Slashdot Mirror


Debian Elevates KFreeBSD Port to First-Class Status

Reader tail.man points out this press release from Debian which says that the port of the Debian system to the FreeBSD kernel will be given equal footing alongside Debian's several other release ports, starting with the release of Squeeze. Excerpting from this release: "The kFreeBSD architectures for the AMD64/Intel EM64T and i386 processor architectures are now release architectures. Severe bugs on these architectures will be considered release critical the same way as bugs on other architectures like armel or i386 are. If a particular package does not build or work properly on such an architecture this problem is considered release-critical. Debian's main motivation for the inclusion of the FreeBSD kernel into the official release process is the opportunity to offer to its users a broader choice of kernels and also include a kernel that provides features such as jails, the OpenBSD Packet Filter and support for NDIS drivers in the mainline kernel with full support."

376 comments

  1. OK by Profane+MuthaFucka · · Score: 5, Funny

    But, does it run Linux?

    --
    Fascism trolls keeping me up every night. When I starts a preachin', he HITS ME WITH HIS REICH!
    1. Re:OK by Anonymous Coward · · Score: 0

      Yes, and may run them faster too.

    2. Re:OK by joe_bruin · · Score: 2, Informative

      But, does it run Linux?

      No, it's GNU/FreeBSD. It can, however, emulate Linux system calls and therefore natively run binaries compiled for Linux.

    3. Re:OK by Anonymous Coward · · Score: 0

      Well, according to half the open source community it should just be "FreeBSD", without the "GNU".

    4. Re:OK by dazjorz · · Score: 2, Informative

      No, because in Debian GNU/kFreeBSD, as the name indicates, the userspace is GNU, not BSD. Therefore, GNU/FreeBSD. (in Debian GNU/kFreeBSD, the k stands for "kernel".)

    5. Re:OK by Jurily · · Score: 4, Funny

      GNU/FreeBSD

      Flamewar starting in 5..4..3..

      Also vi, KDE3, Gentoo and K&R.

    6. Re:OK by Anonymous Coward · · Score: 0

      I was referring, by analogy, to the fact that half the community refuses to call Linux distributions GNU/Linux.

    7. Re:OK by Anonymous Coward · · Score: 0, Interesting

      The FreeBSD is the kernels name, the operating system name. Linux, FreeBSD, NetBSD, OpenBSD, DragonBSD are monolithic kernels, what makes all of them operating systems. Not just kernels. Same thing goes with SunOS.

      GNU is taking so much others fame by forcing that GNU software is on more important part on the software system than others. Example, the Xorg, Firefox, KDE or even Amarok are more important for normal users than any GNU software. Even Linux OS is more important than GNU software. Without Linux OS, none of GNU software would work. Not glibc, not bash, not Gnome and so on.
      And you do not need to use GRUB to boot Linux OS. You can use any wanted bootloader to load OS, GNU is not needed at all.

      GNU's should have their OWN operating system working. But Hurd is not ready. They made bad choise to port a Mach microkernel to Hurd, what made Hurd process slow down to become a completed OS even more.

      GNU project has nice things to say and they have got nice software done. But it is just pathetic that they try to steal others fame and honor while waiting that others respect them and honors their actions. Just like GNU would be the #1 pure thing on universum.

      (And those who even tries to say that GNU software is part of Linux kernel (what is the complete operating system, not just kernel like microkernels are) should explain does they call their computer as CPU/GNU or Motherboard/GNU and do they call United States as England/United States? All the times GNU people wants to people to forget the facts that monolithic kernels are the old way to build the operating system.)

    8. Re:OK by incripshin · · Score: 3, Funny

      You son of a bitch. I use vi, KDE, Gentoo, and K&R.

    9. Re:OK by Anonymous Coward · · Score: 0

      Without Linux OS, none of GNU software would work. Not glibc, not bash, not Gnome and so on.

      There's plenty of operating systems that do not have a Linux kernel that run bash and gnome and link against glibc, like FreeBSD.

    10. Re:OK by Korin43 · · Score: 4, Insightful

      I think you're confusing "important part of an operating system" with an operating system. Linux is definitely not an operating system, it's just a common term to refer to Linux-based operating systems (because the average person doesn't care). Just like the FreeBSD kernel alone isn't an operating system. Debian and FreeBSD are operating systems. GNU appears to be a complete operating system (although not finished). I think you can install it from here: http://www.gnu.org/software/hurd/hurd/running/gnu.html.

      Need a simple proof that Linux isn't an operating system? Download the kernel and boot it (oh wait, you can't because GRUB isn't part of the kernel), do some stuff on the command line (bash is not part of the kernel either), maybe update some programs (apt isn't part of the kernel). Oh wait you say, I don't need apt, I can just download the source and compile it myself... but wget and gcc aren't part of the kernel either.

      And don't take this to mean I think we should call it GNU/Linux. People can call their operating system whatever they want. If they wanted every piece of software that uses GNU to be called GNU/Software, it should be in the license. My point is just that kernel != operating system.

    11. Re:OK by bonch · · Score: 0

      I don't refer to the "userspace" when I mention an operation system. Otherwise, I'd also have to call it GNU/X11/GNOME/Linux.

      The GNU prefix issue is just Stallman's bitterness over others stealing his thunder.

    12. Re:OK by wertarbyte · · Score: 1

      So, it should be able to run User Mode Linux?

      --
      Life is just nature's way of keeping meat fresh.
    13. Re:OK by TheRaven64 · · Score: 1

      UML, as I recall, requires some special system calls that are only present in Linux. I believe someone ported it to FreeBSD a while ago, but the patches were never merged into either project. You can, however, run Linux in QEMU or VirtualBox on FreeBSD (be warned that the VirtualBox port is quite immature at the moment and may kill your host kernel).

      --
      I am TheRaven on Soylent News
    14. Re:OK by shutdown+-p+now · · Score: 1

      It's one rare case where "GNU/Something" actually makes sense. The mainline distribution of the OS is known simply as "FreeBSD". This thing, however, is very different from mainline in that it replaces the userspace with GNU. Calling the result just "FreeBSD" would be both confusing and incorrect. "GNU/KFreeBSD" correctly and unambiguously describes the nature of the project in a way that distinguishes the main point of difference - GNU userspace over FreeBSD kernel.

      This is different from "GNU/Linux" issue, because there GNU userspace is mainstream, and therefore implied and understood by everyone when plain "Linux" in used OS context, with no further clarification required.

    15. Re:OK by sortius_nod · · Score: 1

      Agreed, I think we sometimes get caught up on the semantics a bit too much and forget what the point of the whole Linux/GNU/FreeBSD thing is. Software freedom doesn't include restrictions on naming, otherwise it's not truly software freedom is it now?

    16. Re:OK by geniusj · · Score: 1

      FreeBSD has its own libc..

    17. Re:OK by Jurily · · Score: 2, Interesting

      Who the hell modded this incomprehensible gibberish up?

      Parent is confusing the terms "kernel" with "base installation", "OS" with "kernel", "microkernel" with "kernel", "monolithic kernel" with "base installation".

      And to those of you who think glibc, gcc and autotools are not important, I dare you to build a fully Open Source Linux distro without them, or even just replace them on your own box. I have tried to make myself an uclibc-based Gentoo, and I still have nightmares about it.

      Anyway, let's just call it Debian and be done with it. uname -a will fill you in on the rest.

    18. Re:OK by Anonymous Coward · · Score: 0

      It does for FreeBSD projects, yes. As far as I know Gnome (being a GNU project) on FreeBSD still links against glibc, though.

    19. Re:OK by agnosticnixie · · Score: 1

      There's a hell of a lot of people who ran GNU on SunOS, BSD and other Unixen before GNU started noticing that Linux had ported their shit to a unix clone for micros, why are Linux and FreeBSD any less entitled to say fuck off to GNU because they, years later, figured they were worth the time. The GNU OS is the one powered by HURD, all GNU has are a compiler/clib almost everyone uses anyway and a set of utilities which can be torn off linux distros and replaced by other shit anyway relatively trivially.

    20. Re:OK by agnosticnixie · · Score: 2, Funny

      Yggdrasil LGX was the full name of Yggdrasil, for Linux/GNU/X - apparently GNU people liked it enough, except for that pesky thing saying that the Kernel comes first and giving any credit to X.org

    21. Re:OK by RichardJenkins · · Score: 1

      Sure.

      More seriously, this nicely highlights the silliness and growing ambiguity in saying you "run Linux". It's like an Apple user saying "I run XNU"

    22. Re:OK by x2A · · Score: 2, Insightful

      "oh wait, you can't because GRUB isn't part of the kernel"

      I don't boot my kernel with GRUB.

      I think the whole thing's just silly anyway... people who need to be told "it's linux/gnu" because linux isn't the thing and the whole thing so help me god, are people who are less likely to understand the difference anyway, and aren't going to be looking at it going "wow Linus Torvalds wrote all of this by himself? Wow he must be a god... nobody else, just him, I must give all my credit for doing this to him". No. So putting "gnu" in the name isn't going to make them instead go "Oh it's written by this one guy called Linus and some guy who's parents hated him and called him GNU or something... but still, they're gods and get my respect".

      It's just marketting, and one thing that Linux has accomplished that GNU hasn't, is in having a name that people like. You could replace the ENTIRE stack above the Linux kernel with a much nicer all tied together polished system, call it "DogShitOS" and people are still going to called it Linux. Want people to call it your own name? Come up with something like Ubuntu, Debian, people have no problem saying those. But GNU? People just don't like it... you can't force them to.

      Even apple fans like "apple" or "mac", maybe running "leopard" or whatever. But they're just PC's in "pretty" cases, the main difference now is the operating system: OSX. But who says "I like OSX" rather than "I like apple"? *Not* the general public, they like names, not letters.

      RMS (which for some reason I always read as "root me silly") should just get over it.

      --
      The revolution will not be televised... but it will have a page on Wikipedia
    23. Re:OK by x2A · · Score: 1

      Absolutely, that's why I call my Windows installation "Windows/VisualStudio"

      I wonder what toolchain OSX is built with... cuz it's not in the title.

      --
      The revolution will not be televised... but it will have a page on Wikipedia
    24. Re:OK by Anonymous Coward · · Score: 0

      GNU is not a toolchain, but it contains one.
      glibc, coreutils, gtk+, gnome... that's GNU.

    25. Re:OK by Hucko · · Score: 1

      What Josh04 said and if you are going to be pedantic... please recognise that there are more than 4 linux users. ...I'm assuming that you are a different persona to Mr RMS.

      --
      Semi-automatic amateur armchair Australian philosopher; conjecture ready at any moment...
    26. Re:OK by Tacvek · · Score: 1

      Nobody would ever call it GNU/X11/GNOME/Linux, since GNOME is part of the GNU project.
      rms officially has no issue with with it being called GNU/X11/Linux, although he might note that by either alphabetical order or code size, it might be better as GNU/Linux/X11.

      --
      Stylish sheet to fix many problems in Slashdot's D3: https://gist.github.com/801524
    27. Re:OK by Pseudonym · · Score: 1

      Exactly. That's why "Linux" isn't trademarked.

      --
      sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
    28. Re:OK by Anonymous Coward · · Score: 0

      And you have unambiguously described yourself as a pedantic douchebag. Burn in hell and argue semantics with the devil, you fucker.

    29. Re:OK by RCL · · Score: 1

      No it does not.

      [rcl@dellgx260 ~]$ ldd `which gnome-panel` | grep libc
      libc.so.7 => /lib/libc.so.7 (0x29322000)

    30. Re:OK by Anonymous Coward · · Score: 0

      No, it's GNU/FreeBSD. It can, however, emulate Linux system calls and therefore natively run binaries compiled for Linux.

      GNU/whoosh

    31. Re:OK by Anonymous Coward · · Score: 0

      I agree the kernel itself is not the OS, however, you are being too narrow.

      Methods of loading the OS are not part of the OS. By your measure, because I now run C64 on X64, the bits needed to kick off the OS are part of the OS. x64 (emulator, and method to boot C64 OS) is now part of the OS.

      I don't find that follows. Your separation is arbitrary.

      The operating system is the kernel and all the bits needed to produce the Linux ABI (libraries and such). If those interfaces were different, they would be a different OS even with the Linux kernel. Moreover, we include the kernel (and not just the libraries) as those libraries are built for a particular kernel.

      Not convinced? Consider this. I take a Linux kernel. I wrap it in Mac System 7 libraries such that I can only run Mac binaries. Given its 'interface' is purely Mac, you couldn't even tell the Linux kernel was there. We would see it as Mac System 7, not Linux.

    32. Re:OK by alantus · · Score: 1

      I agree completely with you, its just a name thing. What was he smoking when he thought about calling it GNU?

    33. Re:OK by Anonymous Coward · · Score: 0

      I refuse to answer that on the grounds that I might incriminate myself

    34. Re:OK by RichiH · · Score: 1

      > Also vi, KDE3, Gentoo and K&R.

      That you _really_ meant Vim, KDE3, Debian is obvious, but what is K&R?

    35. Re:OK by Anonymous Coward · · Score: 1, Insightful

      Actually, the Linux kernel is the operating system. The operating system is just the software what offers the hardware resources to software and it operates both of them.

      You are doing dum examples there. You are saying that Linux is not a operating system because it needs a GRUB. Okay so you are saying that Intel's i7 processor is not a CPU because it needs a motherboard. The CPU is the motherboard + i7. But then you again notice something, you need electricity. So the CPU is the electricity + motherboard + i7. And then you notice that no, it does not yet work at all. You need other parts as well than those. Like RAM. And when you add these things, you actually get something else, bigger than just a CPU.

      Same thing is with operating systems. You first start with someting, like I/O. Filesystems, Networking and all device drivers what are needed to communicate with hardware devices. You build up a operating system from many smaller features. You do not call a device driver as kernel or operating system. It is just a smaller part of the operating system. Then when you have got the operating system working, you can get all other software working top of it. LIke glibc (needs a filesystem, I/O, scheduler and so on) and bash. Those both software needs already a existing operating system to work on the computer.

      The operating system needs a bootloader to start it. After the OS is started, it starts running all other process on the system. First software what is outside of the OS on Unix systems is INIT. INIT is motherprocess of all other processes. All other software. You can not see the processes of the OS at all.

      http://www.topology.org/human/?a=/linux/lingl.html

      You are speaking about Linux like it would be a microkernel. Microkernel is just one part of the operating system. The operating system is completed by microkernel + modules (sometimes called OS servers).
      The operating system is the software what is running on kernel mode or as supervisor mode.

      Example, read the page 1-3 of that. And if you dont still understand the technology, refer this small short explenation http://www.usenix.org/publications/login/2006-04/openpdfs/herder.pdf

      http://www.amazon.com/reader/0130313580?_encoding=UTF8&ref_=sib_dp_pt#reader

      On top of the operating system is the rest of the system software. Here we find the command interpreter (shell), compilers, editors and similar application independent programs. It is important to realize that these programs are definitely not part of the operating system, even though they are typically supplied by the computer manufacturer. This is crucial, but subtle, point. The operating system is that portion of the software that runs in kernel mode or supervisor mode.

      It is protected from user tampering by the hardware (ignoring for the moment some of the older microprocessors that do not have hardware protections at all). Compilers and editors run in user mode. if a user does not like a particular compiler, he is free to write his own if he so chooses; he is not free to write his own disk interrupt handler, which is part of the operating system and is normally protected by hardware against attempts by user to modify it.

      The shell or libc can not work at all without operating system. The OS does not need a GRUB anymore. The GRUB only thing is to start a OS. After that it release all the controls to the OS. And if following your logic. The GRUB is part of BIOS/EFI. Because without such, GRUB would not work at all either. And by your logic, the BIOS is part of the electricity lines, because without electricity, the BIOS would not work at all because computer does not work.

      http://www.gridbus.org/~raj/microkernel/chap2.pdf

      The operating system structures are so simple. They are not complex "black magic" what computer science can not explain. Only thing what makes problems is people trying to spread GNU propaganda "Linux is just a kernel" like it would be a microkernel li

    36. Re:OK by Anonymous Coward · · Score: 0

      it's just a common term to refer to Linux-based operating systems

      Linux means the operating system. It has always meant only the operating system. But because of GNU. People started to mean with Linux the GNU/Linux as well, because they ate the propaganda of GNU. So Linux lost it's meaning of the operating system and started to mean GNU/Linux as well, what RMS wanted to happend.
      Then there started to come more avarage joes and they believed that Linux is what they see on the screen. Like GNOME desktop and Firefox. It is same way totally false believe about Linux.

      The Linux only means Linux. Nothing else. When people speaks about Linux kernel, they speak about Linux. When people say Linux, they should only mean Linux kernel. The proplem here, is that Linux kernel is the operating system. They are exactly the same thing. The Linux is not a microkernel, how much you are refferring it to be. Linux is a operating system same way like BSD's are. They are all monolithic kernels and that is the whole operating system running alone in the kernel space. None of the OS parts is located to user space or outside of the kernel (structure).

      You can even check the OS name with uname command. But I believe that you use the GNU version of uname what is edited and modified so it does not print truth, but GNU own propaganda.

      Check the GNU version fo uname http://www.gnu.org/software/coreutils/manual/html_node/uname-invocation.html and compare it to standard Uname command. http://www.opengroup.org/onlinepubs/009695399/utilities/uname.html
      If you are lousy to discover something, then read this http://www.mail-archive.com/linux-users@it.canterbury.ac.nz/msg21956.html

    37. Re:OK by Anonymous Coward · · Score: 1, Interesting

      The kernel can boot itself just fine, ever heard of XIP? Likewise for integrated initramfs taking the place of userspace. Before spouting tired rhetoric, get a clue, thanks.

    38. The ONE TRUE indentation style, of course.

    39. Re:OK by Deslack · · Score: 0

      Linux is trademarked.

      Linux Mark Institute, exclusive licensor of the Linux trademark on behalf of its owner, Linus Torvalds.

      http://www.linuxmark.org/

      --
      .sigs are useless; it doesn't protect you from imposters.
    40. Re:OK by jonadab · · Score: 1

      > You son of a bitch. I use vi, KDE, Gentoo, and K&R.

      Yeah, well I use Emacs, Sawfish, Debian, and the Camel book. So there.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    41. Re:OK by kholburn · · Score: 1

      But unfortunately for you, it is: http://www.linuxmark.org/

    42. Re:OK by Per+Wigren · · Score: 2, Funny

      (in Debian GNU/kFreeBSD, the k stands for "kernel".)

      ...and it's pronounced as "nuke free beast".

      --
      My other account has a 3-digit UID.
    43. Re:OK by morgan_greywolf · · Score: 1

      But it can.

      Which is what Debian/kFreeBSD is: Take the userland of Debian, glibc, gcc, binutils, etc., and instead of a Linux kernel, use a FreeBSD kernel.

      BTW--there's nothing to stop the Debian folks from releasing their own fork of the FreeBSD kernel under GPL v3....

      Also, it might be interesting to see an Ubuntu based on kFreeBSD.... hmmm..... kFubuntu?

    44. Re:OK by mehemiah · · Score: 1

      I suggest you no longer waste your time explaining your jokes to those with wet blankets with sense of humor or no wits in general.

    45. Re:OK by ClosedSource · · Score: 1

      "And to those of you who think glibc, gcc and autotools are not important, I dare you to build a fully Open Source Linux distro without them, or even just replace them on your own box."

      But building isn't the issue. There's a long history of cross-compiling software on completely different platforms than the target without any of the compiling platforms run-time code ending up on the target.

    46. Re:OK by ClosedSource · · Score: 1

      Besides it's not as if GNU userland was an original work. It would be more accurate to call it AT&T/Linux if giving credit was the goal.

