver (if one doesn't already exist) and people will upgrade if they want to. So this makes for a nice press release but I think that the implications aren't as great as they sound.
When it gets to the point when your developers start flaming hardware developers in public, don't expect much generosity from those vendors in the future. Summary of articles: developer sends a couple of requests to adaptec nothing much happens flaming developer types up a rant with keyboard courage hardware vendor shocked at rudeness of rant. replies saying that firmware needs work. developer does the blockquote rant, and flails arms in the air: Wah! Wah! Wah! Adapted hates us. Summary of real life: Adapted doesn't hate OpenBSD. It takes work to provide a huge technical manual of all your products. And, with firmware changing, the specs for how to interface with that product will change. The articles paint Adaptec as some sort of bad guy, when that isn't the case.
It might be easy to shoot down their arguments, but that doesn't make it worthwhile...
Simple, don't buy from Adaptec until they release information about their hardware or documentation. Hit them where it hurts financially, and they'll listen. Or just spite them by reverse-engineering an existing driver? :)
The fact is we don't have enough pull to demand most of the time
Very true, which is why it really doesn't help that much that OpenBSD take this fight alone. Of cause they did get an amazing result from their campaign for free drivers for wireless networking chips.
Theo says: "We are not asking for support. We are asking for documentation."
Substitute "They" for "We" in that sentence and it could have been me speaking, when I was working at Adaptec and trying to release an in-house version of the starfire (a.k.a. "Duralan" ethernet MAC) driver. I hit that same brick wall over and over again while tying to get some chip specs and a linux driver released. Somehow, in their minds, "support" is translated into not releasing specs and drivers. Releasing such information, in contrast, is a failure to support customers. This wierd Orwellian doublethink seems to pervade the thinking of everyone connected with supporting Linux and other free OS's at Adaptec.
It's so amazing to see that nothing has changed at Adaptec in the last 7 years. My own driver episode was "resolved" (unsatisfactorily, for me) by Donald Becker agreeing to sign an NDA for the chip specs. Not to second guess Donald, but my thinking at the time was, "this just postpones the problem. Maybe it would be better just to boycott these imbeciles."
Not to close on a sour note, I should say that Adaptec was a great place to work in many ways, and I always viewed their attitude toward free software as an aberration. I still tend to do so, and perhaps that's wishful thinking on my part.
Have an old Ultra 1 doing firewall and light server duty for a DSL line. So far its had zero hardware issues and everything has worked. Wish I could have said the same for NetBSD. It locked up randomly on the same box.
I haven't used OpenBSD in a few years and was really impressed with their rewrite of packet filter. You linux folks should check it out.
Only the State obtains its revenue by coercion. - Murray Rothbard
But he's belligerent in the right direction. He does more than Stalmman does on the Open Source front, he calls a spade a spade, and is one hell of a coder.
This guy is way out there
And why is it that it's the smallest team that must make these demands?
My guess would be because of the security auditing that the Open BSD team does. They go over every line of code. You can't guarantee the safety of something if you haven't personally reviewed the source code.
Linux and FreeBSD's foci aren't on safety. They're on operability. If the drivers work(even if imperfectly), those two sets of developers are happy.
LK
"Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
Many, actually. And if it gets supported on obsd, chances are it will be ported to free/net.
So, it's in everyone's best interests.
Why adaptec isnt releasing detailed specs is obvious. If people had them they could better evaluate the product. Apparently the marketing dept. at adaptec fears transparency and complacency.
Look at the small and medium end raid market now. Theres not many players, Adaptec,promise,3ware and a bunch that adaptec bought up. Adaptec gains nothing by opening up itself to a point by point comparison with lesser competitors. Their name recognition is carrying them much the way IBM's used to. Further if the hardware is bugged and tricky and adaptec knew about it then they open themselves up to liability.
Their reasons are obvious keep the barriers high and keept those that can't climb them out.
whatever
Big Wolf on Campus. Aimee Castle in Sandman. Sleep. Mind Control Sleep Dust. Aimee Castle is Sandman's slave. Sleep state.
... but they don't help to do open drivers.
Are they scare of something? they don't lose anything helping to build open drivers, they win.
ajf
And you are enough of an asshole to post a comment like that. Search some past activism from Theo/OpenBSD and you might come across something you seem to ignore.
"But he's belligerent in the right direction."
In this case it's a good thing. Sometimes it's not.
I rarely criticize things I don't care about.
Of cause what really annoys me is that the Linux developers seem to care even less. Why is it that the developers of free software can't stand togther and demand documentation? And why is it that it's the smallest team that must make these demands?
One is to whine and complain, the other it to move forward with what you do have, making your OS so big it is something you simply can't ignore. In short, talk is cheap. Show them a market share and lost revenue they aren't getting (and no, something in the zero point something percent doesn't count) and they will come.
Kjella
Live today, because you never know what tomorrow brings
You get BSD and drivers and you don't have to worry about this stuff.
:-)
Plus it runs Office !!!
