Slashdot Mirror
Sun Opens Cobalt Code
Bush Kanaka writes "It looks like Sun has released the user interface and back-end custom code for the Cobalt Raq550 under a BSD-like licence. The BIOS code is also, apparently, now open source and is being maintained by Sun engineer Duncan Laurie in his own time. This has to be good news for all those Cobalt devotees who were annoyed when Sun killed off Cobalt last month, but is anybody going to actually pick up the software and start making their own Cobalt clones?"
Previous article about Sun taking the Cobalts off the market.
...not too bad, some duplicates found by CPD.
The Army reading list
Was I the only one who accidentally read, "Sun opens Cobol Code," and thought, "Egads, some things just need to stay closed-source" ?
--TheOrangeSquid Is it any wonder things seem so awry? We swim in a sea of confusion and don't have to think to survive
Umm, that's probably because it IS the same Cobalt that produced the Cube. Sun bought them several years ago (and subsequently killed them off by failing to release timely security patches or regular updates anymore). I've got several RaQ4r's and I should find out how this affects us when we keep them around to use as workgroup toy servers after we upgrade to general purpose Linux servers. They were great and I would keep using them if only Sun kept patches up to date. Oh well.
The number of sites on Cobalt has declined since August 2002, when it reached its peak of 3.1 million hostnames and 942K active sites. Our November hosting survey found Linux-Cobalt serving 871K hostnames and 527K active sites.
Percantage of Linux Active Sites with Known Linux Distribution:
Redhat 51.7%
Cobalt 19.6%
Debian 15.4%
Suse 10.5%
Mandrake 1.9%
Gentoo 1.0%
From NetCraft
I've got a three-system (used) computer rack that cost in total around $350. It helps to know somebody in the computer department of a large business when it comes time for them to unload their stock, especially when the alternative is for them to pay to dump the stuff.
Try not. Do or do not, there is no try.
-- Dr. Spock, stardate 2822-3.
I mean, think of all the $$$ in license fees that can be extorted, I mean, earned from users of this newly open sourced platform that probably steals heavily from UNIX contributions by licensees and not actual work performed by SCO...
Blogging Weight Loss, Distance Education, and more at verlin.com
The hardware on the other hand was really shitty and ended failing no matter what version cobalt raq/qube you had. I have a cobalt raq that doesn't work and also a cobalt qube 3 that has some misc. hardware issue. I hope someone can work the source and make a good UI for a generic platform.
It's rack-mountable. So if you're a hosting company you can stick a ton of them in a small place. Hence the name 'RAQ'
Symantec will still develop it, all the Gateway Security firewalls and Raptor firewalls were built on cobalts
Is it just me, or does this thing sound exactly like the Cobalt Cube from a few years back
It _is_ the Cobalt Qube software. Jesus, I know nobody reads the articles, but is it really too much to ask that you read the title of the message you are responding to?
- Old Man of the Mountain ---- "I want to disturb my neighbor"
http://open.cobaltqube.org/
Featuring Blue Quartz
Is it just me, or does this thing sound exactly like the Cobalt Cube from a few years back?
This is the first first post I've seen that was dumber than a "First post!" first post.
- Old Man of the Mountain ---- "I want to disturb my neighbor"
Sun has not owned Cobalt for a very long time IIRC. Do they still have anything substantial they got out of the purchase, or was the whole this money down the drain at this point?
Did they need to purchase Cobalt to "get their heads around" Linux? Did they need Cobalt to figure out how to make a low end Intel server? Maybe Sun has less between their ears than I gave them credit for, but I don't believe either is the case.
Seems like a bad business decision to purchase them to me, but maybe I'm missing something.
Soccer Goal Plans
The Cobalt software can be found at: open.CobaltQube.org and the ROM can be found at SourceForge.
This gesture is far more useful to those who are "stuck" with an end-of-lined server. If there are flaws with the software (and with Raqs, there are always a few that creep out over time), the admins can do the fixes themselves.
I really doubt that anyone is going to use the Raq software to start a new project when there are cheap, well-exercised alternatives on the market.
Thus inviting disaster to my oh-so-dainty home DSL line, I nevertheless boldly post a link to a set of humorous uses for a dead Cobalt Cube. These pictures were done by a former coworker of mine, Scott Lundberg (Hi Scott!) about four years ago.
http://www.mbuf.com/deadcube.php
In a related rant, to heck with the upper level source code. I want them to release the schematics, BIOS, and boot code for the ARM cube! Having a turnkey ARM system like that would be an incredible boon to the embedded Linux world. Open Source doesn't necessarily have to mean just source code, IMO.