    47. Re:OK by Korin43 · · Score: 1

      You seem to be making the same argument I am (the kernel is not the OS). If you took Linux kernel + System7 userland, you don't get Linux, you get an operating system that is the combination of both of them. Same as when you take Linux + GNU or kFreeBSD + GNU or Linux + BSD userland. Linux fanboys want to say that the kernel defines the operating system, and GNU fanboys want to say the userland defines the operating system. I say it's the combination. So the Linux kernel is not an operating system. GNU without a kernel isn't an operating system, but Debian, Fedora, FreeBSD and GNU+HURD are all operating systems. When I say "Linux is not an operating system", I mean the kernel, but because no operating system is officially named "Linux", there's just a group of them that are commonly called Linuxes.

    48. Re:OK by Anonymous Coward · · Score: 0

      I think you're confusing "important part of an operating system" with an operating system. Linux is definitely not an operating system, it's just a common term to refer to Linux-based operating systems (because the average person doesn't care). Just like the FreeBSD kernel alone isn't an operating system.

      I was going to point out the fundamental error you are making...

      And don't take this to mean I think we should call it GNU/Linux. People can call their operating system whatever they want

      ...but you did it yourself! What you've overlooked is that a name can be used for both an operating system and the kernel of that operating system, if they people who have the naming rights decide to do so. The earlier builders of operating systems that comprised a Linux kernel plus many GNU utilities plus other open source software decided to name the OS after the kernel. They are the ones who put together the complete package, so they got to name it. Hence, it is Linux.

      Some people go the other way, and name their kernel after the operating system. Or rather, they don't give their kernel a name, and it gets called the Foo kernel, where Foo is the name of the OS.

    49. Re:OK by DragonWriter · · Score: 1

      Linux is definitely not an operating system, it's just a common term to refer to Linux-based operating systems (because the average person doesn't care).

      Actually, I think its a common term to refer to operating systems that use the Linux kernel collectively, because more specific designations (either like "Ubuntu" or like "GNU/Linux") don't generally work for that purpose, though sometimes some of them may be appropriate fro some subset of Linux-based operating systems.

      For instance, if I release software that is designed to run on a Linux based operating system with specific identified dependencies, that does not rely on features of a particular distribution or tools or libraries other than those in the listed dependencies, it makes more sense to describe it as being for Linux, with the specific required dependencies, than to say it is for "GNU/Linux" (since, insofar as it requires anything from GNU, that's already covered by the listed dependencies), or for "Ubuntu" (since it doesn't require anything specific to Ubuntu) or for anything else.

    50. Re:OK by DragonWriter · · Score: 1

      GNU is not a toolchain, but it contains one.

      Right. GNU is an operating system. It contains many things, including a kernel (Hurd) which is not the kernel which is at the heart of any Linux distribution (the Linux kernel.)

      Now, because some of the components of GNU are open source and useful, some of them -- including lots of other stuff not from GNU maintained by people other than Linus Torvalds -- are often packaged together with the Linux kernel to make a complete OS distribution. Usually, such a distribution has a name (like "Ubuntu" or "Fedora") and often the distributor tags on "Linux" because identifying it as such helps people understand what kind of operating system it is.

      Certain people want those distros to be called "GNU/Linux" because they include GNU components, but rarely argue that distributions that include components other than the Linux kernel besides GNU components should also have those other components honored alongside GNU in every discussion.

    51. Re:OK by Korin43 · · Score: 1

      What you've overlooked is that a name can be used for both an operating system and the kernel of that operating system, if they people who have the naming rights decide to do so.

      They can, but didn't.

      The earlier builders of operating systems that comprised a Linux kernel plus many GNU utilities plus other open source software decided to name the OS after the kernel. They are the ones who put together the complete package, so they got to name it. Hence, it is Linux.

      Wrong. The operating systems are NOT named Linux. They are named Ubuntu, Fedora, Gentoo, Debian GNU/Linux (yes this one has Linux IN its name, I'm hoping you understand the distinction between Linux being part of the name and not being the name). Just read their pages:

      • openSUSE is a free and Linux-based operating system
      • Ubuntu is a community developed, Linux-based operating system
      • Fedora is a Linux-based operating system
      • Debian is a free operating system (OS) for your computer -- note: Debian is somewhat inconsistent, because they refer to the "GNU/Linux" operating system, but they still never call Linux an operating system

      Enterprise Linuxes tend to be less clear about it (Red Hat Enterprise Linux, SUSE Linux Enterprise), but the name of the operating system is still not just Linux.

    52. Re:OK by True+Grit · · Score: 1

      The ONE TRUE indentation style, of course.

      Please Philip, only one flamewar at a time, it gets really hard to follow the action with two or more happening simultaneously... :)

  2. Cool by Faw · · Score: 5, Interesting

    First apt based distro with ZFS? Something worthy of a post about...

    I know about Nexenta, but FreeBSD has more drivers than OpenSolaris, right?

    1. Re:Cool by Darkness404 · · Score: 1, Insightful

      There has been a lot of hype about ZFS but what use is it in a desktop system? And honestly, while APT is great for desktop systems, I really wouldn't use it much on a server. So unless there is some amazing benefit for the average user with ZFS why even have this port as a main system?

      --
      Taxation is legalized theft, no more, no less.
    2. Re:Cool by spun · · Score: 4, Insightful

      First apt based distro with ZFS? Something worthy of a post about...

      I know about Nexenta, but FreeBSD has more drivers than OpenSolaris, right?

      You seem to be asking some interesting questions, but fail to do so in a timely fashion.

      --
      - None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
    3. Re:Cool by characterZer0 · · Score: 5, Interesting

      Why would you not want to use APT on a server? What part of automatic dependency handling, automatic unneeded package pruning, easy security update application, and secure package retrieval do you not want on your servers?

      --
      Go green: turn off your refrigerator.
    4. Re:Cool by Anonymous Coward · · Score: 0

      You don't like having snapshots of your files, just in case something happens to them?

    5. Re:Cool by PCM2 · · Score: 1

      Care to explain why you wouldn't use apt on a server? You don't like security patches?

      --
      Breakfast served all day!
    6. Re:Cool by Timothy+Brownawell · · Score: 3, Insightful

      Why would you not want to use APT on a server? What part of automatic dependency handling, automatic unneeded package pruning, easy security update application, and secure package retrieval do you not want on your servers?

      Possibly the "automatic unneeded package pruning". It could be dangerous if your custom apps don't specify their dependencies correctly (say, they rely on something that had been automatically installed by one of their other dependencies).

    7. Re:Cool by ducomputergeek · · Score: 1, Funny

      Because BSD has had this since the 1990's, it's call Ports....

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
    8. Re:Cool by gehrehmee · · Score: 1

      If your custom apps required you to install a package, it'll already be listed as manually installed, so it'll never be automatically uninstalled.

      --
      "You know, Hobbes, some days even my lucky rocketship underpants don't help" -- Calvin
    9. Re:Cool by idontgno · · Score: 1, Flamebait

      Possibly the "automatic unneeded package pruning". It could be dangerous if your custom apps don't specify their dependencies correctly (say, they rely on something that had been automatically installed by one of their other dependencies).

      That's OK, that's the kind of thing which is easily caught in non-production test server evaluation of the config change.

      You do non-production evaluation of config changes, don't you?

      C'mon, no professional just pokes "apt-get update" into the root shell on a live production server. That's just asking for hilarity, fail, and unemployment.

      --
      Welcome to the Panopticon. Used to be a prison, now it's your home.
    10. Re:Cool by shish · · Score: 1

      First apt based distro with ZFS?

      Do we need ZFS any more, now that btrfs is nearly here, based on the same ideas but with zfs's design problems known about and worked around?

      --
      I mod down anyone who says "I will be modded down for this", regardless of the rest of their comment
    11. Re:Cool by Timothy+Brownawell · · Score: 4, Informative

      If your custom apps required you to install a package, it'll already be listed as manually installed, so it'll never be automatically uninstalled.

      The idea is that $app depends on $foo and $bar. But because $foo also depends on $bar, someone was able to goof up and only document that $app depends on $foo. So when $foo gets updated and drops its dependency on $bar, $bar goes away (due to being automatically installed) and $app stops working.

    12. Re:Cool by Anonymous Coward · · Score: 0

      I thought open source apps weren't $app, I thought they were 0$app? And if it costs money to fubar my system then I would hope that 0$app would drop its dependency on $fu and $bar and go with 0$free or something...

    13. Re:Cool by idontgno · · Score: 0

      Come to think of it, "apg-get update" is mostly harmless, but I still wouldn't do it on a production server until the entire "upgrade" process is beaten thoroughly down on a test server.

      IMHO, you shouldn't do anything with apt on a production server until you're sure you're not shooting yourself in the foot.

      --
      Welcome to the Panopticon. Used to be a prison, now it's your home.
    14. Re:Cool by GenP · · Score: 2, Interesting

      Sweet, I didn't know they added raid5/6 support to btrfs!

    15. Re:Cool by Chris+Mattern · · Score: 4, Informative

      If your custom apps required you to install a package, it'll already be listed as manually installed, so it'll never be automatically uninstalled.

      Not if the required package was already installed because a third package that required it and correctly specified it was installed. Uninstall that package, which seems to be utterly unrelated to your custom app, and BOOM, custom app breaks.

    16. Re:Cool by joe_bruin · · Score: 1

      For units with bitlength a multiple of 4, (0x2B | ~0x2B) == 0xFFFFF...

      x | ~x = 0xFF..FF, which is equal to -1 in the 2's complement signed version of the resulting type. Why not (0x2B | ~0x2B) == -1

      example:
          main() { printf("%d\n", (0x2b | ~0x2b)); }

    17. Re:Cool by Chris+Mattern · · Score: 2, Informative

      That's a case for having apt on the test server, not the production server.

      Come to think of it, you'd have apt on the production server solely because you don't want it to look different from the test server, but you'd still never use it, instead simply copying the package changes from the test server.

    18. Re:Cool by fuzzyfuzzyfungus · · Score: 1

      I just make sure to have the money ready when Vinny and Guido come knocking. Nothing bad has happened yet.

    19. Re:Cool by Timothy+Brownawell · · Score: 1

      C'mon, no professional just pokes "apt-get update" into the root shell on a live production server. That's just asking for hilarity, fail, and unemployment.

      Because "professionals" are perfect and have unlimited time and resources, right?

      A couple days ago, I had something failing in test that worked in production... turns out someone had put an emergency fix for this in the production system, and because it was an "emergency" they hadn't bothered to update the test environment first (or at all, actually). Shit happens everywhere, so it's a bad idea to rely on shit not happening.

    20. Re:Cool by Chris+Mattern · · Score: 4, Insightful

      "nearly here". No other words can strike such fear into the heart of a production system sysadmin. How about something that's seen production use for years, instead?

    21. Re:Cool by flydude18 · · Score: 2, Insightful

      And new installations of $app don't work either because users are never told they need $bar. This is a fairly obvious bug that gets noticed, and an update for $app adds the dependency to $bar.

      Sure, I'm assuming that someone will fix it, but you're assuming that someone will goof up in the first place. Seems fair.

    22. Re:Cool by Anonymous Coward · · Score: 0

      That's not how we do it. We apt on test for testing and then apt on prod.

      Copying packages/files and manual installs, etc. adds a point of failure. If apt works on test, then it will work on prod.

    23. Re:Cool by petermgreen · · Score: 2, Informative

      Which is a good reason to either use apt-get which defaults to the sane behaviour of tracking packages it thinks are unused but not removing them until it's explicitly told to or reconfiguring aptitude to do the sane thing.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    24. Re:Cool by Jurily · · Score: 4, Insightful

      Do we need Linux any more, now that HURD is nearly here, based on the same ideas but with Linux's design problems known about and worked around?

    25. Re:Cool by Sancho · · Score: 4, Insightful

      Wow, you got modded Troll and Timothy got modded Offtopic for describing a legitimate concern. Man, this place has gone downhill.

      The correct rebuttal to your statement is that you don't mess with things on production machines. You don't uninstall that third package. If you want to make changes like that, you do it on your test machine first. Timothy was concerned with packages dropping dependencies, but that shouldn't happen within -stable.

    26. Re:Cool by rbanffy · · Score: 1

      - Checksums to prevent/warn of silent corruption
      - Lots of options for redundancy
      - Zero-cost snapshots (Something effectively better than Apple's time-machine)

    27. Re:Cool by oldhack · · Score: 4, Funny

      God damn foobar apps.

      --
      Fuck systemd. Fuck Redhat. Fuck Soylent, too. Wait, scratch the last one.
    28. Re:Cool by JoeCere65 · · Score: 1

      And don't forget about near-daily tzdata updates!

    29. Re:Cool by Anonymous Coward · · Score: 0

      HURD isn't anywhere near nearly there, though. It's currently a large piece of shit, and will remain so for a very long time. brtfs is actually going to get decent soon.

    30. Re:Cool by tayhimself · · Score: 1

      I think this is a little bit of an exaggeration. Or I am very mistaken.
      I update nightly on test servers using aptitude safe-update and weekly on the production ones but I don't check the logs it's all automated.
      It just depends on the situation, some people might be better served with weekly or monthly cron based updates. FYI, this is Ubuntu 8.04 server with Xen.

    31. Re:Cool by gehrehmee · · Score: 1

      Quite, I missed that case.

      --
      "You know, Hobbes, some days even my lucky rocketship underpants don't help" -- Calvin
    32. Re:Cool by MichaelSmith · · Score: 1

      The way Linux is going at the moment we pretty much need a microkernel in the next major version (3, I assume). Linus pretty much stated that a couple of weeks ago.

    33. Re:Cool by bfields · · Score: 3, Insightful

      Sure.

      Options are to keep the autoremoval turned off, or build simple custom packages for your third-party aps with the proper dependencies.

      Or you could do without any package management entirely. That doesn't strike me as likely to make anybody safer....

    34. Re:Cool by mdf356 · · Score: 1

      Why not (0x2B | ~0x2B) == -1

      Okay, it's completely off-topic, but:

      1) 2's complement is the norm but not guaranteed
      2) never do bitwise arithmetic on a signed variable; it's asking for trouble

      --
      Terrorist, bomb, al Qaeda, nuclear, yellowcake, kill, assassinate. Carnivore is dead... long live Echelon.
    35. Re:Cool by Anonymous Coward · · Score: 0

      Linux will never have a microkernel. Microkernels suck. They look nice on paper but in actual use they're truly evil.

    36. Re:Cool by Antique+Geekmeister · · Score: 2, Insightful

      Would everyone who actually manages to run their servers with only "-stable" please raise their hands? One, two... no wait, that person is scratching their head.

      I'm afraid that "running servers with only -stable" is like "never typing 'rm -rf' as root". It would be nice if we could do so, but far too often we need to get work done and risk a "non-stable" package to get particular critical features, and this approach breaks down very quickly.

    37. Re:Cool by Anonymous Coward · · Score: 2, Informative

      Sweet, I didn't know they added raid5/6 support to btrfs!

      ZFS also has triple parity ("RAID 7"?), as well as the ability to shim in SSDs transparently. There's also 'zfs send/recv'.

      How long, and how many petabytes of data has been in production on Btrfs?

    38. Re:Cool by CarpetShark · · Score: 1

      It could be dangerous if your custom apps don't specify their dependencies correctly

      Which virtually never happens on Debian, which is why it's so nice to have so many packages in the distro, all properly catalogued.

      By the way, there are ways to hold packages too, if you never want them to be removed or upgraded, or to make lists of packages required, even for your own scripts.

    39. Re:Cool by CarpetShark · · Score: 1

      Sweet, I didn't know they added raid5/6 support to btrfs!

      If ZFS has raid5, it's more of a chimera than I thought. RAID belongs below the FS, not in the FS.

    40. Re:Cool by rivaldufus · · Score: 1

      If you ever see ZFS snapshots in action, you'd want ZFS on your desktop, too. It's trivial to create a snapshot. No need to pre-allocate space like on LVM.
      Snapshots don't seem to impact performance, either.

    41. Re:Cool by ToasterMonkey · · Score: 5, Interesting

      There has been a lot of hype about ZFS but what use is it in a desktop system? And honestly, while APT is great for desktop systems, I really wouldn't use it much on a server. So unless there is some amazing benefit for the average user with ZFS why even have this port as a main system?

      You must be kidding. You can snapshot your whole root, or your home directory, or anything and automate it for backups. There is even integration with GNOME's Nautilus to browse ZFS snapshots at the file level. You can create new filesystems and snapshots on the fly, compress them, enforce quotas, export via NFS, send snaps to a remote system or dump as flat file, etc, etc, etc.. if you don't have enough disks to use single or dual parity RAID-Z, you can even have ZFS record multiple copies of each block. ALL of those have uses on desktop or workstation systems.

      I'm not sure if freeBSD supports ZFS root or not, but Solaris does if you want a taste. If anything, ZFS-root is under-hyped.
      When troubleshooting updates, instead of booting into your old kernel with a trashed userland, you can boot into an old BOOT ENVIRONMENT. As easy as picking a different grub entry.

      From OpenSolaris (other ZFS-root capable systems may use different commands, output BUTCHERED for junk filter)
      beadm list
      BE Active Mountpoint Space Policy Created
      opensolaris 25.05M static 2009 04 01 2033
      snv_111b 111.89M static 2009 06 03 1846
      snv_121 38.18M static 2009 08 31 1617
      snv_122 42.10M static 2009 09 14 1522
      snv_124 NR / 19.09G static 2009 10 01 2354

      Those are whole root file system snapshots I can boot, consuming a piddly ~50MB each.

      Junkfilter is retarded.. there is a moderation system for a reason, and this is a technical forum.

    42. Re:Cool by Bill,+Shooter+of+Bul · · Score: 1

      Shit does happen everywhere, but that shouldn't be your policy. If your redundant set of routers turns into baseballs, that can't really be prevented and that's sort of okay, because electrical equipment usually don't turn into sporting equipment and having it happen to both is a pretty odd day. But having a web server compromised because every service in the apt repository was installed and running with root users remote access enabled and passwords set to 'password', that should be prevented. And if it does happen, "Shit Happens" isn't the appropriate response.

      --
      Well.. maybe. Or Maybe not. But Definitely not sort of.
    43. Re:Cool by rivaldufus · · Score: 2, Interesting

      RaidZ3 is nearly upon us...
      Most slashdot folks will prefer btrfs just because it's GPL and native to Linux... whether or not it's better.

    44. Re:Cool by Anonymous Coward · · Score: 0

      I'm pretty sure it automagically checks for that case when they build $app.

    45. Re:Cool by agnosticnixie · · Score: 5, Funny

      HURD is, and will always remain, the OS of the future.

    46. Re:Cool by ToasterMonkey · · Score: 1

      If ZFS has raid5, it's more of a chimera than I thought. RAID belongs below the FS, not in the FS.

      WHY does software RAID belong underneath the FS? Because HW RAID does and that's what we're emulating? That's a very silly reason.

      On the topic of merging layers no longer of use to us, what is your opinion on Converged Ethernet *cough*Fibre Channel*cough*?

    47. Re:Cool by stevey · · Score: 3, Insightful

      Raises hand.

      The only time when I've worried about running (Debian) stable software on production servers has been towards the end of a releases' life.

      In that case signatures for things like clamav were often useless. However right now I'm running entirely stable software and don't expect that to change for the forseeable future.

      A lot of people seem to want the latest and greatest releases of software for no appreciable reason. (If they were hitting specific bugs I'd understand ..)

    48. Re:Cool by Chris+Mattern · · Score: 1

      If apt works on test, then it will work on prod.

      That would be true, except for one little difference you can't fix: You're not running apt on prod at the same time you ran it on test. Unless you're working from a private package repository that you froze before running apt on test, you can't say prod will retrieve the same package versions when you run apt on it. You can certainly do it that way, and in a large installation it has a lot to recommend it, but if you're not that big, it's just simpler to apt-get test from the public repositories and then copy the packages to prod when you're ready to promote.

