Info : http://www.linuxhq.com/kernel/change...6/3/index.html
Download : http://www.linuxhq.com/kernel/v2.6/index.html
Printable View
I'm already half way there to downloading it :D
Thanks SDK, I never knew about the log file.
Is it a stable version? Anyway i am still trying to compile the 2.6.2....(dont have time to rtfm!:( because of f**king exams)
*Trying* to compile 2.6.2? What seems to be the problem? Compile went very smoothly for me.
Cheers,
cgkanchi
Nice, time to update the server!
That's what I love about linux, it's constant attention towards a better goal.
Windows could learn a lot from that.
good news
-stays with his 2.4.x kernel- No reason I see to switch :)
You mean like their patch releases, software upgrades, and higher distro releases? Wait, Windows does that too.Quote:
Windows could learn a lot from that.
pooh, you forgot one item: bloat
:argue:
devpon, you forgot one counter: configuration of all services, dependencies, kernel configurations, and program resource demands.
All avalaible in windows. Bloat only exists if you do not understand the Windows OS.
I've been running it on a couple of boxes, and my shell server in user-mode Linux for a couple of days. It seems good.
Nice to see a working crypto api and cryptoloop built in at last.
Slarty
Just a thought:
Relyt@linux:/etc/permissions.d> cd /
Relyt@linux:/> su root
Password: Not3nou6h-tyme
linux:/# make moretime
make: *** No rule to make target 'moretime'. Stop whinning!
linux:/ #
:)
True, it does....it also has problems where update X which fixes problem 1 breaks program Y because of whatever. Remember XP's initial start and the UPNP issue? I don't like updating to fix a problem and then find out that the update opens a BIGGER hole and now I have two issues to worry about. That's what I meant, should've disclaimered it better.Quote:
Originally posted here by pooh sun tzu
-stays with his 2.4.x kernel- No reason I see to switch :)
You mean like their patch releases, software upgrades, and higher distro releases? Wait, Windows does that too.
You mean like when nix OSS groups do the same thing? Or a kernel patch is released one day after a previous kernel patch? Or do you honestly think these same events do not take place in nix releases?Quote:
I don't like updating to fix a problem and then find out that the update opens a BIGGER hole and now I have two issues to worry about
Unless you are running Gentoo, or on another bleeding-edge distro, then you won't recognize those patches because the primary distributors simply won't allow them until they are tested stable. Mind you, the coders still release fixes, which still create more problems. So, you could take the Debian approach and simply not install up to date packages, given them a month or two to pan out, and then get the stable build.
Or you can take the Gentoo Linux path, and download all the latest patches right away without waiting to see if they are stable for < /. joke > fun and profit. < /. joke>
The choice is yours, and it's opinion really. But my point is, you need to point that finger right back at *nix, because Windows is far from being the only coding project that releases patches that create new patches later to fix the old patches. (thinks of nmap, beep, kernel 2.4.3, XFree 4.3.x, etc etc)
Oh I agree, there are a lot of similarities between win32 OSes and *nix OSes. However, the vast majority of people use windows, pure and simple. At least with *nix, I have the ability to do something about the code. With windows, I can't. I have to put my full trust and faith in a company that's written code to crash out Netscape (Windows 95), has trojan deals with various companies, and according to a friend of mine who worked as a contractor up in Redmond, have meetings once a month or so where the developers/coders get together and only then do they finally see what their code does. That's probably changed (the last part), and I hope so, because this is the same group that I put my faith in when I hit windowsupdate.microsoft.com so that I can get the latest fix for whatever.
Both sides have their problems. One side allows me much more of a choice to fix something if I wanted to. I'll choose that side, IMHO. And you're right, it's a matter of opinion and I respect yours. Fortunately for me, I haven't had too many issues with kernel problems, however, X is a different story, hehe....
I don't see that as a Bad Thing©, since at least it is getting people connected.Quote:
However, the vast majority of people use windows, pure and simple.
