2CPU.com » News » April 2005 » Kernel Changes Draw Concern from Open-Source Community

eWeek is running an article discussing the growing concern in the open source community that the Linux kernel is becoming too fat.

Sam Greenblatt, a senior vice president at Computer Associates International Inc., in Islandia, N.Y., said that while the kernel is evolving for the desktop, server and embedded markets, more and more technology is being included, and the kernel is "getting fatter. We are not interested in the game drivers and music drivers that are being added to the kernel. We are interested in a more stable kernel."

Is the desire to get Linux more acceptance in the desktop world compromising its viability as a solid server OS? Discuss.

03/28/2004 10:00 PM: Linux Kernel Comparison: 2.6.4 vs. 2.4.25 @ 2CPU.com! by Jim_
I finally found the time to put together a short Linux kernel comparison for the SMPers out there. I ran a variety of benchmarks including Apache, MySQL, and dbench version 2.0 on my dual 3.2GHz Xeon ...

05/15/2000 01:56 PM: Linux Kernel configuration by J0rdan
I had a request to go over some kernel configuration/tweaking in the Linux forum. I went through a general workstation configuration of mine that should bring your kernel down to the 400k range. I had...

jives BP6 User Posts: 2419 Joined: 2001-05-18

#35875 Posted on: 04/20/2005 11:10 PM If you don't want something in the kernel build your own without it. Of the linux Distros I've used starting from 2.2 up to 2.6 now while yes the kernel has gotten bigger I haven't seen a drop in Stability using the vanilla kernel.

cmost Back in St. Olaf.... Posts: 416 Joined: 2002-08-13

#35876 Posted on: 04/21/2005 12:13 AM I couldn't agree more! The beauty of the Linux kernel is that it's like Burger King...you can have it your way. If one doesn't need the so called desktop features for his/her server, then compile the kernel without them. Sometimes I think people are just getting too lazy to do things themselves...everything has to be prepackaged for them. AMD Athlon 64 X2 3800+ 4096 MB DDR 400MHz DDR2 RAM GeForce 9400 GT w 512 MB DDR2 Primary OS: Parsix GNU/Linux (built from Debian Testing) Compiz-Fusion Git Other OS: Windows Server 2003 SP2 (as workstation); Windows Fundamentals for Legacy PCs; Windows XP Pro. SP3 - via VirtualBox

Glock Glock Operator Posts: 205 Joined: 2001-06-13

#35877 Posted on: 04/21/2005 12:38 AM Jives, while I understand your reasoning, it is this mindset that keeps Linux out of the mainstream and keeps commercial developers from porting their software to Linux. * For those who think The Gimp == Photoshop, hit the "Back" button now, there's nothing interesting for you in this post... There is a difference between programmers and users. The first class makes the programs, while the second class uses them. Compare it to an ecosystem. The one can't survive without the other but still have to find a way to co-exist. Evolution is slow, Linux is not there yet. I think it will, someday, but not quite yet. Not everyone has the knowledge (or time) to compile customized kernels. Those who have, hats off to them, as they are the elite programmer-class I was talking about. They don't need general acceptance, as this class of user can take care of itself. Also don't expect Linux to rule the market if it tries to be just another Windows (adding feature after feature, only to find out that years later, its codebase is bloated and mutilated beyond recognition - full of exploitable bugs and stability issues). That is why, for Linux to succeed, it must first find equilibrium. I don't know if I can get away with another analogy to natural evolution, but maybe mutating is a viable option. IMHO, Linux has to split into a clearly defined SERVER and WORKSTATION class OS and evolve from there. The difference has to be at kernel-level and not just included apps and colorful themes, as is now the case. While most of us agree that the MS OS'es are technically inferior to Linux etc., you can't deny that they have a clearly defined product line (XP Home, XP Pro, 2003 Server, Embedded). And this is something Joe Sixpack and the corporate bobo's need: clarity. And for those other people, clarity would not hurt, wouldn't it? "What is understood, need not to be discussed..." Loren Adams