Just buy a Mac
Unfortunately I cannot purchase Adaptec controllers anymore. No, it's not because they aren't supported in OpenBSD, nor is it a new decision. It is because a couple years ago I purchased several Adaptec raid controllers for some webservers and the drivers included didn't work. I managed to obtain, after much pain, a better driver. To make a long story short, they had to come out of service because the driver updates took so long that I had to run really old kernels just to support the raid driver. Sounds like they haven't changed. Too bad, I used to buy a fair number of raid controllers from them.
You know, the LSI SCSI cards are rather nice, they work with Linux; I don't know what their deal is with docs, but they seem to have contributed code.
(OK, so not directly related to Adaptec - but it seems to be a reasonable place to give their competitor a pat on the back!).
Being able to do whatever we want with source code is a good thing. Thats how I see the BSD license.
The GPL is about making money. Look at mysql, redhat, novell, or even apple. They profit from GPL code, and only follow the GPL when it suits them. Mysql documentation for connector j for example is NOT GPL licensed!
I'm not trying to say that you can't make money on BSD licensed software, i'm just saying that its possible for EVERYONE to make money on BSD license software which means its more open!
Then why does Linux have piles of binary-only drivers, while OpenBSD would never allow that kind of thing?
Parent is not Insightful, it is a Troll.
Interestingly, Theo of OpenBSD is more insistant that the entire base system be open source than the most popular Linuxes. GPL software is tolerable if there is no alternative, but commercial software is not tolerated.
The only exception is firmware binary blobs (which all OSes need, as it is not practical to create open source replacements), they are tolerated if they are released under a license that allows OpenBSD to distribute them.
That's similar to what Linuxes like Debian demand, and that's a lot more than Linuxes like Red Hat and Suse demand.
I rarely criticize things I don't care about.
Signature.
I wish the Linux people would have enough balls to make a stand with us. No such luck there.
Oh well
you use..... windows ?
* To: Charles Swiger
...deliberately breaking OpenBSD's support for Adaptec hardware as some
* Subject: Re: Adaptec AAC raid support
* From: Bob Beck
* Date: Sat, 19 Mar 2005 13:56:41 -0700
* Cc: Theo de Raadt , Sean Hafeez , misc@openbsd.org, Scott Long , freebsd-questions@freebsd.org
* In-reply-to:
* Mail-followup-to: Charles Swiger , Theo de Raadt , Sean Hafeez , misc@openbsd.org, Scott Long , freebsd-questions@freebsd.org
* References:
* User-agent: Mutt/1.5.6i
>
> sort of ultimatum is a childish and self-destructive action. I hope
> the other OpenBSD committers veto any such action as being
> counterproductive and harmful to your users.
Horsecookies. What was done was remove AAC support from GENERIC,
because users know what is in GENERIC is supposed to be stable and a
good candidate for use. I've got AAC's. They aren't at the moment.
they die, and you can't do anything with the raid management without
rebooting, and Adaptec has shown no signs of releasing documentation
so that situation can be corrected.
Sure, there's a "free" driver, and a non-free management interface,
so it's only half a driver. Pretending to have a production system
using a raid card that with no supportable management interface so you
have to reboot to fix anything is like buying birth control pills in
packs of 20. Pretty soon you're going to take a good fucking on a day
you really can't afford it. Period.
As such AAC isnt' any more broken than it ever was. OpenBSD
just chooses not to encourage users to purchase a non-supportable
card by including support for it in the GENERIC kernel. Are you
saying it's more honest to leave unstable and incomplete support in
there? People who wish to use it anyway can always compile it in.
> Otherwise, you're likely to discover that most people choose to run an
> OS which works with the hardware they have, rather than sticking with
> OpenBSD.
Or choose to replace the hardware that isn't supportable by the
OS they want to run. Thank you LSI and Dell. LSI cards seem to work
fine.
-Bob
emphasis added by poster
Somewhere in the void of Internet, a message is quietly being sent to /dev/null, for that it did not pass someone's procmail "crook" filter.
I once had a signature.
This post isn't flamebait, it's either informative or redundant.
The guy is a ginormous prick. I would never even think of using openbsd because of him, as I don't want to be part of building any sort of support for a project with him involved.
Can you imagine how incredibly stupid things would be if, say, Linus was as insane as he is? Surely the code would be forked by rational people before openbsd ever got that big, but still I want nothing to do with it or him.
It's like the guy didn't even read Theo's email, just chopped it up and repeated portions of previous emails in response.
I really don't get this... it's win/win for adaptec unless they have something huge to hide. OpenBSD has been around for a while and has a good reputation for getting stuff done. If they would just forward Theo and his buddies to some guy on the back end of things, they'd generate some sales and also make the OpenBSD people shut up for a while.
Wierdest thing is you KNOW the guy they're emailing speaks excellent English because of the typos he makes. He really must be deliberately misunderstanding the request.
What the Adaptec guy actually said was:Using the word "buggy" like it was used in the Slashdot front page article implies that the cards are flaky and that non-Adaptec cards aren't (as) buggy. This isn't outright stated, but similarly saying something like "I don't use Microsoft Office because it is buggy" tends to leave the listener with the impression that other office suites are less buggy, even though that isn't stated outright.
The Adaptec employee stated only what we already know--that different revisions of firmware have different bugs (in ALL products that use firmware, not just Adaptec RAID adapters), and that they must be worked around. If different revisions of firmware didn't have bugs, then different revisions of firmware wouldn't exist--the first one would have worked fine (aside from occasional feature additions and tweaks).