    49. Re:Cool by Grishnakh · · Score: 2, Informative

      You might want to read up on ZFS. It's not just a simple filesystem; it also includes LVM and RAID features, all in one. I'm not a professional sysadm, but my understanding is that, unlike Linux where changing a RAID5 array to have more disks or more space requires messing with the logical volumes using LVM, then messing with the RAID arrays with mdadm, then redoing the filesystems with the e[34]fstools, whereas doing this in ZFS just involves a few quick commands without having to unmount the filesystem.

      In fact, a quick Google search shows that resizing an ext3 partition is a bit of a pain, involving converting it to ext2 (removing the journal), running resize2fs, and converting it back. Obviously that's not good if the power cuts out while you're doing this. With ZFS, you should be able to resize arrays, add disks, etc., without even interrupting service.

      If btrfs can do this too, then it sounds great to me. I wish they'd pick a better name though.

    50. Re:Cool by x2A · · Score: 1

      foo/bar/gnu please

      --
      The revolution will not be televised... but it will have a page on Wikipedia
    51. Re:Cool by CarpetShark · · Score: 1

      You might want to read up on ZFS. It's not just a simple filesystem; it also includes LVM and RAID features, all in one.

      This is my point.

    52. Re:Cool by domatic · · Score: 3, Informative

      There are more and less correct ways to go about running "non-stable" packages. Just jamming in a foreign binary deb and it's deps is generally the wrong way to do it. The right way to do it is to basically do what backports.org does....and check if backports has it first. If you're not running a Debian derivative then research the equivalent for your distro or BSD flavor:

      1. add an apt-src line from the "non-stable" distro. I've even done this when Ubuntu had the latest SpamAssassin and I wanted a late model debianized SpamAssassin. Note well I said "apt-src" and NOT an "apt" line. apt-get update, yadda yadda

      2. mkdir package-of-interest; cd package-of-interest; apt-get build-dep package-of-interest. If you're lucky this pulls in everything needed to build the package. If not you'll have to recurse with apt-get build-dep package-of-interest-newer-dependency; apt-get source package-of-interest-newer-dependency and come back to step 2 until you have everything you need to build your "non-stable" package.

      3. apt-get source package-of-interest (or -newer-dependency)

      4. cd package-of-interest-version; fakeroot dpkg-buildpackage -b. If all goes well a backport of the newer software that is built against your nice stable distro's libraries will be built one directory level up.

      5. If you had to build a newer library your package needs install it then go back to #2. rinse and repeat until you have your newer software. Myself, I'll build one or two before giving it up as a bad job. Generally I succeed at this. Bread and butter stuff like Clamav and SpamAssassin that is well behaved tends to work well with this procedure.

      The point of this is to avoid having to replace heeby jeeby inducing things like glibc, libstdc++, or even Perl since half your system scripts depend on such things. You avoid replacing packages that are contributing to the stability of the rest of the system and that are still getting critical bug and security updates. It keeps the "foreign" footprint to a middle and gives you a "foreign" package that is more or less compliant with the rest of the system.

    53. Re:Cool by Ant+P. · · Score: 1

      How about because rewriting the same functionality as a special case for each and every FS is retarded?

    54. Re:Cool by DragonDru · · Score: 1

      *Hand Up*

      Everything being dependent on application and such, for my applications running Debian stable only works just fine.

      --
      20 characters max for the password? How will I use my favorite poems as passwords?
    55. Re:Cool by DragonDru · · Score: 1

      Then package $app as a deb and list $bar as a dependency. Then removing $foo, or upgrading it, $bar stays in place. Plus one could install $app repeatedly and easily.
      I know it is intimidating, but it honestly is not that hard.

      --
      20 characters max for the password? How will I use my favorite poems as passwords?
    56. Re:Cool by synthespian · · Score: 1

      Not it's not "harmless". Read the research on package management.

      Debian has been documented to uninstall the kernel (yes, the odd event, but it happened).

      Start here and then use THE GOOG for more research (BTW, this doesn't reference the paper with the find I mentioned above - might wanna search /. and dig up other posts of mine):

      http://www.mancoosi.org/edos/manager.html

      Now for the fun parts: "This shows that the solution set is order dependent, which is not stated in the documentation, and is quite surprising for a user. (...) " 99.9999% of Debian "developers" do not know how their system works, and it undocumented.

      The authors continue, with a trace in which apt-get wants to zap the users x-windows-system and x-windows-system-core.

      I paste the results for your reading pleasure:

      "Another point we want to stress is the extreme user-unfriendliness of the APT tool in the rare occasions when upgrading or installing one package ends up into a major overhaul of the user installation⦠Here follows a real-world example recorded by one of the authors of this report during his daily running of his beloved Debian-based machine.


      sudo apt-get install debhelper
      Reading Package Lists... Done
      Building Dependency Tree... Done
      The following extra packages will be installed:
      armagetron armagetron-common autoconf bonobo-activation codebreaker
      debconf debconf-i18n debconf-utils dialog esound-common fb-music-high
      fontconfig frozen-bubble-data grepmail gv intltool-debian
      libaiksaurus-data libaiksaurus0c102 libatk1.0-0 libatk1.0-dev
      libbonobo-activation4 libbonobo2-0 libbonobo2-common libdb3
      libdbd-mysql-perl libdbi-perl libeel2-data libesd0
      libfilehandle-unget-perl libfontconfig1 libforms1 libfreetype6
      libfreetype6-dev libgcc1 libgcrypt1 libgdbm3 libgladexml-perl
      libglib2.0-0 libglib2.0-dev libgnome-perl libgnutls7 libgsf-1
      libgtk-imlib-perl libgtk-perl libgtk1.2 libgtk1.2-common libgtk1.2-dbg
      libhtml-parser-perl libice-dev libice6 libidl0 liblinc1
      liblocale-gettext-perl liblzo1 libmagick5.5.7
      libmail-mbox-messageparser-perl libmysqlclient12 libncurses5
      libncurses5-dev libncursesw5 libnet-daemon-perl libnet-perl libnewt0.51
      libogg-dev libogg0 liborbit2 libpaper1 libplrpc-perl libpng12-0
      libpopt-dev libpopt0 libsdl-console libsdl-gfx1.2 libsdl-image1.2
      libsdl-ttf1.2 libsdl-ttf2.0-0 libsdl1.2debian libsdl1.2debian-oss
      libsm-dev libsm6 libsmpeg0 libssl0.9.7 libstartup-notification0
      libstdc++5 libt1-5 libtext-charwidth-perl libtext-iconv-perl
      libtext-wrapi18n-perl libtiff-tools libwmf0.2-7 libwww-perl libx11-6
      libx11-dev libxaw7 libxaw7-dev libxcursor1 libxext-dev libxext6
      libxft1 libxft2 libxi-dev libxi6 libxml-parser-perl libxml2 libxmu-dev
      libxmu6 libxmuu-dev libxmuu1 libxp-dev libxp6 libxpm-dev libxpm4
      libxrandr-dev libxrandr2 libxrender-dev libxrender1 libxt-dev libxt6
      libxtrap-dev libxtrap6 libxtst-dev libxtst6 libxv-dev libxv1 lyx
      lyx-common lyx-xforms perl perl-base perl-modules perlmagick pkg-config
      pm-dev po-debconf render-dev tcl8.4 tcl8.4-dev tktable transfig ucf
      whiptail x-dev xaw3dg xbase-clients xfig xfree86-common xlibmesa-dri
      xlibmesa-gl xlibmesa-gl-dev xlibosmesa-dev xlibosmesa4 xlibs xlibs-data
      xpdf-common

      The following packages will be REMOVED:
      autoconf2.13 frozen-bubble frozen-bubble-lib gconf2 gnomemeeting
      itk3.1-dev libbonoboui2-0 libbonoboui2-common libdigest-md5-perl

      --
      Main difference between the BSD license and the GPL license: one is from California and the other is from Massachusetts
    57. Re:Cool by DragonDru · · Score: 2, Informative

      ZFS is intended to prevent data corruption of *any* kind. Its use of double and shortly (now?) triple parity is just the start of it. One can also store up to 3 independent copies of the same data transparently. (One a mirrored array with two drives, this means 6 copies of the same file.)

      Having one system that provides the same interface and same capabilities across many sets of hardware is very handy.
      In action, ZFS is something to behold.

      --
      20 characters max for the password? How will I use my favorite poems as passwords?
    58. Re:Cool by Wizzu · · Score: 1

      *raises hand*

      I run three servers with Debian stable. One has some backports packages (that's my home server), but the other two are plain stable (remote hobby servers). One of those used to have PHP5 installed from some non-stable repository, but with Lenny I could get rid of the extra apt config line.

      Of course, it depends what you want your server to do. For me, it's as simple as eg. Apache, MySQL and PHP. You could argue that no "real" server can be that simple, but .. well, it depends on your needs.

      I don't think what I do is really that rare.

    59. Re:Cool by Antique+Geekmeister · · Score: 1

      This is closer to what I have seen in my Debian and Ubuntu experience. Critical components, not part of "-stable" have to be integrated with some attention to detail. And _uninstalling_ them is a potential nightmare, particularly due to packages that are not well integrated or closely based on the current "-stable" repositories.

      It's more difficult and requires more attention than merely sticking to "-stable". And notice that of the others who've responded so far, one had to step away from "-stable" for ClamAV when the next release was looming, and another had to grab PHP 5. That _counts_ as deviating from "-stable".

    60. Re:Cool by Anonymous Coward · · Score: 0

      Possibly the "automatic unneeded package pruning". It could be dangerous if your custom apps don't specify their dependencies correctly (say, they rely on something that had been automatically installed by one of their other dependencies).

      "Automatic" in my experience* means "It looks like you aren't using these packages, please run apt-get autoremove to uninstall them". And even then, you can stop apt from showing that message by executing "apt-get install package" (will mark package as manually installed, removing it from the unneeded packages list).

      *: Mostly with Ubuntu, but I've seen that message too in Debian on a VM.

    61. Re:Cool by sydneyfong · · Score: 1

      I do have servers under my wing that runs on stable, or sometimes old-stable.

      Why is it not feasible to do so?

      For the one or two packages you absolutely need a more up-to-date version, there's always backports.

      I wonder whether you know what you're talking about, really.

      --
      Don't quote me on this.
    62. Re:Cool by X0563511 · · Score: 2, Informative

      This kind of thing does happen. I've reported a package that did so... we had a fix on the official repositories within days.

      Why this is considered a bad thing is beyond me. Debian on my server, other shinier toys on my desktop.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    63. Re:Cool by X0563511 · · Score: 3, Informative

      You know about Debian-Volatile right? What's even better, is this is part of -stable and has been included in your default repository config since Lenny released.

      This is meant specifically for moving targets like ClamAV.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    64. Re:Cool by X0563511 · · Score: 1

      Of course, apt and friends prompt you before they go about doing something. It's not some magical black box that issues random version numbers every time you run it.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    65. Re:Cool by X0563511 · · Score: 3, Interesting

      By the way, there are ways to hold packages too

      This is called pinning, if anyone is looking for the solution.

      or to make lists of packages required, even for your own scripts.

      "equivs" can be used to create empty packages for the sole purpose of manipulating dependencies. I usually use it to kill packages that are otherwise demanded in other important metapackages, though you could also use it to 'hold' dependencies for a broken third-party .deb package.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    66. Re:Cool by hardwarefreak · · Score: 1

      And honestly, while APT is great for desktop systems, I really wouldn't use it much on a server.

      "I'll rip off your head and shit down your neck!!" Pfft, fucking desktop pussies... APT was fucking _MADE_ _FOR_ _SERVERS_ you twit!

      I've never, not once, installed desktop Debian, nor installed console Debian via anything but a floppy disk or CD boot net install from the mirrors. All from an 80x25 text only VGA console or a serial terminal. And, afterward, for years, installed all needed packages and updates via apt-get, ON AN 80x25 TEXT CONSOLE. And you're telling _ME_ that apt is for desktops? Fucking LOL! I don't have current numbers, but I'd bet real money that there are vastly more headless Debian servers around the world than GUI Debian desktops.

    67. Re:Cool by hardwarefreak · · Score: 1

      Lenny, Postfix, period. (and yes, racked/headless)

      Can't get any more serious or "real" a server than a mail firewall/gateway. Sure, there's more in the basket, little necessary stuff like ntpd, postgrey, perl, jwhois, and tons of other little net tools and scripts. But _every_ server has a little basket filled with applicable tools. They aren't applications in a real sense, which is what you're referring to.

      For old hands, the _real_ servers are the single purpose boxen. Like guards at the gate, standing watch, turning away those who will do ill deeds.

    68. Re:Cool by Anonymous Coward · · Score: 0

      *raises hand*
      Our small compute cluster runs a basic debian-stable install. Plus some third party libraries (GotoBLAS and FLAME). But the rest is plain debian stable.

    69. Re:Cool by Anonymous Coward · · Score: 0

      raises hand.

      I generally compile applications if I need a newer version. It doesn't break down for me.

    70. Re:Cool by anilg · · Score: 1

      First apt based distro with ZFS?

      Nexenta has had support for both apt+zfs and full zfs root before FreeBSD. I wonder what FreeBSD's equivalent of apt-clone is.

      --
      http://dilemma.gulecha.org - My philospohical short film.
    71. Re:Cool by RichiH · · Score: 1

      > Sure, I'm assuming that someone will fix it, but you're assuming that someone will goof up in the first place. Seems fair.

      Hey, I wanted to post that :(

    72. Re:Cool by petermgreen · · Score: 1

      sudo apt-get install debhelper
      Reading Package Lists... Done
      Building Dependency Tree... Done
      The following extra packages will be installed:

      75 packages upgraded, 80 newly installed, 42 to remove and 858 not upgraded.
      Need to get 67.1MB of archives. After unpacking 26.9MB will be used.
      Do you want to continue? [Y/n] n

      It is quite clear that a careful user is not going to let such an upgade go through unless some hint is given by the system that core functionalities like those suggested by x-window-system and x-window-system-core, which the tool wants to remove, will not disappear, but will be properly replaced by some of the 80 new packages whose installation is suggested.It is quite clear that a careful user is not going to let such an upgade go through unless some hint is given by the system that core functionalities like those suggested by x-window-system and x-window-system-core, which the tool wants to remove, will not disappear, but will be properly replaced by some of the 80 new packages whose installation is suggested.

      If that happens it means the user is either running testing/unstable (they are called testing and unstable for a reason! DO NOT USE THEM IF YOU DO NOT HAVE THE SKILLS TO DEAL WITH BREAKAGE), has a sources.list not matched to what they have installed (the debian installer USED to do the stupid thing and putting stable/testing/unstable in sources.list but now it correctly uses the release codename) or is trying to upgrade between releases (in which case they should be reading the release notes, not randomly entering apt commands with no background information).

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    73. Re:Cool by drsmithy · · Score: 2, Informative

      You might want to read up on ZFS. It's not just a simple filesystem; it also includes LVM and RAID features, all in one. I'm not a professional sysadm, but my understanding is that, unlike Linux where changing a RAID5 array to have more disks or more space requires messing with the logical volumes using LVM, then messing with the RAID arrays with mdadm, then redoing the filesystems with the e[34]fstools, whereas doing this in ZFS just involves a few quick commands without having to unmount the filesystem.

      ZFS can't resize arrays at all, so comparing it to the process in Linux isn't really relevant.

      ZFS's "volume management" is certainly easier than in any other OS, but it's real killer feature IMHO is the block checksumming.

      In fact, a quick Google search shows that resizing an ext3 partition is a bit of a pain, involving converting it to ext2 (removing the journal), running resize2fs, and converting it back. Obviously that's not good if the power cuts out while you're doing this. With ZFS, you should be able to resize arrays, add disks, etc., without even interrupting service.

      Whatever article you found is wrong. The process for growing a FS in Linux (with LVM and ext3) is (can be done online):
      Increase size of LV (lvresize)
      Increase size of FS (resize2fs)

      To shrink (needs to be done offline):
      Unmount FS, run e2fsck
      Shrink filesystem (eesize2fs)
      Shrink LV (lvresize)

      The process is straightforward and not difficult, though it *is* easier with ZFS.

    74. Re:Cool by Anonymous Coward · · Score: 0

      C'mon, no professional just pokes "apt-get update" into the root shell on a live production server. That's just asking for hilarity, fail, and unemployment.

      Sure I would. Have you ever run apt-get update and examined its results?

    75. Re:Cool by Anonymous Coward · · Score: 0

      The correct rebuttal to your statement is that you don't mess with things on production machines. You don't uninstall that third package.

      That's only a possible correct rebuttal, and not even the most accurate. I have another one, coming from the man page of apt-get: stop worrying about it.

      from man apt-get, section upgrade:
      "under no circumstances are currently installed packages removed, or packages not already installed retrieved and installed"

      from man aptitude, section upgrade:
      "Installed packages will not be removed unless they are unused"

      If you want to make changes like that, you do it on your test machine first. Timothy was concerned with packages dropping dependencies, but that shouldn't happen within -stable.

      Even if it does happen in stable, using apt-get upgrade is perfectly safe.

      The moral of the story: don't believe everything you read on /. even if it is modded +4 insightful

    76. Re:Cool by smoker2 · · Score: 2, Informative

      I use ext2online to resize my LVM based ext3 filesystem. To clarify, when I have added a new disk to the LVM group, I use ext2online to extend the filesystem to cover the new disk. Also, I have never used lvresize. It's the same procedure as either lvreduce or lvextend. Here is my mini howto for adding a new disk to an LVM volume.

      adding a brand new disk to the lvm volume

      install the disk
      start system
      open terminal as root :

      check what disks (physical volumes)already exist as part of LV :
      pvscan

      run fdisk on the new drive :

      fdisk /dev/sdc # if this new disk is the 3rd sata or scsi disk in the system
                      # modify accordingly and use the same label throughout this # howto! (sdd, sde, sdf etc)

      create a new partition using the whole disk
      n
      p
      1
      when thats done enter the next command
      t
      partition will be 1
      hex code 8e

      write the partition table with
      w

      this exits fdisk

      then create a new physical volume on that disk
      pvcreate /dev/sdc

      then add the physical volume to the volume group

      vgextend my_movies_group /dev/sdc

      run pvscan again to check its in there

      then run vgdisplay and take a note of the allocated PE for the logical volume
      and for the free PE
      then add them together ! (this figure should already be showing in vgdisplay as Total PE )

      then run
      lvextend -l22222 /dev/my_movies_group/my_media /dev/sdc
          # where 22222 is the total you just calculated

      you can run vgdisplay -v again to check its been added ok

      lastly, run this, to resize the filesystem to cover the new drive

      ext2online /dev/my_movies_group/my_media

      That should be it !

    77. Re:Cool by drsmithy · · Score: 2, Informative

      Just for reference, if you're going to be using the whole device as a PV, there's no real need to create a partition on it - and in some cases it can be detrimental (eg: if it's a RAID device you will probably introduce partition alignment problems that will impact performance).

    78. Re:Cool by stevey · · Score: 1

      Yes - it wasn't available in Sarge, and it wasn't well supported in Etch.

      Still now it exists for Lenny it will be a good resource - though personally I always made my own backports when I realiased I needed them.

    79. Re:Cool by Anonymous Coward · · Score: 0


      apt-get install $(apt-file search `ldd $app | grep -v \/` )

      fixed it tho

    80. Re:Cool by Antique+Geekmeister · · Score: 1

      Let me quote you, here:

      > For the one or two packages you absolutely need a more up-to-date version, there's always backports.

      Then it's not purely "-stable", is it? Backports is useful, I approve of backports. But it's like saying "We run on pure RHEL" and then installing components from RPMforge, or installing components from CPAN: you get more usable software, but lose that assurance of server grade software and of component compatibility and consistency.

    81. Re:Cool by Nevyn · · Score: 1

      Why would you not want to use APT on a server? What part of [...], easy security update application [...] do you not want on your servers?