cmost Back in St. Olaf.... Posts: 416 Joined: 2002-08-13

#35878 Posted on: 04/21/2005 01:14 AM MHO, Linux has to split into a clearly defined SERVER and WORKSTATION class OS and evolve from there. The difference has to be at kernel-level and not just included apps and colorful themes, as is now the case. While most of us agree that the MS OS'es are technically inferior to Linux etc., you can't deny that they have a clearly defined product line (XP Home, XP Pro, 2003 Server, Embedded). And this is something Joe Sixpack and the corporate bobo's need: clarity. And for those other people, clarity would not hurt, wouldn't it? I believe this is already the case to some extent. While Microsoft does indeed have the aforementioned discreet versions of Windows: Home, Pro, Server, etc, you are quite right that differences exist at the kernel level as well as the bundled software. With Linux, there are also different distributions geared toward discreet purposes. For example, Red Hat Advanced Server and Enterprise Server are clearly aimed for the back room while Red Hat Workstation or Fedora are aimed squarely at the desktop. In these cases, different features may be compiled into the kernels of each type of operating system by Red Hat. The problem is that ANYONE can recompile their own version of kernel in any of the discreet versions adding or removing features at will. This is not possible with a Microsoft offering. In other words, one cannot recompile XP Home to gain Server 2003. While some believe the ability to modify the kernel to be a boon, others believe it to be a hinderance to Linux taking the mainstream by storm. My opinion is somewhere in the middle. If the Kernel maintainers fork the code into discreet versions aimed at different uses, then we'll eventually wind up in the same boat as Microsoft. Therefore, I think the answer to this conundrum is somwhere in the middle too. P.S. Recompiling a kernel isn't all that bad or time consuming, even for a novice. The only time i've ever run into problems is when i've had to merge a lot of custom patches from a particular distributor. AMD Athlon 64 X2 3800+ 4096 MB DDR 400MHz DDR2 RAM GeForce 9400 GT w 512 MB DDR2 Primary OS: Parsix GNU/Linux (built from Debian Testing) Compiz-Fusion Git Other OS: Windows Server 2003 SP2 (as workstation); Windows Fundamentals for Legacy PCs; Windows XP Pro. SP3 - via VirtualBox

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35879 Posted on: 04/21/2005 01:20 AM Originally posted by Glock While most of us agree that the MS OS'es are technically inferior to Linux etc., Anyone who says that is part of the lower class of the programmer class (to continue your analogy) or just a dumbass. If you say that it means you know neither the internals of *nix (Linux, BSD) nor of Windows NT/2k/XP. I have no idea how people can be so arrogant as to think that the employed programmers at Microsoft are just bumbling idiots (or that management is so stupid as to prevent them from doing anything). I agreed with everything you said up until that point. As for the article, it should come as no surprise that the kernel is getting "fatter". New features to improve desktop/workstation performance become dirty hacks without kernel support. Somehow people think that you should have a microkernel when the kernel is already monolithic. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

cmost Back in St. Olaf.... Posts: 416 Joined: 2002-08-13

#35880 Posted on: 04/21/2005 02:25 AM As for the article, it should come as no surprise that the kernel is getting "fatter". New features to improve desktop/workstation performance become dirty hacks without kernel support. Amen to that! AMD Athlon 64 X2 3800+ 4096 MB DDR 400MHz DDR2 RAM GeForce 9400 GT w 512 MB DDR2 Primary OS: Parsix GNU/Linux (built from Debian Testing) Compiz-Fusion Git Other OS: Windows Server 2003 SP2 (as workstation); Windows Fundamentals for Legacy PCs; Windows XP Pro. SP3 - via VirtualBox

i_wolf labhair dom as gaelige Posts: 2097 Joined: 2002-11-19

#35881 Posted on: 04/21/2005 03:56 AM Anyone who says that is part of the lower class of the programmer class (to continue your analogy) or just a dumbass. Come on, be nice, I don't think he was attacking you personally Arguing the software engineering pro's and con's of the NTKernel is pointless given the fact that its a closed source product. There are a few areas where it is reasonable to extrapolate conclusions from and make basic assessments. One I've made to myself a few times is that I am hoping that Longhorn will improve the load balancing over Win XP Pro. FBSD seams to be the standard to beat in this regard presently. As for the size of the Linux kernel. I think its somewhat rediculous to argue that kernels have gotten too big. They have gotten bigger to provide more functionality. Thats something I like to see in the computer industry... evolution. The codebase hasn't gotten bigger because the standard of coding has plumeted. It has expanded in size to provide support for the ever growing number of uses people have for their computers today... a lot of which are long over due. Hung like a donkey. Go like a horse!