However, to the original poster's credit, Adaptec RAID cards really do suck, and they really are buggy (not to mention slow, especially in RAID 5, compared to almost every other brand--and Adaptec's entire SCSI line is pretty consistant in that regard), but that is beside the point. Slashdot shouldn't participate in the same FUD that we so often criticize--just let the facts speak for themselves, and leave the interpretation up to the reader.
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
Plus, it is likely a more expensive one for Adaptec (vs. binaries) over the long-haul as they need to make sure that whatever they release for the public's eyes are relatively "clean".
The end result of this could very well be Adaptec throwing something over the fence that is only guaranteed to work with a certain generation of cards with only a certain generation of firmware. Is that really better for their customers than a set of binary drivers?
They go over every line of code.
Of course they do, unlike most other open source projects which just check in whole lumps of code without caring about what it does... Sorry which bit of this makes your comment relevant to the discussion?
Try NetBSD... safe,straightforward,useful.
I once found the Linux Adaptec driver maintainer's site. I was appalled to see a matrix of mostly unique drivers for combinations of Linxu kernels and Adaptec HBA models. I got the impression that every time Adaptec designs a new HBA they ignore everything they've done before.
Contrast that with Symbios SCSI HBAs (now LSI Logic). There has always been just one driver for all the Symbios HBAs that use a given SCSI chip, and more recently I think I've seen some integration into a driver that covers multiple 53Cxxx chip types.
I use Symbios / LSI Logix SCSI HBAs in RS/6000 AIX systems, Linux boxes and Win PCs, always without trouble, without driver nonsense, and without configuration or script bullshit. I wouldn't use an Adaptec HBA even if I got it as a gift.
Look at the bright side: there's always seppuku.
All free OSes combined don't really add up to that much market share for many of these hardware devices.
Don't they, really? What's more likely to need a multi-hundred dollar Adaptec RAID adaptor?
a) Grandma's Desktop system (i.e. 90% of computers)
b) General business-oriented Windows Desktop system (i.e. What every PHB has)
c) A Linux desktop
d) A Linux/BSD server stuck away in a closet, responsible for databases, mail, user files, web serving, and whatever else they can think to make it do.
b) A Windows based server stuck away in a closet doing the same as above.
Maybe it's just me, but I think that OSS Servers are a huge consumer of this type of product. Every business class RAID / Multiport Ethernet card that has open documentation pretty much has a driver. Sometimes they manage to hack together a driver that works just as well as the original, even though there is no documentation. I'd say that with IBM's help, the market for Windows/Linux servers is more or less becoming about the same.
Then there's the odd-power user that uses server class stuff, just because they need it to do whatever they're doing... I don't think this makes up that large a percentage of the market, so probably best to simply ignore them for the sake of this argument.
Why anyone would plain ignore such a market is intriguing to me. Here we have some jerks that want to write drives to your product for free, all they want it some documents. They have NOTHING to gain, and YOU have a chance to get your elbow in a big market for basically nothing. If they cooperated, and therefore made their product the best supported amongst their competitors, then they suddenly have that part of the market bottled up. Anything else is financially irresponsible to your shareholders!
... not to unnecessarily push customers to choose a competitor's products, when keeping those customers costs nothing more than an e-mailed PDF or two.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
We primarily use FreeBSD (and a few linux machines), and we have yet to find a RAID controller that makes us trust it, except for 3ware. IDE performance just isn't there yet, so we need SCSI.
The raid monitoring utility raidutil from adaptec would crap out after a while, leaving us with unmonitorable RAID units. The monitoring app for a well known FreeBSD only vendor doesn't work under unix, only windows.
The dell raids (rebranded adaptec I think) will lock up under heavy load.
Are there *any* decent mid-range SCSI RAID cards that:
a) allow for some sort of remote monitoring, can be via the OS/snmp/whatever
b) Work under FreeBSD and Linux
c) Don't die under heavy sustained load
d) Support hot swap
It's getting very frustrating.
The 500 error is new, its there because they don't want their customers complaining. Originally the email address worked, and for 4 months they got the run around and nobody would give them a "yes" or "no" about documentation, that's what he is "snippy" about.
From this piece (linked to from the post):
"...and though we would like to support "all" of the various flavors of these new operating systems..." (emphasis mine)
The first release of the Linux kernel was in, what, 1992? OpenBSD has been around for how long? NetBSD has been around for how long? Debian has been around for how long? Red Hat has been around for how long?
Windows XP (released 2002?) is "newer" than any of the "major" Linux-based or BSD-based operating systems. The mental midget who wrote this piece referred to "[OpenBSD] as well as many other flavors of Linux/Unix" as "new operating systems" (his words), implying that Unix itself is a "new" thing. Reminder: Unix has been around since 1969; Win32 systems have been around since... what, 1993?
With such a PHB-style, Windows-centric philosophy, is it any wonder these guys are basically dodging all requests for help or support?
With spending like this, exactly what are "conservatives" conserving?