      APT has a security update application?

      Doing a search the only things I found were like: this, which is from 2006 and just talks about the "special" security repo. Debian.org/security doesn't point to anything like a security update application either.

      --
      ustr: Managed string API with ave. 44% overhead over strdup(), for 0-20B
    82. Re:Cool by EvilRyry · · Score: 1

      It certainly has a cost, but I wouldn't go so far as to call it retarded. On top of the cost it also has some very strong real world advantages such as write hole elimination(safer), variable block size (better performance), selective resilvering (faster recovery), in FS snapshots (snapshots are actually usable on a production system, unlike LVM).

    83. Re:Cool by Anonymous Coward · · Score: 0

      > I'm not sure if freeBSD supports ZFS root or not

      It does and it doesn't.

      The same bloke often posts about ZFS on FreeBSD, his blog's worth a read.

    84. Re:Cool by Anonymous Coward · · Score: 0

      This almost never happens in practice because most dependencies are automatically generated. Also packages are removed only if you dist-upgrade (to upgrade to a new version of debian or are running unstable). The "normal" upgrade never removes packages. So the scenario you described is 1. rare, 2. only happens during upgrades to a new version of the distribution.

      If you are running unstable on a server you should be prepared for minor breakages every now and then during upgrades and if you do a major upgrade you should do proper testing and allow for some unplanned downtime in the process.

    85. Re:Cool by smoker2 · · Score: 2, Interesting
      Yes I realise you can do without a partition, but when I first set up the LVM volume, I followed best practice as found in the howto I linked.

      Using the whole disk as a PV (as opposed to a partition spanning the whole disk) is not recommended because of the management issues it can create. Any other OS that looks at the disk will not recognize the LVM metadata and display the disk as being free, so it is likely it will be overwritten. LVM itself will work fine with whole disk PVs.

      In future on dedicated machines I probably won't use that method, but then again, LVM has issues anyway. I set up my first volume without using RAID, as the disks were of differing sizes. I added a new disk some time later which very quickly began to exhibit problems. Because the disk wouldn't read properly, I couldn't retrieve any data from it and couldn't therefore remove the PV because it had data on. The only way was to shut down, then remove the disk, then remove the volume then use the metadata to rebuild the volume. I lost a few files but the main worry was whether I would get any files back at all. To this day, I have files visible on the volume that correspond to lost data. I cannot find anyway to delete these files as they only exist in the metadata not on the actual disk. I'm not confident that creating a replacement volume (on the same drives) would allow me to inherit an accurate filesystem from the old volume.

      So my only option at present is to buy progressively bigger drives, run them in for a few weeks, then after adding them to the volume, use the free extents to move data off of the older drives. This procedure is fraught with risk. Any tips ?

    86. Re:Cool by Grishnakh · · Score: 2, Informative

      I'm no expert on this, but it sounds like you need to just start over: instead of adding new disks to your existing volume, just create an all-new volume and move everything from the old one to the new one, so you don't inherit any weird problems.

    87. Re:Cool by marcosdumay · · Score: 1

      If you go all the way into creating a test environment, having an inhouse repository (or at least a cache) is the least one should expect.

    88. Re:Cool by jgrahn · · Score: 1

      Would everyone who actually manages to run their servers with only "-stable" please raise their hands? One, two... no wait, that person is scratching their head.

      *Raises hand*. There's nothing I need on my home server that hasn't existed and been stable for a decade or so. Much of the time it runs Debian 'oldstable' because even though an upgrade would be cool, there is no actual need. At work, we recently got rid of Redhat 8, which is from 2002 or something. No new features we wanted; it simply didn't run on new hardware.

      So what is this bleeding edge stuff everyone needs, according to you? Maybe you are a niche user. Or maybe I am.

    89. Re:Cool by agrounds · · Score: 1

      Consider my hand raised then?

      My webserver and email servers both run -stable on the powerpc port.

    90. Re:Cool by Nevyn · · Score: 1

      Sure, someone will fix it ... eventually. But why take that risk for 0.001 cents of disk space? It seems like the most retarded feature in the world to me, and yet I can't deny that people are asking for it in yum ... so you aren't the only person with a big OCD desire to remove every unneeded bit.

      --
      ustr: Managed string API with ave. 44% overhead over strdup(), for 0-20B
    91. Re:Cool by Timothy+Brownawell · · Score: 1

      Shit does happen everywhere, but that shouldn't be your policy.

      No, it should: "given that we can't prevent all shit from happening, how can we be most resilient against the shit that does happen?".

    92. Re:Cool by DragonWriter · · Score: 1

      Possibly the "automatic unneeded package pruning". It could be dangerous if your custom apps don't specify their dependencies correctly (say, they rely on something that had been automatically installed by one of their other dependencies).

      I'm pretty sure both automatically installing dependencies and automatically pruning those no longer needed can be disabled when using apt via the command line tools, though perhaps not through all graphical front-ends. (I've never had to avoid either, but I'm pretty sure I've seen the options to do so.)

    93. Re:Cool by Ant+P. · · Score: 1

      Those are good points and I take that comment back.

    94. Re:Cool by rantingkitten · · Score: 1

      On every apt-based system I've ever used, it doesn't just remove unneeded packages. It tells you "These packages are no longer needed, use apt-get autoremove to remove them." And then you can check to see what packages it is talking about and make sure it's correct before doing the autoremove.. or you can let them stay there forever since they're probably not going to hurt anything even if they really are unnecessary.

      Your scenario only happens, to my knowledge, if you had some sort of cronjob that kicked off the autoremove after every upgrade or something similar, which would be retarded and you'd have no one but yourself to blame. When do packages just "go away" as you claim?

      --
      mirrorshades radio -- darkwave, industrial, futurepop, ebm.
    95. Re:Cool by Timothy+Brownawell · · Score: 1

      It also happens if you use 'aptitude' instead of 'apt-get'.

    96. Re:Cool by koiransuklaa · · Score: 1

      Ok, I read it. It said two things as far as I can tell:
      1. apt conflict resolution is not optimal
      2. apt-get UI is not optimal in some situations

      So, could you now tell me how "apt-get update" is harmful?
      Also, could you please explain how problem #2 is relevant here? It's pretty clear the apt-get UI isn't friendly on huge upgrades but how is that interesting when we are talking about Debian stable administration?

    97. Re:Cool by True+Grit · · Score: 1

      The way Linux is going at the moment we pretty much need a microkernel ... Linus pretty much stated that a couple of weeks ago.

      Bullshit.

      His comment was about the size and complexity of the kernel, which is an inevitable occurrence for any OS that becomes popular. People went nuts over his comment only because at one point he used the word 'bloat' to refer to the problem, and folks started reading more into that than what he actually said or meant. However, anyone who actually read what *he* said, in full, rather than what others claimed he said, just yawned and asked 'Why is this a story?'.

      And at no time did he ever mention microkernels, and since has reiterated his opinion about them.

  3. Linux vs. FreeBSD by Boawk · · Score: 3, Interesting

    As a UNIX/Linux veteran, I have to admit that I've almost no experience with FreeBSD. Could someone summarize why one might prefer it over Linux?

    1. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 5, Funny

      Features such as jails, the OpenBSD Packet Filter and support for NDIS drivers in the mainline kernel.

    2. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      You will be more l33t for using it.

    3. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 1, Interesting

      why is NDIS considered a feature? assuming we're both thinking of running Windows drivers.

    4. Re:Linux vs. FreeBSD by Bill,+Shooter+of+Bul · · Score: 3, Informative

      That depends upon what you mean by veteran, and what you mean by UNIX. FreeBsd is closer to Unix due to its BSDness. So if you are used to kernels that are more Unix-y than Linux-y you may prefer it for that reason. If you are simply a fan of OSS that runs it as a desktop, there may not be any obvious advantages and perhaps some disadvantages due to lack of desktop like software. It should also include ZFS & dtrace which may entice you. Its also just a different kernel with a different schedule that may perform better for your specific tasks. Osnews carried a story about a benchmark between FreeBSD and Ubuntu the comments from osnews readers are also pretty insightful which is why I linked to them and not the source article. .

      --
      Well.. maybe. Or Maybe not. But Definitely not sort of.
    5. Re:Linux vs. FreeBSD by Niten · · Score: 5, Informative

      In my opinion, the biggest advantage of FreeBSD is how coherent the system is. Everything, from documentation to userspace utilities to the kernel, was developed and tested and released as a single project.

      This allows for neat features that require cooperation between several system components, which would be more difficult to implement in the Linux world. For instance, in FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy.

      So it seems to me that this Debian project defeats the most attractive feature of the FreeBSD operating system (by separating its kernel from its tightly integrated BSD userspace), while simultaneously casting aside Linux's advantages over FreeBSD (more drivers, more supported architectures, somewhat better performance, and--this may be controversial--in my experience, better stability). On the other hand, maybe Debian really can improve on the FreeBSD experience; apt rocks, and the Debian project does perhaps a better job than anyone of combining the disparate parts of the GNU/Linux ecosystem into a coherent operating system. So kneejerk reactions aside, I guess I shouldn't judge this until I have the chance to try it... still, I don't see myself trading in my Debian GNU/Linux anytime soon.

    6. Re:Linux vs. FreeBSD by ground.zero.612 · · Score: 2, Informative

      As a UNIX/Linux veteran, I have to admit that I've almost no experience with FreeBSD. Could someone summarize why one might prefer it over Linux?

      FreeBSD is unix-like http://en.wikipedia.org/wiki/FreeBSD.

      You might prefer it over Linuxes for some of the same reasons you might prefer Apple's Mac OS X http://en.wikipedia.org/wiki/OS_X#History.

      Other than that, perhaps rock-solid stability, ZFS, or it's package management system (admittedly I don't use much Linux, but the pkg_add utility and the ports tree in fbsd are excellent). Oh and did I mention Linux binary compatibility?

      --
      "Be prepared, son. That's my motto. Be prepared." --Joe Hallenbeck
    7. Re:Linux vs. FreeBSD by blind+biker · · Score: 5, Interesting

      One thing I noticed: a working and consistent sound system.

      --
      "The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
    8. Re:Linux vs. FreeBSD by XanC · · Score: 3, Informative

      For instance, in FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy.

      Isn't this an internal feature of their cp implementation? I don't see what this has to do with the kernel, or indeed any program besides cp, at all.

    9. Re:Linux vs. FreeBSD by quantum+bit · · Score: 1

      One thing I noticed: a working and consistent sound system.

      QFT.

      Every time I've tried Linux on my desktop, one of the major things that's always driven me back to FreeBSD is the utter mess that is the Linux sound system.

    10. Re:Linux vs. FreeBSD by poopdeville · · Score: 4, Interesting

      On the other hand, maybe Debian really can improve on the FreeBSD experience; apt rocks, and the Debian project does perhaps a better job than anyone of combining the disparate parts of the GNU/Linux ecosystem into a coherent operating system.

      I am not a big fan of the BSD userland, and I typically install "prefixed Gentoo" on my Macs. (Basically, it brings in a GNU user land, a fresh compiler chain, etc. It works well, but the repositories are very basic. It can help set up a Unixy programming environment, not a feature complete Unixy desktop system)

      kFreeBSD Debian can potentially make Apt a real option on Macs. Fink sucks. Debian's repositories are much better.

      --
      After all, I am strangely colored.
    11. Re:Linux vs. FreeBSD by ducomputergeek · · Score: 5, Interesting

      As someone who has had a lot of experience with both, I switched to BSD in 1999. Back then the main reason was Ports. Needed to install MySQL: /usr/bin/ports/databases/mysql/ make && install. Then go grab a cup of coffee come back and it would fetch everything it needed, compile, and run. Or you could fetch a pre-compiled binary via pkg_add -r mysql. Hell, the first few version of PostgreSQL I used, the only way I could get the damn thing to work was to use BSD ports. The best you had with Linux was RPM and that was dependancy hell at times.

      Also, back in the day it had a better tcp/ip stack and was generally more stable as a server platform and decent SMP support. And frankly it was far easier to support than "linux" was back in the day because there was a single FreeBSD, not umpteen different flavors.

      Today it has ZFS and Dtrace from solaris ported over. I know ZFS hasn't made it into Linux as of yet, not sure about DTrace. But both are handy tools.

      Currently we're deployed 100% on FreeBSD for our web, mail, and database servers running PostgreSQL. But that has more to do with using Pair Networks than any other single factor. They've been 100% FreeBSD and consistently in the top 10 in terms of uptime according to netcraft.

      For the past 10 years, I've found FreeBSD to be a stable, secure server operating system that doesn't take a lot of system resources to run. It seems like Linux takes about 256MB of ram these days in most default configs to run a web server whereas our BSD machines were using closer to 150MB for the core OS. And was both systems running Apache 2.

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
    12. Re:Linux vs. FreeBSD by cbhacking · · Score: 4, Informative

      A lot of wireless hardware still doesn't have drivers (at least, not working ones) on Linux (or FreeBSD). For obvious reasons, NT drivers do exist. FreeBSD's kernel supports directly loading those NT drivers. Linux has ndiswrapper, a project to allow the same thing (ndiswrapper itself is a Linux kernel module that attempts to load the NT driver binary) but the FreeBSD NDIS support is a feature of the kernel itself, and supported as such.

      For TL;DR folks: if you've ever had trouble making WiFi work in Linux, this might help.

      --
      There's no place I could be, since I've found Serenity...
    13. Re:Linux vs. FreeBSD by VGPowerlord · · Score: 1

      This allows for neat features that require cooperation between several system components, which would be more difficult to implement in the Linux world. For instance, in FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy.

      I've looked up SIGINFO (which doesn't exist as such on Linux), and I'm not sure why this would require several components to work together. I imagine it works just like SIGINT does. You know, SIGINT, the signal sent to the running process when you hit ^C?

      Or did you mean this when cp is running in the background?

      'course, on Linux, the standard shell (GNU bash) and the standard cp (part of GNU coreutils) are by the same people, so if they really wanted to, there's nothing stopping them from linking the two.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    14. Re:Linux vs. FreeBSD by Sancho · · Score: 1

      I haven't noticed many stability differences between FreeBSD and Linux. What I have noticed is ABI stability--which is much better on FreeBSD.

      5 years ago, I probably would have been really gung-ho about Debian FreeBSD. Back then, managing FreeBSD ports was difficult, and there wasn't a binary update mechanism for kernel or world. Now we have freebsd-update (which rocks), portsnap (for syncing your ports tree) and the port management tools are more fleshed out and stable. You still have to compile for ports updates, which takes time, so that's probably the main place where Debian can improve things at this point.

      Overall, I like the FreeBSD system more than most Linux distributions I've tried. PF is great at what it does, though the modularity and extensions to Netfilter can be pretty useful. In ad-hoc testing of stock FreeBSD and Linux systems, we tend to get more throughput from FreeBSD bridging firewalls, and I personally find them easier to manage than most Linux distributions.

      My "best of both worlds" would be the performance of FreeBSD, the drivers of Linux, and a flexible firewall. Unfortunately, I suspect that you only get one of these with Debian's FreeBSD offering.

    15. Re:Linux vs. FreeBSD by Xtifr · · Score: 2, Funny

      That depends upon what you mean by veteran, and what you mean by UNIX.

      Ayup.

      FreeBsd is closer to Unix due to its BSDness.

      A peculiar interpretation. In the early days, I tended to prefer Linux over BSD because Linux generally acted more like the real UNIX(tm) systems at work, while BSD remained inherently...BSDish. Linux was like a Unix inflicted with a random, confusing scattering of BSDisms (like the operation of ps(1)). Of course, if you consider BSD to be The One True Unix (as many BSD fans do), then Linux looks like a UNIX with a random, confusing scattering of SYSVisms. In conclusion, I think I have to say that your first statement quoted above completely invalidates the second one, and I agree with the first. But I like BSD anyway. :)

    16. Re:Linux vs. FreeBSD by Sancho · · Score: 5, Informative

      The grandparent was trying to make a point, but failed. Similar behavior exists throughout the FreeBSD userland--you can send SIGINFO with ctrl-t to many userland processes to get information on what they're doing. The point is that FreeBSD's kernel and userland were designed as a system, and little touches like this show that off.

    17. Re:Linux vs. FreeBSD by Jurily · · Score: 1

      So it seems to me that this Debian project defeats the most attractive feature of the FreeBSD operating system (...), while simultaneously casting aside Linux's advantages over FreeBSD

      There is one other thing though: the quality of Debian as a whole will improve, since you can't assume anything about the kernel anymore.

      "I chose 1000 originally partly as a way to make sure that people that assumed HZ was 100 would get a swift kick in the pants." -- Linus Torvalds

    18. Re:Linux vs. FreeBSD by vlm · · Score: 1

      .... I switched to BSD in 1999. .... The best you had with Linux was RPM and that was dependancy hell at times. ....

      First entry in /usr/share/doc/mysql-common/changelog.Debian.gz is dated 12 apr 1997

      No dependency hell in Debian...

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    19. Re:Linux vs. FreeBSD by vga_init · · Score: 1

      Apt is indeed awesome, but FreeBSD's package system is GREAT. I never had a complaint with the way FreeBSD handles packages, and there are some fantastic package utilities in ports that make nightmarish tasks freakishly easy, such as package pruning. One great advantage in BSD is that the base system is not packaged, so if you ever start having major package issues, you can simply wipe out all packages and reinstall your applications from scratch; this keeps your systems squeaky clean without ever having to reinstall the OS.

    20. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      So it seems to me that this Debian project defeats the most attractive feature of the FreeBSD operating system (by separating its kernel from its tightly integrated BSD userspace),

      Seems you do not know that the FreeBSD operating system is the monolithic kernel called FreeBSD. To tamper with the OS, you need only to tamper with the kernel. Not touching at all to user space. We are talking about monolithic kernel, not about microkernel, what would mean that other parts of the OS are running on user space among other softwares (daemons, system applications, application programs, libraries etc)

      (more drivers, more supported architectures, somewhat better performance, and--this may be controversial--in my experience, better stability)

      I hope I did read wrong that. Because Linux OS (kernel) supports more architectures, wider device support and stability is still great. And performance, well we can just see that Linux is superior OS to FreeBSD on that (but does it matter on realworld environments? On labs it does) and even to Hurd (When is GNU going to get working their own OS?)

    21. Re:Linux vs. FreeBSD by Guy+Harris · · Score: 5, Informative

      For instance, in FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy.

      Isn't this an internal feature of their cp implementation?

      No. The fact that ^T sends SIGINFO, just as ^C sends SIGINT, is a feature of the "tty driver" (standard tty line discipline). The fact that a particular program catches SIGINFO and prints a progress report is a feature of the program.

      I don't see what this has to do with the kernel...

      The standard tty line discipline referred to above is in the kernel.

    22. Re:Linux vs. FreeBSD by rbanffy · · Score: 1

      Actually, if BSD is dead, using it would be more like necrophilia than archeology

    23. Re:Linux vs. FreeBSD by 0100010001010011 · · Score: 1, Funny

      Nonsense. I mean you install the alsa drivers for your sound card. Then you install pulse audio to load and use the drivers. Then something something alsa user settings to use pulse.

      So then all your app has to do is send stuff to alsa userspace thingy, which sends everything to pules, which CAN send stuff over the network, but most likely sends it to the ALSA device drivers.

      2-3 hours later and XBMC with HDMI audio works great! /sarcasm.

    24. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      As long as you install the sometimes requiered binary only, closed source firmware, wireless driver support in linux is getting pretty good. Most laptops have either an Intel, Atheros, Broadcom or Ralink cards and they are all supported.
      Most of the time, it's easier to get them working than in windows xp because you don't need to search the internet for a compatible driver. I tried windows 7 and it was much better however (working out of the box).