That is correct. Trust is needed. Or, simply understanding how it works. There of course may always be something hidden, but you just have to trust them. Not a big deal. Remember back when the backdoor was placed in the 2.4.1 kernel? Horrible times. I still don't see this as a Bad Thing©, since there are too many 3rd party tools to allow you to trust the Windows OS by monitoring it.Quote:
At least with *nix, I have the ability to do something about the code. With windows, I can't. I have to put my full trust and faith in a company that's written code
And? I don't see how coding ethids or procedures should alter your perception either way. If it works, it works. If it does not work, then you fix it :) And trojan deals? Come on.... show me some documentation before I even believe that.Quote:
to crash out Netscape (Windows 95), has trojan deals with various companies, and according to a friend of mine who worked as a contractor up in Redmond, have meetings once a month or so where the developers/coders get together and only then do they finally see what their code does.
Then you need to understand the Windows OS much more. No, that isn't an insult. It is a literal and friendly suggestion. There is not one single aspect of Windows that can not be modified or changed. There isn't a single thing you can do in linux (looking at source does not count, nor does modifying source code, because individual modules in windows can stil lbe configured without recompiling), that you can't do in Windows.Quote:
Both sides have their problems. One side allows me much more of a choice to fix something if I wanted to. I'll choose that side, IMHO.
I respect your opinion too, but would rather see it be much more in tune and learned in the ways of Windows. The people who call it nonconfigurable simply need to be shown how it is configurable. I'm serious. Name something, and I'll tell you exactally how to do it or configure it in windows.
Hehe, we seem to have quite the convo going. No worries as it's relatively friendly.
Simply "understanding" how something works doesn't mean that I have the ability to do something different. Backdoors and the like all have to be given a trust factor, agreed. I have to trust that MS or RH have a vested interest in their customers.Quote:
That is correct. Trust is needed. Or, simply understanding how it works. There of course may always be something hidden, but you just have to trust them. Not a big deal. Remember back when the backdoor was placed in the 2.4.1 kernel? Horrible times. I still don't see this as a Bad Thing©, since there are too many 3rd party tools to allow you to still not trust the Windows OS.
And how is that "ok"? Regardless of whether it was a developer who thought it cool to write a backdoor in some program or a company sanctioned it "OK" to put hostile code against a competitive product, either way, that is "not a good thing" IMHO. A backdoor in a kernel shows a vulnerability to anyone with that kernel. A program that crashes constantly when a user tries to get onto the internet because some company named Netscape is trying to get into a valid market, well....that shows aggresiveness between two companies and the end user gets a bad view on the product when it's not that company's fault, and it shows bad business practice. Imagine if Ford was able to tell if you had parts that came out of a Chevy and decided to malfunction your engine? Not cool.Quote:
And? I don't see how coding ethids or procedures should alter your perception either way. If it works, it works. If it does not work, then you fix it And trojan deals? Come on.... show me some documentation before I even believe that.
As for the trojan deals, I can name several companies who agreed to contracts where MS would be the only "shop" in there (getting discounted hardware/software/etc) and if they violated that, who knows what would happen. PM me and I'll give you a guy who's going through that right now. After all, if a company has been proven to have written hostile code, why would you NOT believe that they engage in trojan deals?
I've worked on PCs and windows platforms enough to know that the problems I've encountered have resulted from A: end users screwing up (we all live with that one, eh? Hehe...), B: some program doing something wrong/etc, or C: Windows having bad code. And why would modifying source code not work? Above, you said if it works, it works, and if not, you fix it. *nix source allows me to do that. I can't "recompile" explorer.exe to NOT do various things. But, in that same notion, I'll say that I've taken the choice to use other programs like Opera and its inherent email system rather than Outlook and IE. The day when I have people coming to me with their systems in hand saying "My linux box died" and I see it's because of bad coding, and patchwork updating, I'll be the first one on here admitting to the instabilities of *nix.Quote:
Then you need to understand the Windows OS much more. No, that isn't an insult. It is a literal and friendly suggestion. There is not one single aspect of Windows that can not be modified or changed. There isn't a single thing you can do in linux (looking at source does not count, nor does modifying source code, because individual modules in windows can stil lbe configured without recompiling), that you can't do in Windows.
Bottom line for me: *nix gives me a choice. Whether I want to rewrite a section of code or remove entire chunks or add another function, it'll let me. I can't do that with Windows. Sure, I can "modify" certain things but two things apply. 1) It requires that I TRUST MS in that it does what it says it does and nothing more; unfortunately, MS has proven too many times to the world that that's not been the entire truth. 2) Only what MS wants changed or will allow to be changed will ever be changed.