I think it is quite common for vendors to put out hardware that isn't quite engineered up to spec, where, if you looked at its design and interface, you'd see that it really can't work quite right, or get the performance they claim. Another reason is that the documentation may simply not exist, a clear indication of poor engineering practices at the hardware vendor. I suspect that's actually the main reason so many hardware vendors are so secretive about their interfaces: they don't want to air their dirty laundry in public.
As a rule of thumb, if you are buying a piece of hardware, buy one for which known, good, independently-developed open source drivers exist. The existence of such drivers is a good indication that the hardware is well-documented, probably decently designed, and that it probably does what it is advertised as doing. And that's a good rule of thumb even if you are buying the hardware that you only intend to use with closed-source operating systems.
See, you can't turn what I write into binary only crap. Because what I write isn't crap, and you distributing a binary of my code doesn't make my code dissapear. Think, then speak.
At some point, Adaptec was putting out several incompatible versions of their SCSI controllers (I believe the 2940). That wouldn't have been so bad if their own drivers didn't get confused about that: if you used the wrong driver/card combination on Windows, you wouldn't get an error, things would just mysteriously fail.
That was just one of many headaches I have had with Adaptec cards. I just ended up avoiding Adaptec altogether whenever I could. Adapters from other companies turned out to be cheaper and easier to use/install. Several other companies also had better FOSS support.
We have a driver already. We want specs to write raid management and monitoring software. They already have a binary only piece of crap that does this. We do not want it.
What this will result in is either adaptec releasing their docs, or adaptec losing thousands of raid and scsi controllers worth of business. Their call.
OK, for those of you using hardware RAID on OpenBSD, which vendors do you think rock?
Help save the critically endangered Blue Iguana
Try getting specs for the via unichrome chipset....looks like there are at least a few companies that do little to help the open source community develop drivers for their hardware....maybe they don't want to sell their products as much as their shareholders would like them to.
This kind of nonsense only got them 8 or so wireless chip specs. Clearly it doesn't work and just makes them look bad. That's why Theo got the FSF award.
Of course they do, unlike most other open source projects which just check in whole lumps of code without caring about what it does... Sorry which bit of this makes your comment relevant to the discussion?
Binary drivers.
You can't review the source code to binary only drivers. Other open source OS projects don't pay the kind of attention to detail that OpenBSD does.I'm not accusing them of not caring at all, but they don't spend as much energy on it as the OpenBSD team.
Is that clearer?
LK
"Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
OpenBSD email - only if you interested --> www.loftmail.com
It has become a common modus operandi for companies to force-feed binary drivers to open source projects. The Linux kernel is a shameful example of this. FreeBSD followed after.
Linus says it's ok, smug user's don't care, 'cause their nvidia cards works, and developers are cool with Java, because it "runs everywhere", except where it doesn't.
OpenBSD remains true to its origins: quality and freedom.
Main difference between the BSD license and the GPL license: one is from California and the other is from Massachusetts
Sorry, but it appears that the poster totally missed the point. Theo appears to want documentation. They didn't get it. They got palmed by Adaptec, who tried to appeared to state that OpenBSD is just another flavour of Linux (thus missing the point - I guess that Doug Richardson isn't terribly technical and probably better at marketing - doubtful considering his email reply - than understanding the BSD marketplace) and also, as Theo stated, want support.
Exactly how hard is it to provide decent documentation so that developers can create drivers? I guess in this case it's quite difficult.
Oh, and another bugbear. I used to work for a certain printer company. Whenever someone mentioned even documentation for Linux (*gasp*!) they'd immediately think that the person wanted support. As it turned out the head company eventually got a clue, but it took a long time for the organisation in Australia to even provide a link for documentation. There was a most definite anti-Unix bent. When OS X came out certain colleagues nearly had a heart attack. I once heard one of them say that "historically Unix hardware support sucks". OK, I digress, but this rant felt good.
XML is like violence. If it doesn't solve the problem, use more.
It never happened! It was just a figment of Theo's imagination!
Did you ever consider that this was done temporarily to stop a flow of negative email to this particular managers email account - and then the account was reenabled? I guess not.
XML is like violence. If it doesn't solve the problem, use more.
I really don't... The adaptec configuration utility CD runs linux. Surely it's not hard to release the docs to something they've already created...
"We know what happens to people who stay in the middle of the road. They get run over." - Aneurin Bevan
give up lent for lent, you'll be much happier.
I feel somehow second class, my Napoleans have never had pithy quotes, they do taste good however.
Try writing a letter, not an email - you know physical paper w/ stamp. Nobody in whatever call centre offering support would have any notion of what you're asking for - and to even try just makes you out to be a boob.
Write a letter, 9 times out of 10 you'll get what you want or a good explanation why you won't get it.
Over the years I've had great sucess with this when searching for info - in fact some manufacteurs have send complelete reference books on their stuff shipped in the mail for free for for asking. This also applies when you have a problem with a company that's hard to explain - calling in to explain a technical problem (situation) rarely helps.
Well, I have a number of Adaptec's ATA Raid Cards (ATA RAID 2400A), for the longest time they only supported RedHat 7.0. Now that Fedora is somewhat the premiere platform for me (three releases later), they are finally supporting Redhat 9.0.