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35882 Posted on: 04/21/2005 04:22 AM Originally posted by i_wolf Arguing the software engineering pro's and con's of the NTKernel is pointless given the fact that its a closed source product. There are a few areas where it is reasonable to extrapolate conclusions from and make basic assessments. There's plenty of public documentation regarding its design and inner workings. You don't need the source to come up with conclusions that need no more than a black box. Most people take it on word that Linux 2.6's scheduler uses multiple linked lists to achieve O(1) status, and it can be easily tested using blackbox testing techniques. The same applies to NT. Whether it uses Linked Lists, a dynamic array, or whatnot is inconsequential. Come on, be nice, I don't think he was attacking you personally I didn't attack him, I said anyone who makes that statement doesn't know the inner workings of either system. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

Glock Glock Operator Posts: 205 Joined: 2001-06-13

#35883 Posted on: 04/21/2005 06:00 AM Originally posted by AssKoala I didn't attack him, I said anyone who makes that statement doesn't know the inner workings of either system. But you misunderstood my statement (about the inferiority of the MS OS'es). Probably my fault, I was trying to say too much in a too short sentence. What I was trying to say is that when something gets more complex, it gets more prone to errors. Entrophy kicks in and before you know it, you're generating new bugs while fixing old ones. * Arrogance is saying that, as a programmer, it doesn't happen to you. I was not flaming the MS programmers or project managers (if you read my post, you'll see that I even applauded their effort to produce clarity in their product line). Microsoft's OS seems to be caving in under its own weight. It is a victim of its own success. Why do you think more and more features in Longhorn are being cancelled or delayed? They're too busy cleaning up the 'old' code. But this is a GOOD thing. I would very like a working XP and not some buggy OSX clone * Also, one does NOT have to be an 1337 kernel hacker, to comment about the performance and reliability (in)capabilities of any OS, or that it is improving or degrading. To fix it, yes, but to comment on it, no. Even lower-class users can, to some extent. Bottom line and referring back to my original post: I'm slightly worried about where Linux is going. Like it is now, Linux and Windows are growing towards each other. Windows has the features, software base and ease-of-use (talking user level here) and Linux has the performance, security and reliability fighting for it. Logically, as Windows becomes more secure, Linux will gain some weight. What impact this will have on its security and reliability is yet to be seen, but the signs are beginning to show. That's why I think we need a kernel-fork. One sleek and sexy Linux for those who know what a computer is really for, and one for the masses (read: desktop) to compete with the toy-OS'es of the world. The good thing about the latter is that it will convince commercial application developers, to port their software to Linux (Ahead Nero Burning Rom anyone?). Users of the sleek and sexy Linux can then run that app on their more reliable and better performing machine and actually get some work done (but without the fading windows and warping icons). Last but not least: my main platform (at home and at work) is Windows XP. While Linux is promising, it can't imagine it on my desktop YET and if I wanna get down to some real computer action, I fire up my SGI and get some IRIX action (a real UNIX). That's the stuff that gets my blood pumpin' "What is understood, need not to be discussed..." Loren Adams

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35884 Posted on: 04/21/2005 06:19 AM Originally posted by Glock and Linux has the performance, security and reliability fighting for it. That's where I disagree with you. In workstation applications, Linux performs no better than Windows. In most, Windows will perform better in terms of user performance (higher priority for in focus windows, etc). Security in Linux is a lack of information (in other words, it's practically nonexistant for the "Workstation" user). If you want a truly secure box, you have to practically make it unusable for anything other than running services. If you don't do that, you have new vulnerabilities being found quite often. If someone wants to own your box, they can fairly easily pull it off on all but the most paranoid users. As for reliability, that is hard to quantify. I won't argue that Windows is more reliable than Linux, but I certainly haven't had to reboot, deal with a crash, etc in Windows in...well... I can't remember the last time I've had a BSOD or crash of any sort. On the other hand, my Linux box and the Linux boxes in the various computer labs here seem to do quite well at crashing themselves. Not quite as good at recovering from errors (not system errors, I mean people yanking network cables and the like, but the same systems dual boot Windows), so long as everything is going well they'll be fine. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