    25. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 1, Interesting

      Apt is indeed awesome, but FreeBSD's package system is GREAT. I never had a complaint with the way FreeBSD handles packages, and there are some fantastic package utilities in ports that make nightmarish tasks freakishly easy, such as package pruning.

      Apt, ports, pacman, and the like are more-or-less converging, feature-wise. I'm sure package pruning is freakishly easy with Apt by now too. The nice thing about Apt is the fact that the Debian team is behind it. Essentially, Apt and the Apt repositories are the heart of Debian. You get the same kind of quality control for packages that the all the other Debian ports have.

      Of course, Ports has the BSD teams behind it, but no "central repository" for quality control of programs outside of their domain.

      One great advantage in BSD is that the base system is not packaged, so if you ever start having major package issues, you can simply wipe out all packages and reinstall your applications from scratch;

      I don't see how that is an advantage. I can do that with Debian. Heck, now I can take a Debian Linux system, create a chrooted environment from which to build FreeBSD, install FreeBSD, install Debian for FreeBSD, and upgrade to the latest Linux kernel, without ever shutting the machine off. The point being, you can always do what you suggest.

    26. Re:Linux vs. FreeBSD by EvanED · · Score: 1

      I've looked up SIGINFO (which doesn't exist as such on Linux), and I'm not sure why this would require several components to work together.

      It requires (1) the kernel to know about SIGINFO, (2) the tty or whatever generates the SIGINFO to raise the signal, and (3) the program to catch it. ...there's nothing stopping them from linking the two.

      (1) If the kernel doesn't know about SIGINFO, what would it send to the program? Sure, you could use SIGUSR1 or something, but on BSD it gets a distinguished signal that won't confuse other programs if you send it to them. (E.g. say ctrl-T sent SIGUSR1 to a program that responded to it in an entirely different way. As a user, you'd need to know what signals a program expects; you wouldn't be able to just safely try it and see if it works, because SIGINFO would be ignored if the program wasn't going to do something reasonable.)

      (2) How would the shell know to raise the signal?

      It's possible that this has answers I don't know; I don't know how, say, the SIGINT signal flows around when it's raised. But by my understanding, it at least starts out being noticed by the TTY driver = kernel = not something GNU could do on its own.

    27. Re:Linux vs. FreeBSD by sinclair44 · · Score: 1

      Don't mix up OS X's BSD history with its current state. A friend of mine put it rather well: "OS X was BSD once just like orcs were elves once".

      --
      Omnes stulti sunt.
    28. Re:Linux vs. FreeBSD by TheRaven64 · · Score: 1

      Project Evil is a kernel module on FreeBSD as well, it's just maintained in the main tree.

      --
      I am TheRaven on Soylent News
    29. Re:Linux vs. FreeBSD by Narcocide · · Score: 1

      Personally I'd say that if you're running the system as any type of firewall/router Packet Filter is reason enough.

    30. Re:Linux vs. FreeBSD by EXrider · · Score: 1

      This allows for neat features that require cooperation between several system components, which would be more difficult to implement in the Linux world. For instance, in FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy.

      Wow, nifty, I never knew about that! I just tried it out on Mac OS 10.5.

      --
      grep -iw skynet /etc/services
    31. Re:Linux vs. FreeBSD by TheRaven64 · · Score: 5, Informative

      I would say mod parent up, but you're already at +5. This was what made me switch from Linux around 2002: Sound Works. I had a cheapy AC97 CODEC in my computer at the time. It didn't do hardware mixing. I installed Linux. There were two drivers, an OSS one and an ALSA one. Neither one let the majority of my programs play sound at the same time. For example, I couldn't have xmms playing music while I played BZFlag. Both KDE and GNOME came with their own sound daemon, which meant that either KDE programs could make sounds or GNOME programs could make sounds. Oh, and I think if I used the ALSA drivers then two programs that had had their sound output rewritten to use ALSA could play sounds at once... sometimes. Then I tried FreeBSD.

      In FreeBSD, there was one driver for the sound device. This was back in the 4.x days, so sound devices needed a little bit of extra configuration. I had to set the number of virtual channels and then tell each device to talk to a different one. I set up 4, one for GNOME, one for KDE, one for xmms and left the default one for whatever apps just wrote to /dev/dsp. I could have music playing, BZFlag sound effects in the game, and notification beeps when I got an email or IM. Then came FreeBSD 5 and all of that manual configuration went away. To play sound, a program opens /dev/dsp and writes audio data to it. That's it. No libraries to link against; it's all done via standard UNIX system calls (open(), read(), write(), and ioctl()). It's trivial to program for and it just works. With FreeBSD 8, you now get per-channel volume controls and a rewritten high-performance mixing algorithm, as well as support for all of the new OSS 4 APIs (backwards - binary - compatible with the old OSS 3 ones) if care about those things (i.e. if you are a programmer).

      A few other things that I like about FreeBSD:

      1. Documentation. The man pages and the handbook are written by people who seem to understand both English and the subject that they are writing about. They've also been translated into other languages, but I've only read the English ones.
      2. Sane and simple init system. RCng isn't as all-singing, all-dancing as something like Launchd, but it is sufficiently powerful and easy to understand.
      3. Interfaces don't change. Once you learn how to do something on FreeBSD, you know how to do it. Tools don't get arbitrarily replaced with ones that are in some way better but have completely different interfaces.
      4. Clear separation between FreeBSD and Other Stuff. Ports go in /usr/local/, everything else goes above that.
      5. Jails. If you want to isolate something, or you want a simple testing environment, jails are great. They are almost the same as chroot, except each jail has its own set of users. Root in a jail is not root outside the jail and so can create arbitrary other users inside the jail but can't escape. With FreeBSD 8, jails have virtualised network stacks and now work recursively, so root in a jail can create a new jail. They work especially well when combined with ZFS, because you can make an O(1) ZFS clone of a clean install of the base system to create new jails.
      6. ZFS. RAID without the write hole, per block checksums, top-to-bottom transaction support, O(1) snapshots and clones - what's not to like?
      7. Ports / packages. Not really anything special anymore, but nice and clean. Ports build from source and can have custom build options easily set. Packages are built from ports. Use them interchangeably.

      There are probably some things I've forgotten. I was going to say that the only thing I miss on FreeBSD was valgrind, but it turns out that Valgrind 3.5 has been in ports for a little while now and I just wasn't paying attention (so, obviously, I don't miss it that much).

      --
      I am TheRaven on Soylent News
    32. Re:Linux vs. FreeBSD by Hurricane78 · · Score: 1

      And Portage was why I chose Gentoo. It has some features that Ports hasn't (Paludis and others are even better), but still has what Ports is loved for.

      Actually ZFS has made it to Linux, as I'm currently using it for my archive disks. It's implemented as FUSE. But it's eating crazy amounts of CPU and memory. An example: It can easily take one of two cores and 600 MB RAM, in normal usage. (It's worth it, as I have data where I absolutely can't tolerate bit errors, and where RAID and backups simply aren't enough.)

      --
      Any sufficiently advanced intelligence is indistinguishable from stupidity.
    33. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      I tended to prefer Linux over BSD because Linux generally acted more like the real UNIX(tm) systems at work ...

      You should have stated your preference for SysV beforehand. UNIX is many things. Even UNIX(tm).

    34. Re:Linux vs. FreeBSD by Bill,+Shooter+of+Bul · · Score: 1

      Yeah, I'd agree with that. IMHO, Unix means UNIX (TM) which means its been formally submitted and certified as UNIX by the Open group. I find BSD systems to be more like those official UNIX systems like Solaris and Mac OS X ( well the unix-y underlying parts) at least with the userland utilities.

      --
      Well.. maybe. Or Maybe not. But Definitely not sort of.
    35. Re:Linux vs. FreeBSD by notamisfit · · Score: 1

      Agree on your seven points above. Sadly, I've got one that pretty much vetoes all of them and puts me in the Linux camp: the turnaround time for hardware SUCKS. When I last used FreeBSD around 6.1 or 6.2ish, I was having to import out-of-tree drivers for a stock Intel onboard chip (on a two-year old motherboard). The last time I tried to install it on a laptop, it pretty much gave up at the boot screen, and I shudder to think of whether or not it would even get that far on my Qosmio. It's kind of a shame, cause I really miss the system.

      --
      Jesus is coming -- look busy!
    36. Re:Linux vs. FreeBSD by Grishnakh · · Score: 2, Interesting

      So what's keeping the Linux folks from just merging the OSS4 drivers anyway?

      I believe there's a Linux kernel fork that does just that, but obviously it hasn't been accepted into the mainline.

    37. Re:Linux vs. FreeBSD by eqisow · · Score: 1

      First thing I do on my Linux desktop: install OSS v4.

      No sound problems at all.

    38. Re:Linux vs. FreeBSD by poopdeville · · Score: 1

      Well thanks for that suggestion, but I didn't mix anything up.

      FreeBSD also has a minor history of merging good Apple tools in... FreeBSD is moving to (or has moved to) Apple's launchd initialization system over the old rc scripts. Apple uses the same user land tools that FreeBSD does, so far as possible (and it is very far, from a "command line" perspective.

      In short, the only major differences between OS X and FreeBSD are (or soon will be): the kernel, and all that that entails (for example, OS X will eventually have ZFS, but not because FreeBSD does); Cocoa and the other Apple specific API's; ports package management instead of none.

      Debian has clearly abstracted the kernel issue away. Apt is a ports replacement, so that difference is irrelevant. The only possible issue would be binary compatibility, which is relatively straightforward to solve. In fact, it is solved, since both use ELF and the same instruction sets.

      --
      After all, I am strangely colored.
    39. Re:Linux vs. FreeBSD by cbhacking · · Score: 1

      Linux is working on it, but they're not there yet.

      For example, I have an older Compaq laptop that uses a slightly modified Broadcom chipset that the "native" driver specifically does not support, and that ndiswrapper doesn't really support (you can see networks, but not connect). Were I to bother trying again to get the WiFi working in something non-Windows, I'd try FreeBSD next.

      In fact, across 4 different attempts (on different hardware), I've gotten ndiswrapper to work once, and am 1 for 3 trying to get the Broadcom driver to work. Granted some of those were using older versions of the software, but outside of Intel chipsets, I've yet to actually see WiFi "just work" under Linux. That said, I haven't tried Ralink - but one older Atheros-based card (yes, removable card) never worked at all for me in Linux.

      --
      There's no place I could be, since I've found Serenity...
    40. Re:Linux vs. FreeBSD by toejam13 · · Score: 1

      This is one of the main reasons why I prefer FreeBSD as well. When I tried GNU/Linux a few years ago, I kept running into issues with apps using sound under KDE. There were like six or so sound subsystems that I had to fuss with. Each program seemed to want to use a different system. In the end, I got it to work, but it did seem to be an annoying PITA.

      Under FreeBSD, it just works. I've used a number of audio cards, from Creative SB-16 and AWE64 cards to Intel HDA compatible embedded chipsets. Never had a problem with an app, and the only problem with the kernel driver is that the snd_hda driver doesn't support SPDIF output as of 7.2-RELEASE.

      Another thing that I like about FreeBSD is that its entire memory footprint at a login shell, including audio subsystem, isn't much more than what is consumed just by the RealTek audio driver and processes alone under Windows XP. 26MB just for the RTHDCPL.EXE process? I don't care how cheap memory is these days. That's simply a disgusting amount of memory to use for a friggin sound UI app.

      Oh, the final reason why I like FreeBSD more. Centralized bug reporting. I had an issue with an AMI MegaRAID controller when I first tried a Linux distro over a decade ago. I went to the distro, and they said to talk to the kernel guys. Talked to the kernel guys, they told me to talk to the vendor who wrote the driver. Talked to the vendor, and they told me that it was essentially as-is. Dropped the same card into FreeBSD, and it worked fine. It even caught the weird IO firmware issue with the Quantum SCSI drive I was using.

    41. Re:Linux vs. FreeBSD by Ant+P. · · Score: 1

      The OSS developer appears to be schizophrenic.

      Either that or they just don't care.

    42. Re:Linux vs. FreeBSD by juuri · · Score: 1

      The "real" unix systems at work? What you start using unix in the early 90s or something?

      SysV is the devil... unfortunately the wrong kind.

      --
      --- I do not moderate.
    43. Re:Linux vs. FreeBSD by Sir_Lewk · · Score: 1

      Clever.

      Meanwhile those of use not running Linux From Scratch generally don't have to do shit to get sound working. It's been working "out of the box" with practically every distro I've used since I've had this computer, nearly 3 years.

      Hell, even in Gentoo all I had to do was run `alsaconfig`.

      --
      "linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
    44. Re:Linux vs. FreeBSD by Xtifr · · Score: 1

      You should have stated your preference for SysV

      I didn't prefer SysV. I was forced to use SysV. I preferred not to confuse myself by using a less-compatible product at home when there was another perfectly decent and more compatible option available for the same price. :)

    45. Re:Linux vs. FreeBSD by Ant+P. · · Score: 1

      It seems like Linux takes about 256MB of ram these days in most default configs to run a web server whereas our BSD machines were using closer to 150MB for the core OS. And was both systems running Apache 2.

      Linux 2.4 or 2.6?

      My Gentoo server's running at about 160MB right now. This includes a Lighttpd/PHP/MySQL/Postgres web server, along with running most of my LAN (wifi, BIND, dhcp), and a few screen sessions. Then again, it's not really a default config...

    46. Re:Linux vs. FreeBSD by bonch · · Score: 0

      cp, like the rest of the userland tools, are a shipping part of BSD and not a separately maintained package, so talking about BSD's cp is talking about BSD.

    47. Re:Linux vs. FreeBSD by a09bdb811a · · Score: 1

      With FreeBSD 8, you now get per-channel volume controls

      Cool. How and where is this exposed? Does it show up in the KDE mixer? Or perhaps an equivalent of alsamixer?

      (Questions like this keep me on Linux because I dread having to re-familiarise myself with everything, even if it's only slightly different on BSD.)

    48. Re:Linux vs. FreeBSD by k8to · · Score: 1

      Linux is as Unix-y as BSD. They're both Unix.
      Duh.

      Perhaps you mean closer to the first implementation of UNIX? There've been like 8 at least by now.

      --
      -josh
    49. Re:Linux vs. FreeBSD by blind+biker · · Score: 1

      The funny thing is, if you have been using Linux for long enough, you already went through a phase of re-familiarization, from OSS to Alsa. But now there's this new kid in town in Linux, PulseAudio. It is supposed to work together with ALSA, but there are problems. I mean, an entirely new category of problems which you will have to get familiar with.

      --
      "The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
    50. Re:Linux vs. FreeBSD by TheRaven64 · · Score: 1

      There is a command-line mixer tool which exposes them. No idea about the KDE mixer, but if it uses the standard OSS APIs then it will work. They are exposed via the OSS4 APIs to programs too, so apps can set their own volume without affecting the system-wide volume. Unlike the official OSS4 code, it also has some hacks that make apps that were using the old OSS3 mixer APIs for setting volume work correctly with per-channel volume controls.

      --
      I am TheRaven on Soylent News
    51. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      Well, I can't help you if you use "slightly modified Broadcom chipset" since the b43 driver works only with devices from broadcom which haven't been modified by the user.
      When I first tried a broadcom PCMCIA card, it worked with ndis wrapper. It now also works with b43, so I don't really understand what you are talking about.
      When was your experience? Wifi under linux improved a lot in the last years.
      As for your atheros card, again, which chipset is it? The madwifi driver as been working well for a while, and there is now ath5k and ath9k which are 100% free and included in the kernel.
      I think I would have hard time going to the store and finding a laptop with wireless not supported by linux...

    52. Re:Linux vs. FreeBSD by Frankenvince · · Score: 1

      Let me add to that list:

      - A kick ass mascot that your girlfriend would like.

      - Netgraph: try it and never look back.

      One bad point that hasn't been mention: nvidia-freebsd-amd64's non-existance. This long-lasting feud seems to be resolved but Nvidia has yet to show the goodie.

    53. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      VMS also has a similar feature when you press ^T which I find myself using all the time. As someone who has been using FreeBSD as their desktop OS for several years now, I can't believe I didn't know about it in FreeBSD!

    54. Re:Linux vs. FreeBSD by petrus4 · · Score: 1

      The OSS developer appears to be schizophrenic.

      Either that or they just don't care.

      You make it sound as though that's a unique issue. There are a huge number of projects which wouldn't get used if the mental health of the lead maintainer was an issue. Glibc and cdrtools are the main two that immediately come to mind. Ulrich is fairly well known for being as mad as the proverbial March hare. I don't blame him for that, though; I've felt a certain wrench in my own sanity when I've tried to patch and compile Glibc for Linux From Scratch. I can't even begin to imagine what working on that code for years on end would do to a person, but it can't be anything good. ;)

    55. Re:Linux vs. FreeBSD by Hatta · · Score: 1

      On the other hand, maybe Debian really can improve on the FreeBSD experience; apt rocks, and the Debian project does perhaps a better job than anyone of combining the disparate parts of the GNU/Linux ecosystem into a coherent operating system.

      So how good is the KFreeBSD apt repository today? Can I expect most of the packages I use under Debian proper to be available on KFreeBSD?

      --
      Give me Classic Slashdot or give me death!
    56. Re:Linux vs. FreeBSD by Bill,+Shooter+of+Bul · · Score: 1

      No. I was alluding to the fact that BSD branched off unix main line ( resulting in the trial with AT&T) where as Linux merely is an adopted child in the unix family. Its also closer to Solaris, HP-UX, and other Open group certified UNIX(TM) OS es. But compared with windows, they are both unix-y.

      --
      Well.. maybe. Or Maybe not. But Definitely not sort of.
    57. Re:Linux vs. FreeBSD by ecliptik · · Score: 1

      I've had nothing but trouble with Intel wi-fi chipsets on Linux, even with the latest kernel and firmwares it's still not stable and after a while becomes frustrating. I'm running OpoenBSD on my laptop for the sole reason of having a stable wi-fi connection.

    58. Re:Linux vs. FreeBSD by Guy+Harris · · Score: 1

      VMS also has a similar feature when you press ^T which I find myself using all the time.

      I wonder whether they picked it up from TOPS-20 - which was based on Tenex; BSD got ^T from Tenex.

      As someone who has been using FreeBSD as their desktop OS for several years now, I can't believe I didn't know about it in FreeBSD!

      It's also in the other *BSDs and in Mac OS X; I'm not sure whether all of them default to having the "status" character set to ^T, so I'm not sure whether they all default to having ^T work, but, if they don't default to setting the "status" character, you can do "stty status '^T'" to set it.

      Some of the ^T output is handled inside the tty driver, so that shows up in response to ^T even if the foreground job isn't catching SIGINFO.

    59. Re:Linux vs. FreeBSD by cbhacking · · Score: 1

      I said the Compaq uses a modified chip, not that I (or any other user) had modded it. It's the OEM chipset, it's just not exactly the same as the normal BCom chipset. I last tried to get this working about 2 years ago, and tried both ndiswrapper and the b43xx driver.

      Another laptop with a OEM Broadcom chipset (but one that I thought was un-modified) didn't work either as recently as a year ago.

      A Broadcom PCMCIA card might work in Linux, but the Atheros-based one (which was old enough I don't think it even did 802.11g) wouldn't detect when using ndiswrapper. Madwifi didn't work (and would cause the system to hang if I tried for too long). I don't think ath5k was available yet - this was probably 2.5 years ago.

      I've seen both ndiswrapper and the native Broadcom driver work exactly once, and it wasn't my hardware. I follower the other guy's steps exactly, to no avail.

      Intel Pro Wireless, so long as the chipset isn't newer than the distro I'm trying to use it with, is the only one that's worked reliably for me (and even then it doesn't always work out of the box, depending on how seriously the distro in question is about avoiding non-free binary blobs). These days, I am just very careful about what WiFi a new computer uses.