We play our cards however we can. I know what I'd like to stick with and I'll give full credit and say that my XP Professional (legitimately licensed I might add) has been great. With my choice of using other programs and its increased stability, it's been a much more pleasurable experience, especially considering I'm a hardcore gamer and just about every game known to man is written natively in win32 environments. It was a good step, IMHO, from Win98. They just have longer to go, hehe...
I'll respond to your entire statement with this:
Then you have a lot more to learn about Windows. Everything you just said above, can be done. Want to make kernel configurations in windows? Possible. Want to change gui functions? Possible. Want to disable certain API or module calls? Possible.Quote:
They just have longer to go, hehe...
So I am fine with you not trusting MS, that's completely cool with me. But, there is a difference between their Windows OS and Micorsoft as a company. :) Little by little, I'm proving to people on this board, through tests, tutorials, and whatnot, that Windows can do anything you wish. So long as you take the time to learn. Mind you, not in a zealot sense of the word *prove*, as I very much adore my fedora/gentoo box.
Then you need to visit Bugtraq, Bugzilla, and every help forum of every nix distro :)Quote:
The day when I have people coming to me with their systems in hand saying "My linux box died" and I see it's because of bad coding, and patchwork updating, I'll be the first one on here admitting to the instabilities of *nix.
Good conversation, even if this didn't go anywhere. :D If you ever have any questions about modifying and configuring windows, let me know
Haha, true...I want to learn more on Windows, but I fall victim to working in *nix all day as that's my job. One that I love, and one that's sidetracked me to the various naysays of Windows. But, it's my choice. :D
I hear that, only the vice is true :D Cheers on this working out well.
PST, I was just thinking, is it possible to chroot (or similar) a user on a Windows box? If it is, I'd love to know how :).
Cheers,
cgkanchi
cgkanchi:
Part of that would involve doing a PATH command, and rewriting the PATH to include the new root. that would create a root semi-jail, as it eliminated remoce command calls. A simple batch file could do that. I'll certainly look further into changing the actual current dir into the base dir. :) give me a moment, as changing a force on the base dir is the 2nd half of chroot.
edit: Cygwin allows the use of chroot on any directory, so that would cover it in according with a chroot jail. A temp solution, but it works.
I think what I need to know is, what do you want to accomplish with chroot? Since it has so many uses and functions, of course there will not be a command duplicate to emulate that said command. But the process in which you can choose to use chroot, can be emulated.
So, more specifically. What do you want to do with chroot :)
My main use for chroot would be to provide a shell (Terminal Services) to a person, who could do whatever he liked with the chrooted part of the box without touching the actual XP install.
Cheers,
cgkanchi
Ahhh, that makes much more sense :D
Okay, here we go.
I just send you my IP via PM, use that to ssh into me. That's right, ssh into my windows box, on port 22.
user: sshjail
pass: sshjail
You will be in a protected directory that is password protected, and rooted to that main dir. If we also wanted to alter the executiton settings we could also merely have a batchfile autorun on start up. Give that a shot, and then I explain how.
edit: Unplugged my firewall from port 22 just now for you to try. enjoy!
edit2: damn firewall, give me a moment
edit3:
Okay, so he now understands it, I will explain it here.
1. Two different ways to chroot, command line and GUI based.
2. For a command line based chroot on remote, simply install a good old Windows SSH server. (example: http://www.networksimplicity.com/openssh/)
2a. For a gui based chroot on remote, simply install a remote access GUI controller, (example: PCAnywhere)
3. Set proper settings to their home dir, the cmd.exe(or otherwise for PCAnywhere) executable for the terminal, and then proced to make proper User configuration settings. Around that home directory. If you know how to work the user and group policy settings, locking them in that home dir specificed by your ssh server/PCanywhere is a peice of cake.
4. After permissions are set and server is on, run it and login for the test. If you set permissions and policies correctly (just like good old chmod) then they are locked tight in their directory.
5. Finally, if you want to modifiy executable commands, simply have the connection shell run a batch script to redefine PATH and the Enviromental Variables to the appropraite directories you want them to access.
6. A bit longer to do than chrooting someone, but the process get's done just as securley and just as well :)