Mr. Hahn Unregistered

#35885 Posted on: 04/21/2005 06:36 AM I dunno dude. I can't compare the boxes in the lab to my own. Generally since my own are better kept. My window/bsd/osx boxes I have at home and the dorm are all rock solid, but goto the coc and see how fast one of those windows boxes crashes, or even the osx boxes for that matter. Finder is never stable in that damn fruit lab. Not to mention the network setup sucks balls, the tcp traffic is slow as hell and I bet it is because some router allong the way isn't allowing mtu discovery or something of the like. As far as the reliability of any system goes: any os is unreliable as a workstation. There is just too much user interaction and screwing around that goes on. Windows gets screwy, osx gets screwy, linux gets screwy. Back to the original thing about the way systems work. I agree with the koala's ass. Anyone that knows how the bsds and system Vs and NTs work doesn't say one is better than the other. At this point they for the most part good at managing system resources, so the real thing to go after is what platform gives you an upper hand at doing whatever task it is you are doing.

cmost Back in St. Olaf.... Posts: 416 Joined: 2002-08-13

#35886 Posted on: 04/21/2005 06:08 PM Security in Linux is a lack of information (in other words, it's practically nonexistant for the "Workstation" user). If you want a truly secure box, you have to practically make it unusable for anything other than running services. If you don't do that, you have new vulnerabilities being found quite often. If someone wants to own your box, they can fairly easily pull it off on all but the most paranoid users. This isn't true. The way Unix permissions and security are setup by default make it more inherently secure than Windows' default setup. While it's true that a determined hacker can own your box if you don't take precautions, you hardly need to be "the most paranoid" to keep your box secure. I read recently that when a Unix box and a Windows box were both placed on the Internet, the Windows box was attacked by bots and other malware and taken over within an hour while the Unix box required at least a month before it was compromised. Granted, the number of malware targeting Windows is much more prolific even though Unix predates Windows by about a decade, this experiment is still telling of Unix default security. AMD Athlon 64 X2 3800+ 4096 MB DDR 400MHz DDR2 RAM GeForce 9400 GT w 512 MB DDR2 Primary OS: Parsix GNU/Linux (built from Debian Testing) Compiz-Fusion Git Other OS: Windows Server 2003 SP2 (as workstation); Windows Fundamentals for Legacy PCs; Windows XP Pro. SP3 - via VirtualBox

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35887 Posted on: 04/21/2005 08:24 PM Originally posted by cmost This isn't true. The way Unix permissions and security are setup by default make it more inherently secure than Windows' default setup. While it's true that a determined hacker can own your box if you don't take precautions, you hardly need to be "the most paranoid" to keep your box secure. I read recently that when a Unix box and a Windows box were both placed on the Internet, the Windows box was attacked by bots and other malware and taken over within an hour while the Unix box required at least a month before it was compromised. Granted, the number of malware targeting Windows is much more prolific even though Unix predates Windows by about a decade, this experiment is still telling of Unix default security. Ah, but now that's a different discussion. The fact that Windows, by default, has you running as Administrator is a problem. However, that is changing with Longhorn. And when I talk security, I don't mean "how long it takes to get attacked". The time to get attacked just means people aren't attacking you. It doesn't mean you're secure. When I say a system is secure, I mean it is free of known exploits or should it have exploits, it has had workarounds put in place to avoid them. Because you aren't attacked, means in NO WAY WHATSOEVER that you are secure. Your example about the Windows box and Unix box is meaningless. They are BOTH insecure as they were both compromised. The fact that Windows has more attacks placed on it just means it'll take less time for one to break through. Also, keep in mind that most of those tests are done with an unpatched Windows and Unix system out of the box. I have no idea how running an unpatched system proves anything. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