Has anyone made an "Internet Appliance" to replace the Cobalt RAQ line? Although the RAQ line was not perfect, it's ease of use was unmatched.
That's pretty cool, I guess we'll see a slew of userfriendly server distros based around that... And dedicated server companies installing them on cheap hardware instead of the non-free alternatives!
Gotta admit that it's pretty nice for a company to open-source what they will not use for profit anymore..
First, those of us who rely on Cobalt appliances will stand a better chance of finding useful updates and peer support as no doubt from the existing Cobalt users communities a thriving community will appear around this project. I have spoken in the past about Sun's lacklustre approach to providing software updates for these boxes so from there any progress can only be an improvement.
Second, the Cobalt www based admin software is not that bad. In fact, I'd go as far as to say it's quite good. It allows people who would normally have no idea to administer a www connected server appliance and having at times seen some of the competing commercial offerings I'd say it does that job well and I'll certainly spend some time poking around inside it. To have this project in the public domain as open source can only be of benefit to the open source world as a whole.
I await with interest further developments upon this piece of software.
Oxford Dictionaries Online
Aside from Genome integration, Solaris userland has been static for many, many years (far too long IMHO - Sun relies upon sunfreeware.com far too much).
Why would the Cobalt code not be useful as part of the base install?
Actually, what I'd like to see is Sun pick the best GNU distribution and wrap the Solaris kernel and libc around it. Sun releasing GNU/Solaris would prove that there is life in the old girl yet.
MOD PARENT UP!
.php.inc files in the user/admin interface were left browsable, so that administrator passwords could be easily viewed if attacker knew their location on the filesystem. And with opening the source to the UI, things like that will happen.
He speaks the truth.
He might be wrong about 3 users, but now all the people that ARE using RaQ's will be that much more open to attack.
Besides, Cobalts arent exactly well-known for security, either. I've heard that
sure the code may be open now, ( which is nice ) but how would you get a hold of cobalt hardware?
Last I heard it was a custom Mips setup.. Not exactly a 'commodity' item...
Seems more like a PR move then something practical...
( and no, ive NOT looked at the code yet. so my statement may be off )
---- Booth was a patriot ----
They should open their eyes and realize how valuable the cobalt line was to the market it served and how that market is coming right back again in the form of consumer level servers. Consumers are gathering large amounts of content (mp3s, movies, photo albums, etc) and they are starting to want to host this content - THEMSELVES! This is a nice market for low cost, quiet, easy to manage servers. The average consumer doesn't want to administer Linux, the want to administer something along the lines of a Cobalt server (or easier).
Sun once again pisses away an opportunity because they can't see the forest for the trees.
I have a Qube-1 (2700). But I can't use it anymore; consequences of a failed attempt to upgrade the hard drive. It will only boot with the original disk, and it's really not clear what you're supposed to do, to get another disk to boot (or even to wipe and reinstall the original). It's a royal pain to try it. Netboot doesn't actually work for me, and I don't think the original ones did that. There's no serial interface, and the circuit to add one are based on a later version of the Qube. Basically, I got stuck there, so the Qube is just a bookend today.
Even to the extent that it works, the MIPS version of the linux kernel doesn't work on this flavor of MIPS. I followed the cobalt-bsd list for a few years, and there was never a real success story about bsd on one of these things either. Now that I have a Shuttle XPC and Mini-ITX, the only thing that's really interesting about the Qube is that it's blue, and it's something that "nobody" has.
Aside from Genome integration, Solaris userland has been static for many, many years (far too long IMHO - Sun relies upon sunfreeware.com far too much).
:-) You yourself say that it's all on sunfreeware, and since all sane solaris users use jumpstart and customise the builds, you can have GNU tools on anything if you want them anyway. It's flexible - and I like that. The GNU tools are there is you want to use them (and I agree you probably would :-)) - doing it this way is win/win, IMHO.
:-D
Some people like Solaris because of this. Migration is easier when few things change. Others also think that it cost Sun a big part of the workstation market. Some people's good choices are other people's bad choices.I personally am glad it doesn't move too much, but I can see why you're not.