With the the latest Fedora, there is no way to see if the raid array has a failed drive. So I instead use the card as a quad ATA controller, using software RAID. Guess if I'd buy another Adaptec piece of hardware???
Love many, trust a few, do harm to none.
The idiot-in-question posted anonymously (IP be-damned) with that claim.
It's simply a BS excuse from Adaptec for taking down Richardson's known point-of-contact.
It also had the added benefite of portraying OpenBSD and Theo supporters as The Bad Guys(tm).
As always the suckers bought it, as suckers always will.
Who said anything about vendors supplying drivers? They don't want drivers! They want documentation!
XML is like violence. If it doesn't solve the problem, use more.
They obviously were just looking for a scapegoat to write a new release song about. What would an OpenBSD release be without an homage to de Raadt's latest exercise in bridge-burning.
Please provide examples of binary drivers in the Linux kernel that are officially endorsed by Linus Torvalds. Thanks.
XML is like violence. If it doesn't solve the problem, use more.
theo de raadt
:)
d. j. bernstein
others?
releasing specs does *NOT* give a competitor an advantage /cheaper product, which would work with the same driver so it can be easily considered as an alternative? Customers would love it.
Let's see, if a vendor A has a big market share and a vendor B would like to have part of it, would not it be interesting to create a better
I understand something like that has happened not so long ago to Cisco as a vendor A and Huawei from China as a vendor B.
I do not know if it is easy to create a RAID controller to work with an existing driver... but you see where I am trying to get?
...a stunned silence fell upon the hall.
Doung's mail account is now filtered. But we, on misc@openbsd.org, just received this:
From: Theo de Raadt To: misc, freebsd-question
Since the original Adaptec guy Doug has blocked his mail, here is the
email address of the next person at Adaptec who is involved in this.
He has also previously indicated that he would be involved in any
decision to provide documentation on the aac RAID management
interface.
Marty Turner
marty_turner@adaptec.com
Product Manager
Adaptec, Inc.
(919) 287-2045
Sorry Marty, but you are only getting comments from your customers.
Would you get more of a chance of getting documentation from IBM then? Maybe Theo should pester them instead of Adaptec...
XML is like violence. If it doesn't solve the problem, use more.
The sooner SCSI support is ripped from BSD & Linux, the sooner it'll be ready for the desktop.
It was made very clear already that there are some very big customers who will stop buying adaptec because of this. 1800 RAID controllers counted so far just in OpenBSD users, nobody is even bothering to count the SCSI controllers because its so many.
I've been reading the posts on the misc mail list for OpenBSD, and I think a few things need to be said.
First, Theo and the other developers, although making good points, are being quite rude to employees. I think that its important for them to push this issue, but I think they are handling it immaturaly. Flaming Adaptec (ex-)employees is not a good move, even if Scott did make a post on OSNews -- attack the companies economic base through a boycott instead.
Secondly, I think that if Theo and the gang started an organized boycott of Adaptec raid controllers in a professional manner, then got those people to sign a petition, write to Adaptec, and such along with getting a pretty accurate count of how many of Adaptec's raid controllers have been purchased by those boycotting Adaptec, they might be able to show themselves to be consisting of a large enough market to cause a dent in Adaptec's profits. Not only OpenBSD, but also FreeBSD, NetBSD, and Linux users who feel that its important to use open source drivers. This may require a bit of work, but its the most effective way to get Adaptec's attention. I mean, how many open-source Unix servers are using their raid cards? How many of those users, admins, etc. realize the importance of an open source driver so it can be maintained by the community, since most companies have been slow (to say the least) to update their binary drivers? Not to mention, the flexibility involved with porting it to different Open-source Unix OS's and using it with different software configurations and versions?
Thirdly, some people are arguing that that Adaptec will release an SDK in 4 months, but given the history of the Adaptec drivers and drivers by other companies, that would probably involve using a binary driver... which wouldn't help.
I think that if the open source OS's are going be taken seriously by vendors, then they need to act in a professional manner and show their economic strength through well-crafted reports and well-organized efforts.
I support the work of Theo and the other OpenBSD developers -- I believe they are right, but I think the open source community has to join together for a common cause and be professional about such things.
If we, as a community, can make this happen in a professional manner, and win, then maybe, just maybe, we can extend this to other vendors. If we can't pull together, then we're fighting a losing battle against closed source OS's such as Windows and venders such as Adaptec, and we might as well give up now.
We can do this, I know we can. But, we have to do it correctly. So, come on folks, act professional, realize what's at stake, and organize. Think of the visibility the grass roots democratic groups got when they organized and acted like a unified front -- they didn't win the election, but that was surely noticed.
You ignore the words you are reading too. NOBODY WANTS DRIVER CODE. Releasing the docs that say "register x is the array status, 1 means failed, 2 means degraded, 3 means rebuilding, 0 means all good" and other such info will not reveal anything private or hidden. Even if they are using another companies patented stuff, the info we want won't reveal that.
I mentioned the RAID card might be an exception. I'm still not sure your average RAIDed server isn't either
a) SQL server (including exchange)
b) Corporate file server
In my experience its rare to mix things like webserver and mail server. In any case I don't disagree with the main thrust of your argument.