      --
      There's no place I could be, since I've found Serenity...
    60. Re:Linux vs. FreeBSD by xiong.chiamiov · · Score: 1

      FreeBSD for Linux Users. It's the clearest explanation I've ever found.

    61. Re:Linux vs. FreeBSD by soup4you2 · · Score: 1

      "FreeBSD you can press ^T while cp is copying some huge file, and this will send SIGINFO to cp, causing it to print a progress report to STDERR. Handy."

      I did not know about this. Thanks for sharing :)

    62. Re:Linux vs. FreeBSD by soup4you2 · · Score: 1

      I really love the FreeBSD project, and it's defiantly an outstanding operating system, over the years though i have gotten really tired of the constant compiling of everything. It's really not efficient when your dealing with over 100 servers.

    63. Re:Linux vs. FreeBSD by True+Grit · · Score: 1

      You make it sound as though that's a unique issue.

      ...

      Glibc and cdrtools are the main two [counter-examples] that immediately come to mind.

      Both of which have since been been forked(1)(2) precisely because of their 'schizophrenic'(3) maintainers.

      1 : depending on your definition of what a fork is.

      2 : 'glibc' -> 'eglibc' , 'cdrtools' -> ( 'cdrkit' | 'dvd+rw-tools' )

      3 : or whatever the actual diagnosis turns out to be.

    64. Re:Linux vs. FreeBSD by True+Grit · · Score: 1

      But now there's this new kid in town in Linux, PulseAudio.

      Which isn't actually necessary for most 'typical' users (whether their sound driver is ALSA or OSS).

      Its talked about a lot only because of the decision of one popular distro to make it default on their system, which given the number of problems since reported, may have been premature, but I don't know the whole story as I don't use that distro.

    65. Re:Linux vs. FreeBSD by k8to · · Score: 1

      That's what I said. The original implementation of UNIX. UNIX is a spec, not an implementation.

      --
      -josh
    66. Re:Linux vs. FreeBSD by Anonymous Coward · · Score: 0

      Are you serious?

      http://www.freebsdnews.net/2007/09/06/os-users-mugshots/

      http://4.bp.blogspot.com/_Cm_GpjzZWBY/Sggtto8TbZI/AAAAAAAAAD0/o43RyRyU0U8/s1600-h/Linux+vs+FreeBSD.jpeg

  4. We have to put an end to their monopoly of awesome by SafeMode · · Score: 5, Funny

    It's nigh time that we look at the sheer scrumtrulesence of Debian and realize that it's reign of End All Be All of OS's must be curtailed and possibly even put an end to. No single OS should be this awesome. And we can no longer ignore the fact that it is.

  5. Awesome! But... by Anonymous Coward · · Score: 4, Insightful

    This is a really cool thing, except that I wonder how much this is going to be used? I'm sure there's a group of people who will be interested in this, and it might be a great stepping stone for those that want to move to/from FreeBSD to/from Linux, but a lot of the FreeBSD community is heavily focused on the fact that FreeBSD is developed as a complete OS. The userland and the kernel are developed by the same people and integrated. So while this is exciting, I'm not sure how much interest you're going to get from the FreeBSD community. Similarly, a lot of the Linux people who use Debian don't think of using Debian but think of using Linux, Debian just happens to be the distribution they choose.

    Now, what may be interesting that'll come out of this is packages with better FreeBSD compatibility. That is something I look forward to.

    1. Re:Awesome! But... by characterZer0 · · Score: 1

      I have been using Debian for years. I have seen comparisons that indicate that I would get better performance on my mail server with FreeBSD, but I am not going to make that leap. However, if I get some or all of the performance advantage using Debian/kFreeBSD, I will try it.

      --
      Go green: turn off your refrigerator.
    2. Re:Awesome! But... by SafeMode · · Score: 1, Informative

      and that's the problem with the terminology of calling Linux Linux and not Gnu/linux and the like. While being pedantic, it separates the fact that what people are familiar with in dealing with linux, has little to do with the kernel and mostly to do with the utilities and apps that run on top of it. With debian, you dont change the utilities and apps , so it's the same OS no matter what kernel you're using. The kernel just makes subtle changes in what those apps can do (and possibly let you run different apps that aren't compatible between kernels). Start thinking of "Linux" in the same light as we think "Unix". It's not an OS, but it's a superset of OS's. Distributions aren't simply repackaging an OS, they are different OS's. Some limited to just one kernel, some limited by the gui even. We need to stop visualizing them as all revisions of the same pseudo OS.

      So no, For the most part, i dont think linux users who use Debian Linux will find themselves in an alien environment with Debian kFreeBSD. And i dont think it would be a stepping stone to using FreeBSD. I think it just gives Debian users (which should be considered it's own OS) a different backend that may do certain things better than the more familiar Linux backend.

      With Debian crossing the kernel boundaries, i think the idea that the OS is the kernel will be even more obviously misguided. It's the Debian OS. It really can't be described as anything else. Debian refuses your limitations. Unless it has a non-free license.

    3. Re:Awesome! But... by at_slashdot · · Score: 2, Interesting

      While packages with better FreeBSD compatibility are nice, I wonder if getting more critical release bugs won't slow down Debian releases even more. If it's all positive development then is nice, but I'd like to know the downside of things too in order to tell if it's a good or a bad decision.

      --
      "It is our choices, Harry, that show what we truly are, far more than our abilities." -- Prof. Dumbledore
    4. Re:Awesome! But... by petermgreen · · Score: 1

      The users I see are those who want stuff that the freebsd kernel has but don't want to learn a completely new userland and package management system*

      Now, what may be interesting that'll come out of this is packages with better FreeBSD compatibility.
      I doubt there will be much of that. It's only the freebsd kernel that Debian are using. The C library and toolchain are still GNU.

      *personally my only experiance with the freebsd "ports tree" has been bad. It seems that once a system had been installed for a while that you couldn't use the ports tree that came with it because you couldn't get the "distfiles" and when I tried to update the ports they borked the system. I then tried to update the core system to match the ports tree but I still ended up stuck with some library issues that I couldn't fix.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    5. Re:Awesome! But... by vlm · · Score: 1

      I wonder if getting more critical release bugs won't slow down Debian releases even more.

      Unlikely, since the Debian method of handling RC bugs is to remove the package from the release if its got RC bugs. And that process seems brutally fast...

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    6. Re:Awesome! But... by petermgreen · · Score: 2, Informative

      One thing about so called "rc" bugs in debian is that they divide into a few categories.

      * Those that while technically qualifying for severity serious aren't actually bad enough for the release team to take any action.
      * Those in packages that the release team considers unimportant enough to kick out.
      * The real rc bugs, those that can't be allowed in the release but aren't in packages that can reasonably be kicked out either. These are a small minority of the so-called "rc" bugs but they are the ones that really hold up releases.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    7. Re:Awesome! But... by Anonymous Coward · · Score: 0

      With debian, you dont change the utilities and apps , so it's the same OS no matter what kernel you're using.

      Stop right there. You do not even know the history how the operating systems have be build.

      The monolithic kernel is the complete operating system. Microkernel is just a part of the operating system. Microkernel needs OS servers to build a complete operating system.

      If you switch Linux to Hurd on Debian, you switch the operating system. If you switch everything else than the Linux, you switch distribution but you maintain same release/version of the Linux operating system.

      The GNU's own operating system, called Hurd, is not is not ready. And Hurd is not a kernel, it is a complete operating system, even not functional, yet. Hurd's kernel is a microkernel called GNU Mach. Or do you want to explein how kernel has a kernel inside of it or how "operating system" (debian) would have operating system (Linux kernel) inside of it? Does your heart have a heart inside of it?

      Start thinking of "Linux" in the same light as we think "Unix". It's not an OS, but it's a superset of OS's. Distributions aren't simply repackaging an OS, they are different OS's.

      Wow, do you trust everything what marketing people say to you? The Linux kernel, is a monolithic. What makes it a complete operating system, alone (no other software is needed).

      Now you are saying that Linux 2.4.11 is different OS than Linux 2.4.12. Or if Debian packages Linux 2.4.11 it is different OS than if Novell packages Linux 2.4.11.

      You should read about operating systems history. The operating systems even exist only because they are there to help run all other software on the computer. They are the bridge between hardware for other softwares. Without the operating system, you should program all the machine code and device commands to them itself.
      All other software needs a operating system to function, like glibc, bash, zlib, Xorg, INIT and so on.

      The problem is that first the operating systems were called as kernels, master programs, cores, nucleolus , supervisors and operating systems. They were monolithic kernels alone running in kernel space.
      Then later came the idea to build secure and stable OS. They sliced the OS tiny kernel what would include only most basic functions of the OS. It was called as microkernel. Then the sliced OS parts were moved to user space among other normal softwares and they are running as protected threads as supervisor mode to microkernel what controls them. Together microkernel + servers build up a complete operating system.

      And people talks about monolithic kernels like they would talk about microkernels. And then some people are refferring OS as what they see on the screen. Problem is that people can not touch the software as they can touch the hardware.

      Like you would not call your CPU as motherboard because you have connected it to such. You would not call your motherboard as power plant because your house gets power from such.

      Or is it so that your computer is part of the power plant, because it needs a powerplant so it is Computer/Powerplant named computer?

      There are different structures for the operating systems. The operating system can be a monolithic, server-client or layered, as example.
      Even monolithic can be a modular on the binary level, but not on the architecture level, like microkernel-based operating systems are with their modules.

      People calls Hurd as kernel, just by trusting RMS speaking truth. Well, the Hurd has a kernel, called GNU Mach. The GNU mach is not a operating system, it is just a kernel. The GNU Mach + OS servers build together a OS called Hurd. And Hurd can run other GNU software like glibc, bash and so on. Difference is that Hurd is partically on kernel space and user space.

    8. Re:Awesome! But... by Bluesman · · Score: 1

      Eh, I've always wished someone would do the opposite of what Debian is doing and use the FreeBSD model for a Linux distribution. I think the ports system on FreeBSD is much better than any similar repository in the Linux world, and system configuration is simple and contained mostly in a single file.

      (And no, Gentoo is not what I had in mind.)

      --
      If moderation could change anything, it would be illegal.
    9. Re:Awesome! But... by Kjella · · Score: 1

      Only because they tried very hard to kill themselves by never getting a release out and finally had to lower their standards and realize it's better to release at some point than not at all. The mantra was "use testing" which was very stable but meant something could break or applications could change drastically any time you did an update. Some of the Ubuntu releases has been rather hairy but they never broke anything between releases, if it wasn't already broken in the release. Honestly, it's better with the devil you know than the devil you don't.

      --
      Live today, because you never know what tomorrow brings
    10. Re:Awesome! But... by Hatta · · Score: 1

      I love debian because apt is awesome and the repositories are enormous. I also love the GNU userland. I really don't care what my kernel is.

      I'm definitely going to give this a try on my next computer. I'd like to try it out now, but I'm heavily invested in a luks/lvm/raid configuration, so I'm stuck for now.

      --
      Give me Classic Slashdot or give me death!
  6. truly a testiment to OSS (not entirely F/OSS) by Anonymous Coward · · Score: 0

    Kernel choices is a good thing. Good job fellas.

  7. Wow by exa · · Score: 1

    Very glad to hear progress on the freebsd front!

    --
    --exa--
  8. Scrumtrulesence by sexconker · · Score: 5, Funny

    Scrumtrulesence is a perfectly cromulent word.

    1. Re:Scrumtrulesence by MonsterTrimble · · Score: 1

      If I had mod points my friend.

      --
      I call it 'The Aristocrats'
    2. Re:Scrumtrulesence by Anonymous Coward · · Score: 0

      Now I'm all discombobulated !

    3. Re:Scrumtrulesence by mpdolan37 · · Score: 1

      scrumdiddlyumptious!

      --
      Facts are useless, they can be used to prove anything.
    4. Re:Scrumtrulesence by sexconker · · Score: 1

      I would have gone with Dale Gribble's "scrumdiddlyGARlicicious".

      (They were fishing with garlic flavored bait, he caught a fish and held it up, moving it's mouth while he said, to taunt Hank. Hank insisted on using earthworms for bait, and was failing.)

  9. I still dont see the point by nurb432 · · Score: 2, Interesting

    if you want FreeBSD, use it.. If you want Linux, use it instead.

    What real advantage is there in mixing things like this? And no im not trolling, i really don't understand the point here.

    --
    ---- Booth was a patriot ----
    1. Re:I still dont see the point by QUILz · · Score: 1

      APT, maybe? I'm sure there are plenty of other reasons to combine Debian's userland with the FreeBSD kernel.

    2. Re:I still dont see the point by Darkness404 · · Score: 1

      About the only thing that I can think of is that a Debian admin good with the few debian-only tools like APT feels more at home. I really don't get why Debian would do this though because of the fact that it will take away from its primary user base (Linux users) to help fill a possible niche of users (KFreeBSD users) that are small in number.

      --
      Taxation is legalized theft, no more, no less.
    3. Re:I still dont see the point by mcgrew · · Score: 1

      Debian is in effect raising BSD from the dead. IMO it's a good thing, the more OSes there are, the better.

    4. Re:I still dont see the point by bzzfzz · · Score: 1
      Well, the whole point behind a distro is that everything -- the kernel, a bunch of loadable kernel modules, the libraries, the userland, any X Window system you might choose to run, and various other stuff -- is assembled and tested by others, and some sort of effort is made to evaluate interdependencies and resolve them in a sensible way. It isn't a trivial job.

      Sure, you could take an older Debian distro or one of the other distros, remove Linux, add FreeBSD, and you might get it to work. But I can guarantee that at least *some* of the packages will be broken. What Debian are doing is treating those problems as release-critical, so that they will fix them either on the kernel side or in the affected packages themselves before issuing the distro.

    5. Re:I still dont see the point by spiffmastercow · · Score: 1

      The point is that Debian is not supposed to be dependent on a specific kernel, but essentially it has been until now because HURD was such a failure.

    6. Re:I still dont see the point by nurb432 · · Score: 1

      Well, the whole point behind a distro is that everything -- the kernel, a bunch of loadable kernel modules, the libraries, the userland, any X Window system you might choose to run, and various other stuff -- is assembled and tested by others, and some sort of effort is made to evaluate interdependencies and resolve them in a sensible way. It isn't a trivial job.

      And this happens now on BOTH sides of the fence, so mixing this improves the situation how? I see it making it worse if anything.

      --
      ---- Booth was a patriot ----
    7. Re:I still dont see the point by PCM2 · · Score: 3, Interesting

      if you want FreeBSD, use it.. If you want Linux, use it instead.

      Yeah, that's what Debian said.

      --
      Breakfast served all day!
    8. Re:I still dont see the point by livingdeadline · · Score: 1

      You'd get a dpkg/apt-based distro with awesomeness that can't be ported to Linux due to GPL constraints (DTrace, ZFS). These two tools alone bring an insane amount of features any sysadmin would love.

      An apt-based system with some of the modifications and default configurations Debian provides is in lots of situations just that much simpler to maintain than a BSD userland. It's way easy to use a rolling release in a Debian environment or upgrade between stable releases with a couple of commands.

    9. Re:I still dont see the point by nurb432 · · Score: 2, Informative

      So just take stock freebsd, rename pkg_add to apt-get and you are done :)

      Ok, im joking of course but you see the point im sure.

      --
      ---- Booth was a patriot ----
    10. Re:I still dont see the point by Anonymous Coward · · Score: 0

      if you want FreeBSD, use it.. If you want Linux, use it instead.

      What real advantage is there in mixing things like this?

      "Mixing" things like what? Most of the software used in Debian e.g. GNU tools and X Window system isn't aimed at Linux in particular.

    11. Re:I still dont see the point by FlyingBishop · · Score: 1

      Debian is designed with absolute stability in mind. Everything should conform precisely to spec and never crash. If they can use multiple full-featured kernels, they have something to test against, and help weed out bugs on the kernel side that might otherwise not have been found.

      Ports are valuable if only for the stability they add (and they're even more valuable if people use them so that they're deployed widely and help root out even more bugs.)

    12. Re:I still dont see the point by acey72 · · Score: 5, Insightful

      Debian is in effect raising BSD from the dead. IMO it's a good thing, the more OSes there are, the better.

      If being made into the un-dead means becoming more like GNU/Linux, I'd rather just keep me and my demonic servers six feet under please.

    13. Re:I still dont see the point by Daniel_Staal · · Score: 1

      One good one is to test your assumptions: Making sure software ports cleanly over a couple of different (although very similar) underpinnings means you are likely to find bugs that you might otherwise overlook. But which would likely bite you later.

      Ok, so that's a reason for Debian to do it, not necessarily for you to use it, but it's a valid reason non the less.

      Also, the Linux kernel and the FreeBSD kernel are tuned differently. It's quite possible that one would be better under some loads than the other. This gives you the choice.

      --
      'Sensible' is a curse word.
    14. Re:I still dont see the point by vlm · · Score: 4, Insightful

      And this happens now on BOTH sides of the fence, so mixing this improves the situation how? I see it making it worse if anything.

      Software that compiles and installs on BOTH BSD and Linux, has not been all that unusual since, perhaps, 1991-1992.

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    15. Re:I still dont see the point by Korin43 · · Score: 1

      Because Debian is not Linux and Linux is not an operating system. Debian is an operating system that uses either the Linux kernel, FreeBSD kernel, or HURD. Don't confuse the fact that the Linux version is most popular with the idea that it's the only one. The Debian project doesn't produce any kernel, so using the FreeBSD kernel is not any more of a "mix" than using the Linux kernel. As to why you'd want it: apt.

    16. Re:I still dont see the point by vlm · · Score: 5, Insightful

      I really don't get why Debian would do this though because of the fact that it will take away from its primary user base (Linux users) to help fill a possible niche of users (KFreeBSD users) that are small in number.

      That type of question makes sense when asked about Microsoft, but doesn't even make sense when discussing Debian. "Why would Debian do this" is like a zen koan, until you're enlightened it makes no sense, or when it makes sense it means you're enlightened.

      Debian developers do what they want to do, within the legal framework and societal tolerance. If the guys doing the port, feel like doing the port, they do the port, and we get a "testing" quality port, and if its good enough, TPtB declare it a release-quality architecture and we eventually get a "stable" quality release. There is no "Debian" borg style hive mind, or if one does exist, instead of "the three laws of robotics" or "the ferengi laws of acquisition", the hive mind has the social contract and the DFSG. There is no top down militaristic business command structure. Very few people in Debian with positions of power have the "wikipedia" attitude of "I'm not personally interested in your work, so I shall gleefully destroy it while laughing, ha ha ha".

      In summary, they felt like doing it, they did it, some folks in positions of power acknowledge it. Its the same deal for all Debian packages, this port is not getting "special" treatment.

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    17. Re:I still dont see the point by rbanffy · · Score: 1

      ZFS for those who are suspicious of FUSE and don't want to give up APT.

    18. Re:I still dont see the point by rbanffy · · Score: 1

      I think you pretty much nailed it: it's about making the userland as independent as possible from the kernel and, in the process, to find any code that makes improper assumptions about the computer it's running on.

      You find all sorts of subtle bugs when porting to a different architecture. It must be the same with kernels.

    19. Re:I still dont see the point by notamisfit · · Score: 1

      It proves/disproves/doesn'treallydoeither some pedantic point about what is or isn't an "operating system."

      --
      Jesus is coming -- look busy!
    20. Re:I still dont see the point by mdemonic · · Score: 1

      Debian is in effect raising BSD from the dead.

      How do you think the number of kFreeBSD users compare to the number of people using otherBSD?
      How do you envision this ratio will change over the years?

    21. Re:I still dont see the point by petrus4 · · Score: 1

      That type of question makes sense when asked about Microsoft, but doesn't even make sense when discussing Debian. "Why would Debian do this" is like a zen koan, until you're enlightened it makes no sense, or when it makes sense it means you're enlightened.

      Why on Earth is this pretentious garbage modded +5, Insightful?