Actually, what I'd like to see is Sun pick the best GNU distribution and wrap the Solaris kernel and libc around it. Sun releasing GNU/Solaris would prove that there is life in the old girl yet.
I disagree. I don't want them to do this because you can do it yourself - that's man/hour's that would be better spent on *real* solaris dev
P.S. All our builds have most of the GNU Tools available, though not installed, on a shared netfiler. Just about everyone mounts them and puts it in the path before most everything else
Try NetBSD... safe,straightforward,useful.
I wonder how long it will take SCO to find something of their IP in the Cobalt code, now that its open source
--
One by one the penguins steal my sanity...
The netbsd port to the qube has some problems with the 2700, but from the mailing list people are getting close to solving them. Give the netbsd cobalt restore cd (works like the old cobalt restore cd but puts netbsd on the disk) a try if you are not using the machine anymore.
I've recently started thinking about how our local Free-Net (which provides Web hosting to nonprofits, among other things) could set up a control panel for domains we host. Is poking through the Cobalt code for stuff to steal a good idea, or are there already better Free control panel programs available?
My Web Page
What does this require hardware-wise? Is it specifically designed for the hardware they ran it on? Can I download the code and have a 'Cobalt' 486 server? And is it distro-dependent?
:)
Sorry if these are answered somewhere, but I can't find them, and I'd like to know if it's worthwhile for me to download a 12 MB file at 4.8 KB/sec, or if it won't work on anything here.
________________________________________________
suwain_2
forgive my "newbiness" but does this mean that one can now take the complete cobalt software and now apply it to one's own hardware? Is it really as simply as that?
I use to work at a small company (about 6 full staff) which used an old Qube 2 box for distributing email around the office. Although restrictve and sometimes basic customisation required diving into the horrible world of 70's style telnet-based command line editing. Despite this, I thought the web-based administration system used was amazing.
No mucking about, simply point web-browser at the Qube, type in user-name and password and then a few simple web-pages allowed common adiministration tasks to be performed by the most unskilled of 16 year old work experience kids.
I always thought this provided the ideal interface for computer/network-controlled home appliances. Personally I'd find it really useful to log into my house from work and make sure lights are switched off (or switched on before I arrive home), pre-heat oven, turn hot water on etc.
Surely such control is already technoloically viable using simply a qube style interface, an always-on home internet connection and a computer linked into lighting/kitchen systems (for examples above).
Does anyone know of any projects/progress for this sort of thing?
You might want to rethink your position for a few reasons:
For these reasons, and a few more, an unsupported release of GNU/Solaris (to compliment Solaris/Classic) would be of great benefit to Sun, and work on it should be commenced as soon as reasonably possible. Debian is probably to be preferred.
Sun is no longer the UNIX market leader, nor are they the SPARC performance leader (being dethroned by Fujitsu). Sun needs to throw out a few new products to regain the lead. GNU/Solaris may become an integral part of Sun's survival (or it could be a white elephant - you might be right).
p.s. Just imagine the fun that Sun could have with "uname -a" under GNU/Solaris!
Also of course, the cobalt rom is immensely useful, since that's how you load your kernel which loads your kernel :) It might be best to just put your actual kernel into the flash, though I've heard of people having had cobalts so long the flash has gone bad or something, and the system dies permanently when they try to update the rom.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
I see a bios image, I see source code.
I guess you can emulate the bios somehow because it won't be flashable in any motherboards, that doesn't take monkey brain to figure out, but what about the rest for non-linux users?
--- Metamoderating abusive downgraders since my 300th post.
Hey, I've got an old RaQ2 MIPS that I've been looking for updates or more importnatly I new kernel but finding LINUX MIPS for the RaQ2 is next to impossible.
I agree with most. These boxes were unbeatable back in the day. Truly plug and play and an unbelievably straight-forward user interface. Many companies today could learn from Cobalt's interface approach.
is anyody going to actually pick up the software and start making their own Cobalt clones
It doesn't matter. If nobody does, then Sun was justified in killing it off. If people do, then Sun should be congradulated on doing the right thing.
This makes me feel a lot better about using Sun technologies. I know they have a tight control on Java now in order to make sure it does fly out of control as some people beleive happened to C++. This makes me feel that if Java does become legacy and unsupported, that the systems I'm building now with have another avenue of support.
-no broken link
Ooh I finally got through. Here's a mirror of the BlueQuartz installer for RedHat 9.