>>The whole MIT/BSD culture can be described with one sentance;
>>Being able to read other people's source code is a nice thing, not a fundamental freedom.
Which is interesting considering that the roots of open source started in MIT's labbs with people like R. Stallman in the 60's. Its a shame they haven't followed the example set by their students.
--- Sig
I though linux already supports the controler. I am not sure if it works like it's supposed to but if it does then I really don't see why the OpenBSD team could not use the linux kernel code with some modifications so that it fits in their case. And as far as Adaptec is concerned they have no reason why they should ever provide the OpenBSD team with info if they don't want to. After all OpenBSD is really unpopular compared to other server/computer OSes (it is some of the best OSS codeing though).
I could see Adaptec going out of their way if there was a Linux issue, because Linux has more mindshare. Even then, I wouldn't bet on a different response unless it was Microsoft making the demand. My guess is that they don't see BSD as a big part of their immediate future, and unless they OpenBSD guys convince them otherwise, it'll remain that way. It's Adaptec's choice to make, and time will tell if the ultimately lose anything from seeing their priorities as being elsewhere.
"Everything is adjustable, provided you have the right tools"
(please ignore my other un-edited, ambiguous comment)
I could see Adaptec going out of their way if there was a Linux issue, because Linux has more mindshare. Even then, I wouldn't bet on a different response. They'd probably take notice of Microsoft timelines, my guess is that MS brings them a lot of income, probably still more than Linux. My guess is that they don't see BSD as a big part of their immediate future, and unless they OpenBSD guys convince them otherwise, it'll remain that way. It's Adaptec's choice to make, and time will tell if the ultimately lose anything from seeing their priorities as being elsewhere.
"Everything is adjustable, provided you have the right tools"
How does running Microsoft Office have anything to do with a headless rackmounted firewall machine in a data center somewhere?
You might not have to worry about drivers in MacOS. Oh wait, you do. A lot of hardware won't work on MacOS due to lack of drivers.
Anyway, you might not have to worry about drivers in MacOS, but you do have to worry about the lack of a decent firewall. And the lack of security features like privsep everywhere. And stack protection everywhere. And execution protection everywhere. These are precisely the features that make people choose OpenBSD over other OSes, and they are no more present on MacOS than they are on Linux (has a decent firewall, but it takes a lot of work to modify with non-standard modules so that it can keep up with pf).
To suggest that MacOS and OpenBSD have even remotely overlapping areas of usefulness (apart from things like the ability to serve web pages slower than Linux and FreeBSD) demonstrates a profound lack of understanding of the issues people need to deal with when they choose an OS.
I rarely criticize things I don't care about.
The OpenBSD crew are looking for documentation, Adaptec is twisting this into demands for support. Theo is fighting so people who use OpenBSD will have a better system that supports more hardware.
Frankly, if it means pissing some people off at Adaptec then so be it. Vendors need to realize that if they don't give the people what they want... people will go elsewhere.
Why Adaptec is resisting providing documentation to a group that will do the work for them so their products are better supported and thus *PURCHASED* more is beyond reason.
Thanks Theo, keep up the good work!
If this is how Theo is dealing with the Adaptec, it's not surprising he isn't getting far. He's like a child throwing his toys out the cot because he can't get an open source implementation from them. Now don't get me wrong; I support Theo's view on the matter, and I agree that having a free (as in speech) implementation is a Good Thing(tm), and I would much prefer this to the current solution being used in FreeBSD. However, Adaptec have already promised an SDK in four months. Sure, it's annoying that we have to wait four months, but at least they've promised something. Frankly, I think Theo is being impatient and hotheaded. Mod me down if you want, but I've seen the dozens of emails come through the FreeBSD-questions list the last few days, and the antagonism on both sides is pathetic. Sure, it's a good thing that people care enough to become emotional, but you sort of expect adults to be able to work through their problems like...well...adults.
no he didn't, dumbass. the quote dealt with incompetence, so it's ironic that the poster would then incompetently misspell.
And that is to ignore and avoid companies who wouldnt release drivers. Look at the pressures on nVidia and ATI currently. Most hardware companies which used to release drivers for unixware, solaris, aix, netware etc are giving linux second or third priority to windows and osx. So as long as the developers arent releasing windows-only drivers, they're under pressure to release OSS drivers.
This is problematic since redhat and suse do distribute binary drivers. If they help, and go OSS-only, people will simply not buy such hardware, and those companies will have to give in.
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
I wonder what the impact, positive or negative of this article will be. Has anyone ever studied what happens when good/bad press on a company hits Slashdot?
Adaptec certainly must know that press like this can't help their sales.
I agree with Theo in this case, since two servers we got several years ago from Dell needed to have their PERC controllers replaced with Adaptec controllers to see the hard drives. I sure don't want to be limited to buying parts from all over to get a server up, I just want to have the standard hardware work for a standard option server from a major supplier. The current platform is Dell, previously it was Compaq, or (shudder) Netframe. While none of our OpenBSD servers need to run RAID right now, the principle fits like a glove.
Theo may have a reputation for being difficult, but I have seen no evidence of that.
What I see is an intelligent, dedicated person who does not suffer fools gladly. And unfortunately, he runs into more morons than he should have to put up with. Those morons then tag him as being "difficult".