      As for why Debian are starting to use different kernels, it's fairly simple.

      A large number of the Debian developers are die hard Stallmanite/FSF/GNU fanatics. FreeBSD (as one example) is a system that has both kernel and userland. Using the kernel of other operating systems which already have their own userland, is basically an attempt to render the userlands of said other systems irrelevant.

      GNU/kFreeBSD is an abomination, and I'm not the only FreeBSD user who thinks so. Most other FreeBSD users I've spoken to want to get (more) free of having to rely on FSF software, not less. The FSF are generally considered a plague by BSD users, and rightly so.

    22. Re:I still dont see the point by Atriqus · · Score: 1
      I can't tell what part of that was more of an insult to the BSD license: the paranoia of Debian trying to undermine FreeBSD, giving a shit about what another project is doing with the code when they are following the license terms set by the author, or the insults thrown about at a lot of work that was far from trivial.

      Most other FreeBSD users I've spoken to want to get (more) free of having to rely on FSF software, not less.

      Great, but Debian has nothing to do with this. If you remember correctly, FreeBSD is the full OS. So if you and the other FreeBSD users that make up that brain trust are annoyed about FSF software in FreeBSD, then complain to the FreeBSD devs because they're the ones putting it in there.

      --
      Hey, look! It's Bono's brother.
    23. Re:I still dont see the point by Anonymous Coward · · Score: 0

      A very good explanation of how Debian works, thanks. I think the Borg is actually a good comparison (usually without the destruction part): "We will add your biological and technological distinctiveness to our own. Resistance is futile."

  10. debian expectation from a "debian" user by Anonymous Coward · · Score: 0

    firstly, i use debian only for that it release "stable updates". at work i use slackware on servers. i hate *rpm files ;) and most of server related support software is linked to redhat (shit) and suse (double shit).

    some thoughts:
    1) linux have more hardware support. i've an amd64 processor and nvidia goes only on freebsd on i386 platform. today linux is an "official" platform. not freebsd. **user people** likes hardware compatibility, not license power :)
    2) why i continue to use debian if slackware arrive to 13Â release? and, checking startup speed it's better?
    3) why someone tell that zfs is usefull for SANs works, when everybody use hardware-based SANs? like EMC or NEXSAN or others (that are embedded windows (EMC :P ) or embedded rtos/linux?

    waiting...

    1. Re:debian expectation from a "debian" user by Anonymous Coward · · Score: 0

      >> 3) why someone tell that zfs is usefull for SANs works, when everybody use hardware-based SANs? like EMC or NEXSAN or others (that are embedded windows (EMC :P ) or embedded rtos/linux?

      False. Many of these high end SAN systems run variants of FreeBSD.

    2. Re:debian expectation from a "debian" user by Anonymous Coward · · Score: 0

      30^6$ are the cost of freebsd? eh? or it's only support?
      Qlogic Freebsd drivers are written, or not?

      bye America. your time is done ;P

    3. Re:debian expectation from a "debian" user by livingdeadline · · Score: 2, Interesting

      ZFS with snapshotting and stuff is usable in any file system.. even root ones. True, ZFS is a memory hog, but man, imagine a root file system where you could have file system provided revision control for *every* file...

    4. Re:debian expectation from a "debian" user by ducomputergeek · · Score: 1

      Actually, that's saved our rears a couple times as we run ZFS on our development machine (freeBSD) and have accidentally over written files. Being able to recover the old ones saved at least a day's worth of work once.

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
    5. Re:debian expectation from a "debian" user by marcosdumay · · Score: 1

      1) That is a problem for FreeBSD, but the overall driver situation doesn't really mater, what matters is whether the OS has the drivers you need. FreeBSD has the drivers lots of people need, so lots of people are able to use it without any problem.

      2) If you choosed Debian because of startup time, you choosed it wrong. Debian is very easy to maintain and upgrade, and has a huge collection of sotfware ready to use. There are other factors here, but startup time isn't one of them.

  11. Yes! by Anonymous Coward · · Score: 0

    I think I just wet myself with excitement!

    Yes. Yes I did.

    1. Re:Yes! by kclittle · · Score: 1

      You are easily excited...

      Yes. Yes you are.

      --
      Generally, bash is superior to python in those environments where python is not installed.
  12. Start the RMS timer... by Anonymous Coward · · Score: 2, Funny

    How long before he tries to shoe "GNU" into the FreeBSD name?

    1. Re:Start the RMS timer... by Anonymous Coward · · Score: 0

      No need, it's already called Debian GNU/kFreeBSD.

    2. Re:Start the RMS timer... by Anonymous Coward · · Score: 0

      http://www.gnu.org/gnu/gnu-linux-faq.html#bsd

    3. Re:Start the RMS timer... by TheRaven64 · · Score: 3, Informative

      It's part of the name, and it makes sense. This is a GNU userland (importantly, a GNU libc) with a FreeBSD kernel. You probably can't run FreeBSD binaries on it, because they will expect different libraries. You can run any GNU programs that don't make system calls directly and you can run most GNU/Linux binaries with the Linux kernel ABI module loaded.

      --
      I am TheRaven on Soylent News
    4. Re:Start the RMS timer... by Anonymous Coward · · Score: 0

      seems you didn't read the GNU/Linux FAQ

  13. What's Next ? by Anonymous Coward · · Score: 0

    Running the OpenSolaris system on the FreeBSD kernel ?

    1. Re:What's Next ? by gujo-odori · · Score: 2, Insightful

      I'll take a pass on that one, thanks :p

      But a Debian system with an OpenSolaris kernel? Now *that* would be nice!

      Of course, if KDE should someday work as well on OpenSolaris as GNOME does (including Timeslider integration into Dolphin and/or Konqueror), then it might be just as well to go with OpenSolaris itself, although I'd still prefer the APT to OSOL's packaging system. Plus, of course, the number of packages in the Debian repositories completely dwarfs what is available for OSOL.

      That said, I like OSOL so much that if KDE _were_ at the same support level as GNOME, I'd likely move from Kubuntu to OSOL now.

    2. Re:What's Next ? by Knitebane · · Score: 2, Informative
      Here you go....

      Nexenta

      --
      "...history will look upon the act of depriving a whole nation of arms, as the blackest." --Ghandi
    3. Re:What's Next ? by gujo-odori · · Score: 1

      I was going to just bypass that because Nexenta Core itself is not - AFAICT - something that's ready to use as a system. It's what the name implies - something ready to be used as the core of a system. But, since I hadn't visited the Nexenta site in at least 6 months and probably longer - I dropped by. Turns out there's a desktop distro now, called StormOS, based on Nextenta Core: http://www.stormos.org/ that I'm going to look into.

      Thanks!

  14. Will Debian move to BSD permanently? by MobyDisk · · Score: 2, Interesting

    Is this a stepping stone to Debian moving from Linux to BSD permanently? I'm trying to figure out if the FreeBSD licenses are more compatible with the Debian philosphy, or less.

    1. Re:Will Debian move to BSD permanently? by IgnoramusMaximus · · Score: 3, Insightful

      Is this a stepping stone to Debian moving from Linux to BSD permanently?

      Unless the elected leaders of Debian all go insane at the same time, not very likely.

      I'm trying to figure out if the FreeBSD licenses are more compatible with the Debian philosphy, or less.

      Far less. If Debian were to ever go BSD as its primary license (see point one), somewhere in the vicinity of 90% of its contributors would leave, probably to start a new GPL-ed distribution.

      Inclusion of the BSD kernel is not the same as an adoption of the BSD philosophy, as the kernel is an interchangeable component of the whole, very much like the much maligned Hurd is, of which there is also a Debian-based experimental distro.

    2. Re:Will Debian move to BSD permanently? by Anonymous Coward · · Score: 0

      No, Debian is not moving from one kernel to another. They are offering a choice between two kernels. Most people will continue to use Linux, but some niche of users' will prefer BSD. Both will be maintained permanently.

    3. Re:Will Debian move to BSD permanently? by vlm · · Score: 1

      Is this a stepping stone to Debian moving from Linux to BSD permanently? I'm trying to figure out if the FreeBSD licenses are more compatible with the Debian philosphy, or less.

      Essentially boils down to, is GPL or BSD closer to the DFSG and the Debian Social Contract. BSD allows closing the source off, thus it permits (but certainly does not require) behavior that is pretty much the opposite of social contract #1, #2, and #4, and DFSG #1, #2, #3. The BSD license does not by any means require a third party to participate in anti-social behavior.

      GPL is kind of like a nerf gun, you can't really hurt the community with it as long as you follow GPL rules, but BSD is more like a real gun, in that its possible to derive software using it to screw over the community, or, perhaps not, its your choice.

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    4. Re:Will Debian move to BSD permanently? by MonsterTrimble · · Score: 1

      As I recall and gleen from both Debian & FreeBSD, Debian will always redistribute it's changes to the community. However, this is a construct of GPL also. BSD does not require redistribution of the source code, only that copyrights are still in the binary & source code IF they are redistributed. That's why OSX can be based off BSD and not release the sourcecode. Long story short, the BSD licence is neither more nor less.

      Off Topic side question: Is OSX's GUI GNOME? And if so, how does Apple get away with not release the source?

      --
      I call it 'The Aristocrats'
    5. Re:Will Debian move to BSD permanently? by cbhacking · · Score: 2, Insightful

      Permanently seems unlikely, especially since Linux still has more users and developers (AFAIK) than FreeBSD. That said, if they maintain FreeBSD as a supported kernel, then more of the software packages that are normally run on Linux will be tested and supported on FreeBSD. This is a good thing. One problem that *BSD has faced historically is that a lot of software isn't actually written for a UNIX-like OS (i.e. written to the POSIX API) but is instead written for Linux specifically. Not only does that make it less portable, it makes it less maintainable - Linux sometimes dumps things when it discovers a better alternative to its current way of doing this. Coding against the common API puts you at less risk of finding the API you use getting deprecated.

      --
      There's no place I could be, since I've found Serenity...
    6. Re:Will Debian move to BSD permanently? by rjstanford · · Score: 1

      Off Topic side question: Is OSX's GUI GNOME?

      Nope... its Aqua.

      And if so, how does Apple get away with not release the source?

      See (1) above.

      --
      You're special forces then? That's great! I just love your olympics!
    7. Re:Will Debian move to BSD permanently? by glwtta · · Score: 1

      Unless the elected leaders of Debian all go insane at the same time, not very likely.

      Given their history, that scenario is not all that unlikely. (Not that I think they would move to BSD...)

      --
      sic transit gloria mundi
    8. Re:Will Debian move to BSD permanently? by Anonymous Coward · · Score: 0

      Hurd is not a kernel. It is the GNU's own operating system.
      Hurd's kernel is GNU Mach what is a microkernel. Or do you want to call it as GNU/Mach/Hurd?

    9. Re:Will Debian move to BSD permanently? by IgnoramusMaximus · · Score: 1

      You mean like they did when they got fed up with Debian and migrated to Ubuntu?

      This is nonsense. Most of Ubuntu packages are unaltered Debian packages. While Ubuntu has a different target audience, it depends in its entirety on the good will of the Debian project.

    10. Re:Will Debian move to BSD permanently? by IgnoramusMaximus · · Score: 1

      Hurd's kernel is GNU Mach what is a microkernel

      You are confused. Hurd is a kernel which uses the Mach micro-kernel as its core. The whole idea of a micro-kernel is that it is a minimalistic foundation upon which you build the actual kernel environment. Similarly, the BSD kernel has a Mach micro-kernel elements, like the virtual memory management system borrowed from Mach. MAC OS X has a Mach/BSD hybrid environment. Etc and so on.

    11. Re:Will Debian move to BSD permanently? by tokul · · Score: 1

      Is this a stepping stone to Debian moving from Linux to BSD permanently? I'm trying to figure out if the FreeBSD licenses are more compatible with the Debian philosphy, or less.

      It is only kernel. Debian Sid also has Hurd and NetBSD kernels.

    12. Re:Will Debian move to BSD permanently? by Anonymous Coward · · Score: 0

      (Not that I think they would move to BSD...)

      Debian Windows? Cool!

  15. Re:I bet the HURD team is turning in their graves. by rickb928 · · Score: 1, Offtopic

    Wait a minit. HURD is just another Plan 9.

    You WANT to do that? BAHAHAHAHAHAHA!

    Seriously, HURD is pointless. If I want more CPUs, I buy a bigger chip. If I want more cores, I code it for GPGPU and get more cores on more video cards. If I want more drives, I get SASI or some other expensive SAN technology.

    HURD doesn't solve anything. Of course, that makes it a great open source project. And some day, somewhere, someone will make something useful out of it.

    Open Source Wins! Yaaaaaah!

    Whatever.

    --
    deleting the extra space after periods so i can stay relevant, yeah.
  16. How long... by NNKK · · Score: 1

    ...before some vocal peanut-gallery subset of Debian developers decide this is a horrible, undemocratic idea and must be put to a vote of everyone who has ever contributed a toenail to the distribution?

  17. test? by Anonymous Coward · · Score: 2, Insightful

    You do non-production evaluation of config changes, don't you?

    C'mon, no professional just pokes "apt-get update" into the root shell on a live production server. That's just asking for hilarity, fail, and unemployment.

    can you tell me more about the potential applications of this "test machine" idea? i've been asking for a test machine for 7 months and my predecessor for the 8 months before me, but since we've had no failures, who can find the money?

    1. Re:test? by vlm · · Score: 2, Insightful

      can you tell me more about the potential applications of this "test machine" idea? i've been asking for a test machine for 7 months and my predecessor for the 8 months before me, but since we've had no failures, who can find the money?

      apt-cache search xen

      Its free.

      Also, lets be realistic here, my test box is a 500 MHz AMD-K6 wiht 384 megs ram from roughly the mid 90s... probably 99.99% of testing only requires verification that it works, not that it works at "full speed".

      --
      "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
    2. Re:test? by TheRaven64 · · Score: 1

      Install VirtualBox on your laptop and duplicate the main server's configuration in a VM.

      --
      I am TheRaven on Soylent News
  18. GNU's not by fm6 · · Score: 2, Insightful

    There's a religious tone to your answer. It assumes the question was: "does it run the Linux [kernel]?" But outside the RMS fan club, "Linux" is the name of the OS, not the kernel. So the guy was really asking "does it run the Linux [operating system]?" Hence the "funny" upmods.

    It just occurred to me, that if you're going to quibble about the synechdoche usage of "Linux", then GNU/Linux is even worse, because that term implies that that one OS somehow completely incorporates the other. But you couldn't really incorporate the GNU operating system in anything, because the stupid thing still isn't done yet. (After 25 years! [Insert Duke Nukem or Harlan Ellison joke here]) What's included in Linux is not GNU, but the libraries and utilities that were originally meant to be part of GNU. So really, it should be "Unix-like OS with Linux kernel, GNU excerpts, and some other stuff", or UOWLKGEASOS for short.

    But your post does answer one important real-world question, one that isn't answered on the KFreeBSD site: what's the darn thing for? I guess the answer is, "So you can run both BSD and Linux (GNU/Linux? UOWLKGEASOS?) binaries on a single system."

    Except I still don't see the point. Is there any software for one system that's never been ported to the other?

    1. Re:GNU's not by incripshin · · Score: 1

      For macrokernels, the OS is the kernel. At most, you can say that the functions in libc like wait() or stat() are part of the OS, but that's it (and it isn't saying much, those functions are a thin interface to the real system calls). For systems with exokernels, the OS is the union of the kernel and the library OS. I don't know much about microkernels, but the kernel is only a part of the OS. An OS' job is to manage resources and provide protection to processes. GNU doesn't do this.

    2. Re:GNU's not by synthespian · · Score: 1

      LOL.

      "UOWLKGEASOS is Not Linux."

      Hmmm. Didn't work.

      "ULOWFREEBSDKGEASOS" is not Linux.

      Hmmm. NOW I GET IT!

      Do you see, people, why Stallman calls Linux his GNU operating system?

      --
      Main difference between the BSD license and the GPL license: one is from California and the other is from Massachusetts
    3. Re:GNU's not by fm6 · · Score: 1

      There are many definitions of "Operating System". (Though yours has to be one of the most convoluted and weird.) But when people say "Linux" or "Unix" they generally mean the whole ball of wax. In this context, it makes sense to use that definition.

      Especially when you're comparing these OS with GNU. Remember, "GNU" stands for "GNU's Not Unix". RMS didn't set out to create a bunch of libraries. His goal was to create a "free" Unix clone, unencumbered by AT&T's expensive and restrictive licensing terms.

    4. Re:GNU's not by incripshin · · Score: 1

      What's so convoluted about it? It manages resources (memory, devices, CPU) and provides protection (resource sharing, virtual memory). That's what an OS is. Take even an introductory course in operating systems and they'll tell you that. With macrokernels like Linux or BSD, the OS is the kernel. It extends out to libc to a very small degree, because libc provides the system call interface. printf() at some point may invoke the read() system call, but it is not a part of the OS. With an exokernel like Aegis (being the first of its kind), the OS is Aegis + ExOS (the library operating system).

      You're thinking of what's more properly termed a distribution, and you can make the case that GNU should be a part of the distribution name. I would never make that case, but that's beside my point: macrokernel==OS.

      Yeah, I remember GNU Hurd. Jokes aside, much of what GNU provides is extra (like bison), not to say it isn't helpful. GNU's also harmful to Unix portability: look at the GNU man page for rm and tell me what recursive option will also work in OpenBSD. It's also less free than a BSD alternative (as in freedom to do whatever you want with it, i.e. freedom).

    5. Re:GNU's not by fm6 · · Score: 1

      You have a private definition of OS that you find satisfying, but that nobody else uses. I could waste a lot of time trying to demonstrate the problems with your definition, but since nobody but you uses it, there's no reason I should care.

    6. Re:GNU's not by incripshin · · Score: 1

      For this reason, computers are equipped with a layer of software called the operating system, whose job is to manage all these devices and provide user programs with a simpler interface to the hardware.

      Most computer users have had some experience with an operating system, but it is difficult to pin down precisely what an operating system is. Part of the problem is that the operating systems perform two basically unrelated functions, extending the machine and managing resources...

      -Chapter one, Modern Operating Systems 2nd ed, Andrew Tanenbaum

      Well, me and Andrew Tanenbaum anyway.

    7. Re:GNU's not by agnosticnixie · · Score: 1

      No, the only reason you should bring up Andy Tanenbaum in a discussion about GNU and Linux is to remind everyone that microkernels are best and that when HURD is done, we will finally have GNU and not need that not invented here eyesore called the linux kernel. Or something like that.

    8. Re:GNU's not by incripshin · · Score: 1

      I will remember that for next time :). A microkernel might be kind of cool, but I will never use a GPL3 OS.

    9. Re:GNU's not by agnosticnixie · · Score: 1

      Heh, I've been considering getting Minix running in VirtualBox for fun (and because I have the book for v3) because I was curious about microkernels, but it kind of amuses me how the debian GNU hardliners are sometimes about Hurd, and so you get a bunch of headbangy nonsense (a microkernel powers a bigger kernel that is not at all an OS, ever, because we'd have to admit embedded linux is an OS) that seems like it was written on drugs (fun fact, that nonsense comes up three times in the lower comments).

      That and you get license wars, which get funny when you consider they're using BSD (which they don't like for one silly reason) because, hm, Linux doesn't want GPL3 (which they like half for the same reason they didn't like BSD) and they're mostly just pining for the second coming of Stallman ;)

    10. Re:GNU's not by Anonymous Coward · · Score: 0

      Tuo add something more

      Next comes a layer of primitive software that directly controls these devices and provides a cleaner interface to the next layer.