i_wolf labhair dom as gaelige Posts: 2097 Joined: 2002-11-19

#35888 Posted on: 04/21/2005 08:42 PM There's plenty of public documentation regarding its design and inner workings. You don't need the source to come up with conclusions that need no more than a black box. Most people take it on word that Linux 2.6's scheduler uses multiple linked lists to achieve O(1) status, and it can be easily tested using blackbox testing techniques. The same applies to NT. Whether it uses Linked Lists, a dynamic array, or whatnot is inconsequential. Thats not what I am talking about. Basically I was commenting on the original comment you had made that it was crazy to imply that Microsoft's coders are any less capable than an Open Source coder. that is why i said Arguing the software engineering pro's and con's of the NTKernel is pointless given the fact that its a closed source product. i.e. You cannot see how the kernel was written so you cannot check to see if silly little programmer induced performance problems have crept in like unnecessary castings back and forth etc... The things that can tell you somebody's capability as a coder is good or not as the case may be. As for the inherent security of each OS.... well I remember somebody here once commented that security is 95% administrator 5% OS. I thought that this was quite apt. At the same time, the permissions system in Unix is an initial major deterrant in the war for security. If a would be hacker does manage to penetrate your firewall and gains access to your box , the intrinsic permissions system will likely limit the damage he/she can wreak... Hung like a donkey. Go like a horse!

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35889 Posted on: 04/21/2005 08:48 PM Originally posted by i_wolf somebody here once commented that security is 95% administrator 5% OS. I completely agree with that. Which is why I think the idea of using the "unpatched box on the net" test is idiotic. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

rmn oh my, it's huge! Posts: 6013 Joined: 2002-01-26

#35890 Posted on: 04/21/2005 09:02 PM Originally posted by AssKoala The time to get attacked just means people aren't attacking you. It doesn't mean you're secure. That's precisely the reason why Llap-Goch was created! RMN ~~~

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35891 Posted on: 04/21/2005 10:43 PM Originally posted by rmn That's precisely the reason why Llap-Goch was created! RMN ~~~ Holy shit...where do you find that stuff Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

rmn oh my, it's huge! Posts: 6013 Joined: 2002-01-26

#35892 Posted on: 04/21/2005 11:08 PM It was on a papperbok. http://www.intriguing.com/mp/_scripts/brandnew.asp RMN ~~~

cmost Back in St. Olaf.... Posts: 416 Joined: 2002-08-13

#35893 Posted on: 04/22/2005 01:09 AM Because you aren't attacked, means in NO WAY WHATSOEVER that you are secure. Your example about the Windows box and Unix box is meaningless. They are BOTH insecure as they were both compromised. The fact that Windows has more attacks placed on it just means it'll take less time for one to break through. It isn't meaningless. I thought we just went through this in one of the other threads? Typical home / non-techie users don't know how to get and/or apply patches, therefore, the security of a default installation becomes VERY important and relevant. The fact that Windows, by default, has you running as Administrator is a problem. However, that is changing with Longhorn. Easily remedied, fortunately! Incidently, i've noticed that every time there's a problem with Windows somebody is quick to point out how it's no big deal because it's going to be fixed in the next version of Windows, whose coming is always billed as right up there with the second coming of Christ. This phenomenon has been happening since the days of Windows 1.0. Yet, with Linux, if it isn't perfect the first time then it means Linux is an inferior product and not worth the time. Why do you suppose this is? AMD Athlon 64 X2 3800+ 4096 MB DDR 400MHz DDR2 RAM GeForce 9400 GT w 512 MB DDR2 Primary OS: Parsix GNU/Linux (built from Debian Testing) Compiz-Fusion Git Other OS: Windows Server 2003 SP2 (as workstation); Windows Fundamentals for Legacy PCs; Windows XP Pro. SP3 - via VirtualBox