Try www.netmax.com. Linux in a box with a remote UI. Works great with a headless PC. Don't even need a CD-ROM past installation.
So when is Microsoft gonna' release the Bob source code? Or are they waiting for Bob's second coming?
I would guess that you guys do not know cobalts very well. Its a linux bases OS (already open) with a apache (alread open) based frontend written in php (human readable code). So all this really changes is the license. The only thing that is of intrest is the CCE portion which was opened with the open qube. Get a clue before you post.
I am a republican not by choice, but rather by lack there of.
Hopefully this will silence some of the Sun == Microsoft trolls for a while.
Seen any meaningful OSS contributions from MS lately?
Such a server is just as valuable with source code available. The target market you described is not going to download the source and build their own from an ITX case. In fact, if it does what I need for a reasonable price with open source, I'm not going to either. I've got too many projects going already.
The business problem with opening the source is that a competitor could offer a slightly cheaper box with the same software. This is unfair for the company developing the software/integration scripts. For this kind of situtation, I think delaying release of the source is appropriate. It seems to work for OpenGL (where last years features are open source).
If the software is almost all open source to begin with, then the companies are competing on hardware features/sexiness/reliability - go to it!
ummmm... dude? Are you on crack or are you just fucking stupid!? Go back to MS Windows you POS, we don't need Sun or anyone interfering with Linux, we're going to take over the world and then IBM/RH/SuSe is going to take us over!
i had emailed apple regarding the cobalt products when osx was in version one, stating that this was the perfect apple like device. hell, the founders came from apple. then they come out with that damn cube. duh! i currently use plesk but, it just is way to complicated for somebody with actual work to do. when i had a t1 and the raq as well as nasraq, my maintenance was really simple. this type of product coupled with a good groupware solution for small business, and perhaps backup solutions would be quite nice. the cobalt os is now generalized as opposed to the first mips based products. mips is great however.
if anybody wants to start a group/business around this, i'm in the chicago area, and can be reached at anything (info) plus simple interfaces dot com.
that's a pretty big percentage of users! certainly they are a bit off, but still...
"is anybody going to actually pick up the software and start making their own Cobalt clones?"
Im not sure, but im sure that someone will be hunting through the sauce, searching for exploitable code...
I know a number of small local ISP's that used them for the same reason and many are dumping them dirt cheap and we are purchasing what we can afford, keeping some for parts, and coloing others, and using some for in house application development.
At least when Sun's products at EOL they release the code, unlike some other companies who's idea it is, sorry WIn..I mean Product X is now no longer supported: please spend more money to upgrade now!
"The problem with socialism is eventually you run out of other people's money" - Thatcher.
there are cheap, well-exercised alternatives on the market.
If you want to use code to do a server appliance, you'd get further, faster using something like Webmin.
"#!/bin/sh" appears at the beginning of several files in both distributions.
You can't judge a book by the way it wears its hair.
Early model Qubes had MIPS CPUs, not ARM...
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Much of the Netwinder drivers, etc. went into the Linux kernel as part of the ARM version of the kernel. This includes framebuffer drivers and framegrabber drivers for the IGST/Tvia CyberPro 2XXX and 5XXX series of set-top display adapters.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
They have yet to really sell to the consumer. Each and every piece that they do is either dedicated to selling more server iron, engineering workstations, or screwing the competition (If you don't think StarOffice/OpenOffice isn't scorched earth intended for Microsoft, you probably ought to look at it closer.).
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
I've been working with the guys over at open.cobaltqube.org to get the damn thing working on Debian.
My biggest problem to date is that the RaQ550 isn't gcc-3.2/3.2 aware. It also ain't updated for newer PHP versions, while the Qube3 release is.
So far we've got the Cobalt Configuration Engine (cce backend) and minor parts of the web interface working.
Performance is quite a bit better than that old Linux code, and NetBSD's pkgsrc package system is pretty slick..
The old MIPS based qubes really should have their bios published/open sourced, since they have quite severe limitations on the size of the kernels one can load, so an open sourced bios could be abused as a second stage bootloader. Or is that already avaliable, and I've missed it?
Were the Cobalt boxes perhaps manufactured by a subcontractor that might still be willing to produce them?
It appears there is still a market (but with new hardware inside!), including a loyal fan community. Who is going grab the business opportunity?