Next time you see someone slam Theo, please consider which of the two has been a more valuable contributor, and judge accordingly.
"The empty vessel makes the greatest sound." -- William Shakespeare; Henry V, 4. 4
Same thing with us and a pair of 6 channel SATA LSI cards. For 6 months after we bought them they didn't have support for any of the 1 year old Linux distro's we were running, and you HAD to go into BIOS to rebuild a drive - which meant 10 hours downtime for the system. Fucking insane. Was only fixed a couple months ago.
Not only that, but the controllers are regularly dropping drives for no reason. LSI blames the drive manufacturers and won't do an RMA, saying the drive manufacturer utilities don't pick up all possible errors and that there is no clear problem with the controller, while most of the drives rebuild and test fine. We're ready to toss them in these cards in the fucking bit bucket and do software raid.
Tell that to the bnetd team, and other like projects that shut down due to legal pressure.
The spirit of resistance to government is so valuable on certain occasions that I wish it to be always kept alive
I watch misc@ on and off.
I've seen him chew out people that deserve it.
I've seen him chew out people that didn't deserve it.
I rarely criticize things I don't care about.
It's not any kind of razor, and it's not even attributed to a 'Hanlon'.
:
I fail to see why you view this vague, consensual, Wikipedia post as authoritative. The three authors cited aren't even pre-Napoleonic anyway (Heinlein,William James (WHO?), Robert J. Hanlon ). Heinlen indeed... I like how they put all of them before Napoleon or Goethe, who seems more plausible.
This quote sums it up (from one of the websites used as authorities)
"I did a search for Hanlon's Razor on the internet and was surprised that no one seems to know the origin. The author was my late friend Robert J. Hanlon of Scranton, Pa. - Joseph E. Bigler joeb43@yahoo.com"
I don't think this is on Wikipedia either
- La vérité historique est souvent une fable convenue
Historical truth is often an accepted fable
NetBSD are the portable people *laughs*
Funny beacuse I can get plenty of mission critical-ness without RAID, using flash memory-based devices. So stop flaming and go back to Linux-Land where you come from, *BSD doesn't need people like YOU.
Let me guess: you are one of them? heh certainly gave you the biggest wake-up call you have had in your entire fucking life.
You sir, are a FUCK_TARD!
I also dislike the parent of your comment, but your point on lack of Mac OS drivers is just plain wrong. Every single piece of hardware I've ever plugged into my Powerbook as worked without a separate driver install needed. Not to mention my Powerbook came with oodles of open source drivers, including all the CUPS printing drivers.
nVidia have said it was NDA from SGI that casued them to hide their driver code. SGI said publicly "there is nothing that we would object to them opening up". Since then, this line has been passed on by nVidia apologists but NEVER themselves.
So come on nVidia, whose NDA is holding things up so we can ask them to relax enough for an OSS driver.
But after reverse engineering, the competitor has MORE information about the hardware than they would have got from the specs.
Also, remember that keeping a driver is a cost centre. If they OSS their specs and the OSS coders build their own, they have reduced their costs and increased potential marketshare.
Remember that 5% market is what RIAA are creating insane laws about losing.
We have several hundred Dell 2650 systems all with the "PERC" based raid card. Originally these cars where an LSI based and everything was nice, we rarely if ever had any problems with the RAID or issues with mirror sets not mirroring and reporting as being ok in both windows and linux.
After we upgraded to the 2650 with the Adaptec based cards (to support higher speed CPUs etc etc) we found that we started to get corruptions with the file systems and often the RAID would report as being ok but the mirror would be broken (one disk working the other empty). After many upgrades and bug fixes new drivers and all sorts of things the platform is now relativly stable. however it does not perform as well as the LSI based raid cards that we have grown to love.
For anyone who hasn't read the mail threads here. The much maligned ex-Adaptec employee here, is actually Scott Long the maintaner/developer of Adaptec support in FreeBSD. His addmission of the buggy nature of adaptec hardware is actually an offer of help from another BSD developer. Which was very rudley declined. Some people really need to grow up.
(A word to the wise: there's drivers for Radeon 8500 and up posted from 16th Feb 2005 on the http://www.atitech.ca/ website. They even have x86_64 versions should you wish them. I don't have a Radeon, so I haven't checked how easy to install or reliable they are -- anecdotes say they're quite a bit better.)
Theo is a smart guy, I've been to a lecture given by him, and I can tell you, _I_ was personally astounded by the technical stuff he came up with.
... then fnck them. Good luck... I'll find a vendor who _does_ understand that I'm right... that wants my money... and is smart enough to realise that, providing me with the means to do _what_I_want_ with my Adaptec/whoever's hardware... is more in their interests ( the company ) then it is in mine... ( the customer ).
Thus, I'm pretty sure, when he says that Adaptec are being beligerent, irrational or just plain stupid... it's probably the truth.
I actually believe that a company has a responsibility to allow me to use hardware in whatever manner I see fit.
That doesn't mean said company has to support my desire to run my Dell poweredge on HURD/L4, but, they'd better provide me & other developers with the knowledge of how to make stuff work on our own... if they expect slashdot-nerds (tm), who have sway over IT budget(y) to spend their dollars with Adaptec/whoever else...