AssKoala Anti-Zealot @ GATech Posts: 3309 Joined: 2002-01-02

#35894 Posted on: 04/22/2005 02:17 AM Originally posted by cmost It isn't meaningless. I thought we just went through this in one of the other threads? Typical home / non-techie users don't know how to get and/or apply patches, therefore, the security of a default installation becomes VERY important and relevant. By that logic, it doesn't matter now that XP SP2 automagically fetches and downloads updates automatically. Now, as to why its an issue and why its meaningless: under that same logic, if the world were to run *nix, they'd be getting attacked left and right. Windows is the dominant system and there more computer illiterate people who run it. If you're going to attack something and expect results, you'll attack such a system. That is in no way a reflection of security, it only imposes the false notion of security. Easily remedied, fortunately! Incidently, i've noticed that every time there's a problem with Windows somebody is quick to point out how it's no big deal because it's going to be fixed in the next version of Windows, whose coming is always billed as right up there with the second coming of Christ. This phenomenon has been happening since the days of Windows 1.0. Yet, with Linux, if it isn't perfect the first time then it means Linux is an inferior product and not worth the time. Why do you suppose this is? Easily remedied? Not at all. Because most software is written with the assumption the user is an Administrator, it is in no way an easy change to make. As for problems with Linux, I hear just as many people say the same thing when there's a problem with *nix. It'll be fixed in "blah" patch or, usually, that such a problem doesn't exist and its just FUD made to hurt Linux. Fact of the matter is, there are maybe 2 zero-day attacks that have been made against Windows. The same cannot be said about Linux, and it's much younger. Of course, that issue is of large debate at many security focused organizations. That is, to increase efficiency in reporting exploits in Linux's case and to increase speed at which patches actually get downloaded and installed in Windows' case. Me Webpage | If you always think like an expert, you'll always be a beginner. | "A handful of knowledgeable people is more effective than an army of fools" -Writing Secure Code, 2nd Ed.

i_wolf labhair dom as gaelige Posts: 2097 Joined: 2002-11-19

#35895 Posted on: 04/22/2005 05:26 AM Both have their holes. Generally speaking (I hate using that word) Linux holes are identified and patched much quicker. Microsoft tends to be much slower to get releases out there... I can understand why though... Microsoft must test and retest every patch they release so that they can be sure it doesn't affect something else. A side result however is that a serious hole may take quite a time to fix. On a side note thought some of you here might find this interesting ~ Suse has become the first distribution to be granted level four Common Criteria certification.... which means it is deamed to be secure enough out of the box to be used for command-and-control applications in US government departments. The Common Criteria are international standards for measuring the security of software. It has a scale that runs from 1 - 7. The award was given to SLES 9 running on an IBM eServer. Hung like a donkey. Go like a horse!

Slaughter111 Registered User Posts: 728 Joined: 2002-12-22

#35896 Posted on: 04/25/2005 08:15 AM Originally posted by rmn That's precisely the reason why Llap-Goch was created! RMN ~~~ lol, rmn you are insane Originally posted by i_wolf On a side note thought some of you here might find this interesting ~ Suse has become the first distribution to be granted level four Common Criteria certification.... which means it is deamed to be secure enough out of the box to be used for command-and-control applications in US government departments. The Common Criteria are international standards for measuring the security of software. It has a scale that runs from 1 - 7. The award was given to SLES 9 running on an IBM eServer. now that's nothing to sneeze at...:eek: go, Novell!!! A+ | MCSA | MCITP | CommVault CVSA/CVSE | VMware VCP 4 Heatware: Slaughter111

proffesso Winner of the Internet Posts: 1532 Joined: 2001-06-25

#35897 Posted on: 04/25/2005 10:35 AM wow...so what the hell is lvl7....an unconnected hdd sitting in a safe? chasing tokyo girls

rmn oh my, it's huge! Posts: 6013 Joined: 2002-01-26

#35898 Posted on: 04/25/2005 08:43 PM No, that's level 6. Level 7 is the same, but without the hard drive. RMN ~~~

Vuke69 Bitpimp Posts: 377 Joined: 2001-03-16

#35899 Posted on: 04/26/2005 07:33 PM http://www-1.ibm.com/servers/eserver/zseries/security/ccs_certification.html

2CPU.com » News » April 2005 » Kernel Changes Draw Concern from Open-Source Community