That's what we call "keeping the customer happy"... it follows on from another thing we call "the customer is always right"... always. The customer is the most important thing... if Adaptec, can't comprehend that
Probably it's a case of public outing and humiliation of said hardware vendor... to get some movement.
Fair play to Theo for dropping aac support, if the company is going to be beligerent and place barriers in the way of allowing Open Source vendors, to provide top level support, then, I agree with Theo... said, hardware vendor can goto hell !
Go directly to hell.
Do not pass go.
Do not collect my 200 dollars...
I'll spend my dollars elsewhere
!!!!!
Getting the message yet Adaptec ?
Adaptec does not support BSD, only Linux.
I gotta say it
Because BSD is dying...
Seriously how can the Linux kernel people have no problems, but BSD can't get docs; Can BSD ask the Linux people for docs; View the Linux code?
2.6.11 Linux kernel seems very happy with both aacraid and I2O.
I dealt with Adaptec over the whole SVR4/UnixWare driver issue. At the time, I worked for myself, then for a major distributor (since defunct) of Novell products. I started working with Microport SVR4 to implement a video store Point Of Sale system. I had nothing but trouble with the fancy Adaptec 1740 EISA board I got for the project. Narrowed it down to the Adaptec supplied driver included with the SVR4. Contacted Adaptec, got run around for ALMOST A YEAR. Went to work at the distributor, as a UnixWare sales support engineer. First meeting for the public on UnixWare, and the second topic was how Novell fixed the Adaptec driver. This was the upshot after I had given Adaptec core dump after core dump showing their issue. I said "Hell with that bunch" and started working closely with BusLogic. Now, I understand that BusLogic on NetWare was not Plug and Play, but on UnixWare, it was a breeze. No issues, and the board was just as fast and reliable as Adaptec.
After a few years of working for the distributor, I was changing jobs. The sales manager (who was the primary contact I had for the product and support (thanks, Dave)), asked me what would be the natural progression for the product line, and I said "Dave, Linux is coming on big time (this was April, 1995.) It will be big on the net, and there will be a world wide market for hardware supporting Open Source." Next thing I know, I got a demo board from BusLogic, along with the code and documentation for Linux, from Dandelion. Later, BusLogic was purchased by Mylex, (RAID host adaptors), then Mylex by IBM, then sold to LSI Logic. I haven't had one of their boards in a while, but I would love to find a retail source for one.
Needless to say, it worked flawlessly. So these days I say, VOTE WITH YOUR DOLLARS. If Adaptec doesn't support Open Source, find someone who does, and pay them. Open Source gives the world a choice, don't give up yours.
dot-sig.
No, I was talking about their policy on using binary drivers that they cannot redistribute and have no control over. OpenBSD refuses them while Linux and FreeBSD accept them, but I don't know NetBSD's stance on the subject.
I'm sick of following my dreams - I'm just going to ask them where they're going and hook up with them later.
Apple doesn't play in the same league as RH or Novell, they're a hardware vendor. And most of their software is not open. The only GPL'd soft they use is gcc and khtml, and they contribute back the changes. Darwin is based on Mach and FreeBSD, none of which are GPL'd, yet they still contribute back.
"Let me guess: you are one of them? heh certainly gave you the biggest wake-up call you have had in your entire fucking life."
Actually no. I don't post on misc@.
I rarely criticize things I don't care about.
It might be the case that all the hardware you've tried works, but there's a TON of stuff that doesn't work.
PCI cards that don't work on MacOS are more common that those that do, and stuff like USB ethernet adaptors are very hit-or-miss.
I rarely criticize things I don't care about.
Ooh, that's right, GP!
MODED!
Yeah, I'm aware of the fact that RAID management involves all these things. Clearly the product manager was under the impression that the OpenBSD project wanted GUI support. He states it in two emails to the list - they're all in the archives. This shows a clear lack of understanding of what was needed and wanted.
However, your post is pretty informative. I only wish I could mod you up...
XML is like violence. If it doesn't solve the problem, use more.
In order to be ISO certified which they claim to be, they must maintain accurate and complete documentation internally. Clearly they have it, and have shown it to auditors, there is no reason not to release it to the public.
While I'm sure this has a lot to do with Adaptec, Theo is a horrible diplomat.
shop.envescent.com - Computer hardware and more.
My interpretation was ... "go steal it from the linux code" -- means there is no documentations, it is writen "ad-hoc"
-Ron
Bill Gates has done very well by producing low cost software that works well enough for the corporate and home user.
It is not unethical or illegal to produce closed source code for a profit.
This includes not giving away your internal file formats, internal alogrithms, internal api calls and anything else that makes your software better than someone else's.
For the case in point, if you find the closed source raid device driver objectionable, then do four things:
1. Reverse engineer it
2. Publish a white box set of specs for using the device
3. Find someone to implement it that has not looked at the internals of the device driver
4. Have no involvement in writing the open source device driver
I live in Calgary and I have Theo(Beer Night at Brewster pub). Snippy does describe this mans attitude as well as severly arrogant and self centered. He does not work at all well with others.