      This software, called microprogram, is usually located in read-only memory. It is actually a interpreter, fetching the machine language instuctions such as ADD, MOVE, and JUMP, and carrying them out as series of little steps. To carry out an ADD instuction, for example, the microprogram must determine where the numbers to be added are located, fetch them, add them, and store the result somewhere. The set of insructions that the microprogram interpres defines the machine language. Which is not really part of the hard machine at all, but computer manufacturers always describe it in their manuals as such, so many people think of it as being the real "machine." On some machines the microprogram is implented in hardware, and is not really a distinct layer.

      The machine language typically has betweeon 50 and 300 instcutions, mostly for moving data around the machine, doing arithmetic, comparing values. In this layer, the input/output devices are controlled by loading values into special device registers. For example, a disk can be commaneded to read by loading the values of the dis address, main memory address, byte count, and directions (READ or WRITE) in its registers. In practice, many more parameters are needed, and the status returned by the drive after an operation is highly complex. Furthemore, for many I/O devices, timing plays an important role in programming.

      A major function of the operating system is to hide all this complexity and give the programmer a more convenient set of insructions to work with. For example, READ BLOCK FROM FILE is conceptually simpler than having to worry about the details of moving disk heads, waiting for them to settle down, and so on.

      On top of the operating system is the rest of the system software. Here we find the command interpreter (shell), compilers, editors and similar application independent programs. It is important to realize that these programs are definitely not part of the operating system, even though they are typically supplied by the computer manufacturer. This is crucial, but subtle, point. The operating system is that portion of the software that runs in kernel mode or supervisor mode. It is protected from user tampering by the hardware (ignoring for the moment some of the older microprocessors that do not have hardware protections at all). Compilers and editors run in user mode. if a user does not like a particular compiler, he is free to write his own if he so chooses; he is not free to write his own disk interrupt handler, which is part of the operating system and is normally protected by hardware against attempts by user to modify it.

      Finally, above the system programs come the application programs. These programs are written by the user to solve their particular problems, such as commercial data processing, engineering calculations, or game playing.

      5.[Banking system][Airline reservation][Adventure games]
      4.[Compilers][Editors][Command interpreter]
      3.[Operating system]
      2.[Machine language]
      1.[Microprogrammin]
      0.[ Physical devices]

      -Chapter one, Modern Operating Systems 2nd ed. Andrew Tanenbaum

      At least one basic books suggested to read before someone comes to say that Linux kernel is not the Linux operating system. People should only use "Linux" to refer Linux itself. What is a monolithic kernel == operating system. While people do wrong by referring Linux a complete software systems (different distributions).

      But people should as well understand that the operating system what has a microkernel is structured by microkernel + OS servers what are located in userspace (or in kernel space). The microkernel ran in supervisor mode, but so does all OS servers ran o

    11. Re:GNU's not by fm6 · · Score: 1

      Uh, no. That definition bears no resemblance to yours.

  19. I want GNU by Xtifr · · Score: 1

    if you want FreeBSD, use it.. If you want Linux, use it instead.

    I don't want either one, I want GNU, but unfortunately, they refuse to create a useful kernel (there's some question as to whether they're even trying, or if they're just too wrapped up in "interesting" research to ever focus on making something useful), so I'm stuck with two options for my kernel, and, while Linux is the path of least resistance for GNU fans, it's also flakey as hell, so a supported GNU/BSD system is a refreshing alternative.

    I could just go with BSD, but I hate their freakin' userspace, and, while Ports sounds good on paper, it can be (voice of experience here) a real PITA in a production environment. In my experience, Debian does the best and most reliable packaging of a wide variety of software, and a combination of Debian and GNU on a BSD kernel sounds a hell of a lot like my dream environment.

    1. Re:I want GNU by Anonymous Coward · · Score: 0

      I don't want either one, I want GNU, but unfortunately, they refuse to create a useful kernel (there's some question as to whether they're even trying, or if they're just too wrapped up in "interesting" research to ever focus on making something useful)

      It's too late for them to really start working on HURD anyway. The only solution I can see is forking Linux or one of the BSD kernels and gradually changing that into the microkernel they desire. I also doubt they actually have people qualified to do core Linux work, so they'd need to steal a couple experienced contributors / maintainers from mainline.

    2. Re:I want GNU by petrus4 · · Score: 1

      I don't want either one, I want GNU, but unfortunately, they refuse to create a useful kernel

      Can I ask why?

  20. Why FreeBSD by vga_init · · Score: 1

    FreeBSD is nice and all, but can't they elevate Mach or L4? Maybe then some work will get done in that area.

    1. Re:Why FreeBSD by Narcocide · · Score: 1

      Mach should have died with Disco, sorry.

  21. Re:I bet the HURD team is turning in their graves. by oatworm · · Score: 1

    Actually, Plan 9 was useful and implemented somewhere - if memory serves, Los Alamos set up a grid running Plan 9 back in the day. HURD, on the other hand, barely even exists as a cohesive system inside the architects' heads. Last I checked, they're still trying to find a microkernel to throw the thing on.

  22. Re:We have to put an end to their monopoly of awes by gbarules2999 · · Score: 1

    Three Awesomes for the Red Hat/Fedoras under the sky,
    Seven for the Slackware users in their halls of stone,
    Nine for Ubuntu rip-offs doomed to die,
    One for the Debian on his deb-packaged throne
    In the Land of apt-get where the repositories lie.
    One Distro to rule them all, One Distro to find them,
    One Distro to bring them all and in the kernel bind them
    In the Land of apt-get where the repositories lie.

  23. Not as insane as it sounds by cracauer · · Score: 5, Interesting

    This sounds insane to people who approach this from the usual angle. Linux has a lot more support for all the junk and semi-junk hardware out there, but some of the GNU core Unix userland is of questionable quality. All of us cursed GNU creeping featurism in the commandline utilities and GNU libc problems at some time or another. You would think people want the Linux kernel and the FreeBSD Unix userland. So why go the other way round?

    There are very specific needs being addressed by using the FreeBSD kernel inside a Debian.

    FreeBSD's ports system for third-party applications only has a devhead, and that has caused an increasing number of problems. FreeBSD has stable branches and releases for kernel, for "core Unix" userland including binutils and gcc/g++, but not for third-party applications. At the time that this was created it was great, because what we wanted at the time was a stable base system to do "server stuff" with, and the ports/applications were just for accessing the things, a light desktop that didn't do much except run xterm and emacs.

    Today, I see two main problems with what worked a few years back:

    1) those "server style" third-party applications aren't sitting flat on a Unix anymore. They are stacks of dependencies of considerable depths. It's not an apache with mod_cgi and the base perl system anymore.

    2) some third-party applications became very aggressive lately and can be unusable in their newest releases. Many people bash GNOME and/or KDE, myself my favorite target is Xorg. The Xorg server has caused the most headaches across all my Linux and FreeBSD machines in the last years.

    So, here's the trick. FreeBSD only has one branch in ports, so even if you use an older -STABLE release branch of the FreeBSD core system you still get the newest releases of third-party applications via ports. That's why my *most* stable OS (FreeBSD) had caused me the most headaches lately, because it upgrades me to the newest Xorg *first*, not last like it should.

    I don't want to distract too much from the point of this posting by giving reasons why people want the FreeBSD kernel, let's just say there are enough of us. But no matter how much you want the FreeBSD kernel, many see increasing problems with ports/applications for the reasons I gave.

    Debian provides stable branches for all applications, and that makes some people who don't generally like Linux still go "PLING!".

    In addition to all that, Debian's packaging system, and the way that it is kept working (few package screwups upgrading), the way that it integrated /etc/* file management are simply first class and blow other Linuxes out of the water, too. Debian's packaging is the best out there, I haven't seen anyone challenge that notion in a long time.

    So, very suddenly you have a demand for the FreeBSD kernel in a Debian application provision system and here we are.

    %%

    (BTW, what blows my mind for real is that FreeBSD is now partially sold based on driver availability. Because they kept their NDIS windoze driver integration system alive and maintained when Linux didn't. That is ... something, I have to think about it)

    1. Re:Not as insane as it sounds by chgros · · Score: 1

      All of us cursed GNU creeping featurism in the commandline utilities
      Funny. I tend to curse the lack of features in non-GNU utilities. For instance, grep -o.

    2. Re:Not as insane as it sounds by petrus4 · · Score: 1

      2) some third-party applications became very aggressive lately and can be unusable in their newest releases. Many people bash GNOME and/or KDE, myself my favorite target is Xorg. The Xorg server has caused the most headaches across all my Linux and FreeBSD machines in the last years.

      Yes, the isolation of the upstream X maintainers, (whoever they are) is legendary. The appropriate response, however, is not for downstream to make excuses for the X people, and for said downstream to burden themselves further with what is rightfully the X project's work. Rather, it's to give the X project's people a (metaphorical, one hopes) motivational foot up the backside.

      It is surely not an unreasonable expectation for X to work, out of the box, with the specific range of hardware that the X project has claimed compatibility with. This is same the standard which (in the past perhaps, at least) any other project has been held to.

      If the X people can't manage this, then they either need to find more developers and testers, or lower their ambitions, in terms of the amount of hardware which is claimed to be supported.

      There should not be a scenario where X.org is releasing a snarled, half-mangled pile of code which downstream is then expected to clean up and get working with their own system. Theo de Raadt mentioned that state of affairs in a BSDCan talk that he gave once, and it should not be accepted. If said scenario still exists, there needs to be some long, loud, and pointed complaining engaged in on X's mailing lists.

    3. Re:Not as insane as it sounds by Rufty · · Score: 1

      So, can I expect debian packages of BSD userland soon?

      --
      Red to red, black to black. Switch it on, but stand well back.
    4. Re:Not as insane as it sounds by cracauer · · Score: 1

      If said scenario still exists, there needs to be some long, loud, and pointed complaining engaged in on X's mailing lists.

      My attempts to even get a single new Xorg bug fixed ended up in frustration and me feeling sure that I have been lied to by the developer in question when he said it works on his installation. I went through all the trouble of doing a git version Xorg build (there's no single module-tree with a central build anymore and that Python metabuilder they have did not go through for me). Only to find that the bug behaved the same way in their devhead.

      I haven't been very active in the FreeBSD project since I now work for a company that uses Linux. The pressure from things that don't work at work is off. I'd like to dig my teeth into something that is broken for me every day. But even though Xorg fits that criteria (too well...) they pretty much excluded themselves from the list.

      And I got a hunch that I am not the only one who feels that way. Why would people who value correctness join a project that doesn't? Human nature says otherwise.

      Maybe a second fork from pre-license change XFree86, emphasizing on correctness instead of features and instead of support for hardware that long-time computer professionals don't use in the first place is the way to go.

    5. Re:Not as insane as it sounds by cracauer · · Score: 1

      So, can I expect debian packages of BSD userland soon?

      What you probably want to do is have a FreeBSD userland in a chroot.

      That way you have binaries built in Debian/FreeBSD binaries at the root. You have regular Linux binaries in /compat/linux (which isn't a chroot, it falls through to the regular filesystem). And if you want to keep your shellscripts portable and/or need something from FreeBSD userland you put that into a /var/chroot/freebsd.

    6. Re:Not as insane as it sounds by cracauer · · Score: 1

      All of us cursed GNU creeping featurism in the commandline utilities
      Funny. I tend to curse the lack of features in non-GNU utilities. For instance, grep -o.

      Well, it goes both ways. GNU dropped `tail -r` which is pretty bad since at the time many scripts were written that was the only portable way to reverse the lines in a file.

      But anyway, this was part of the "what this is not" intro. If you run this you get GNU *utils.

    7. Re:Not as insane as it sounds by petrus4 · · Score: 1

      Maybe a second fork from pre-license change XFree86, emphasizing on correctness instead of features and instead of support for hardware that long-time computer professionals don't use in the first place is the way to go.

      I could be wrong about this, but my gut tells me that a secondary patch project could help.

      When X.org do a base release, then after all the jerry rigging people have to do, to get it working with their various distributions/systems, if said people could make patches of said modifications, and then submit them to a repository for such.

      It would save probably a lot of time and duplication of effort, because if Debian or whoever else manage to fix a certain bug, anyone else who went to the repo and was having the same bug, could get that patch, and either maybe modify it if needed, or simply slot it straight in.

      Another reason why I think that could really help X, is because (as an example) I wanted to re-enable the ability to hit control-alt-backspace to kill an X session when it hangs. I have a 3d video card with bad vram, and so if I'm playing World of Warcraft sometimes, and X freezes, if I can't get out and have to reboot the machine, it means that because the hard disk was still mounted, it can damage the partition.

      The X people however decided to make that disabled by default, to supposedly help new users. Given the size of X, I'm assuming there are probably a lot of obscure options like that, which some people need, and some people don't. A patch repository could offer patches where such functionality was turned on or off by certain patches, as people wanted.

    8. Re:Not as insane as it sounds by evilviper · · Score: 2, Informative

      So, here's the trick. FreeBSD only has one branch in ports, so even if you use an older -STABLE release branch of the FreeBSD core system you still get the newest releases of third-party applications via ports. That's why my *most* stable OS (FreeBSD) had caused me the most headaches lately, because it upgrades me to the newest Xorg *first*, not last like it should.

      Ummm. WHAT?

      Why are you installing all your software from the latest CVS snapshot of FreeBSD ports?

      You're supposed to use the RELEASE-tagged ports tree for your version of the OS (the tar.gz file on every FTP mirror server, everywhere). Then, install portaudit, and only when portaudit complains about security issues in a specific version of a specific port, should you CVS-UP that single port, and build the latest and greatest.

      Even if the package in question is dependant on 20 other packages, and 20 packages in-turn depend on it, it will work just fine.

      So I'll ask again, why do you think you need to CVS-UP the entire ports tree, and build all these new, presumably incompatible and/or buggy newer version of all the software you use?

      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    9. Re:Not as insane as it sounds by Anonymous Coward · · Score: 0

      I've been using grep for a long time. In what instance would this actually be useful? Generally when I personally am grepping a file I'm not looking for grep to return the part of the string that got matched, but the entire line as there is some unknown data on it which I wouldn't be able to search for directly (think something like this crude example: ps fax |grep ps | awk '{print $1}').

    10. Re:Not as insane as it sounds by chgros · · Score: 1

      I use this when I want to get all the instances of a specific regex.
      For instance, to find all the C++ mangled names in some output, I do:
      > grep -o '_Z[A-Za-z0-9_]*'
      The power of regular expressions is that you don't need to know exactly the string you're going to match.

  24. My experience with Linux kernel by Anonymous Coward · · Score: 0

    I had to write some CPU/Mem stats code to monitor our program in realtime. On Linux It's simple a nightmere. The standard POSIX calls always returns 0. So I need to dig into top code and stated to decode the /proc files. There is simple no document at all. I have to reverse engineering the code (fortunately it's open source). Finally, I got it working. Though this is not the end of story yet. On later versions of the kernel, the way how kernel reports the proc stats changed...without any notification anywhere in any doc. I fixed this using some heuristics, To make it even worse, the behavior has change again within the stable branch of Linux kernel. WTF!!!!. Different Linux distros on different patch levels show different behaviors and there is no way to tell which one is which one. So I had to put a --if-not-working-try-this-instead-and-wish-best-of-luck config.

    For comparison, on Solaris, code written once, works fine for Solaris8, Solaris9, Solaris10 without a single change.

    Another experience with BSD. I always felt like XWindow is much more responsive than Linux. On BSD I can open a fresh gnome terminal in under 1 sec. On Linux? No way!
    Years ago I can run FreeBSD + Kde/Gnome in vmware fast like on a real machine. At the same time RedHat9 is totally unusable.

    If we put more drivers in BSD and make a Ubuntu like distro, I believe it's a much better option than the Linux mess.

    (This post is created on a Ubuntu machine. I'm not a Linux hater. I use what works.)

    1. Re:My experience with Linux kernel by Koutarou · · Score: 1

      I had to write some CPU/Mem stats code to monitor our program in realtime. On Linux It's simple a nightmere. The standard POSIX calls always returns 0. So I need to dig into top code and stated to decode the /proc files.

      Could have probably ended the post there. /proc on linux is and always has been a total cesspool.

  25. FreeBSD does run Linux (binaries) by Anonymous Coward · · Score: 0

    But, does it run Linux?

    Yes:

    http://www.freebsd.org/doc/en/books/handbook/linuxemu.html

  26. Re:That's...nice by CarpetShark · · Score: 1

    I assume they have infinite resources.

    Compared to your puny distro? Yes, they do.

  27. ZFS is a killer app by mcrbids · · Score: 1

    My first thought was... why?

    We are a Redhat/CentOS shop, so the pull of BSD kernel is pretty light, but for backups. See, we have a system that does nothing ut backup everything else. Specs are light, 2 Ghz unicore, 2 GB RAM but it has 12 hard disks in there with room for more. Managing all that space gets VERY cumersome with partitions and Ext3.

    I'd consider switching to Deb/BSD if it supported ZFS in kernel land and was VERY stable. That way, I'd get userland environment that would pretty familiar (more than BSD) and ZFS.

    Good work!

    --
    I have no problem with your religion until you decide it's reason to deprive others of the truth.
  28. Jails on Linux by Lemming+Mark · · Score: 1

    Linux containers give you jail-type functionality and - unlike VServer and OpenVZ - are in the mainline. http://lxc.sourceforge.net/

    The OpenVZ people have, I understand, been a strong force in getting the container stuff into mainline.

    There was an LSM implementing BSD-style Jails but I'm not sure what happened to it ... http://kerneltrap.org/node/3823

  29. Re:I bet the HURD team is turning in their graves. by notamisfit · · Score: 1

    Not to mention Plan 9 features like /proc or UTF-8 encoding that made it out to other operating systems. As for the HURD, I don't think anyone's found any good ideas to rip off.

    --
    Jesus is coming -- look busy!
  30. Re:I bet the HURD team is turning in their graves. by Lemming+Mark · · Score: 1

    "Wait a minit. HURD is just another Plan 9."

    I think that's doing a little of a disservice to both of them. Particularly Plan 9, which was both finished and sold as a commercial product - and is still developed actively and deployed. It's also a very nice OS, which to a geek possibly matters more than its commercial success. But HURD as I understood it had quite a nice architecture and was basically the first (to be planned, at least) GPL OS so it's certainly significant even if not successful.

  31. Go HURD go! by Anonymous Coward · · Score: 0

    If BSD can go it, so can the HURD. Go HURD go! I know you can achieve first class status!

  32. This is not how a sane admin uses apt. by brunes69 · · Score: 3, Informative

    Any sane admin has their own local apt repository, that they point all production and testing servers at. That repository has both "stable" and "testing" branches, like any apt server. All of the production servers grab off of the "stable", and all of the testing off of "testing".

    The trick is, this repository SHOULD NOT be a mirror of the actual debian repository. Rather, the "testing" of your internal server should be a mirror of the "stable" debian tree. Then, weekly or daily or whenever new debian "stable" packages come out, you update your testing boxes, and TEST the packages against your local software. If something breaks, no harm no foul - you wait till the next update.

    Once everything is tested OK, you sync those packages over to YOUR "stable" branch, and then that night all of your production servers will automatically get those updates. No fuss, no muss.

    1. Re:This is not how a sane admin uses apt. by synthespian · · Score: 1

      The need for the repository separation and testing, which you correctly point out, seems to indicate that the tool is not reliable.

      --
      Main difference between the BSD license and the GPL license: one is from California and the other is from Massachusetts
    2. Re:This is not how a sane admin uses apt. by swillden · · Score: 1

      No, it indicates that you don't understand the reason for using the two repositories.

      --
      Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
    3. Re:This is not how a sane admin uses apt. by Carewolf · · Score: 1

      Exactly, and you also need your own repository to host your own packages for custom apps, so that your custom apps will have the correct dependencies and upgrade smoothly in Debian. This completely invalidates the concern of auto-pruning. Not that I ever activate auto-pruning, it is not run automatically and requires the administrator to put it into cron himself if he wants it done.

    4. Re:This is not how a sane admin uses apt. by GWRedDragon · · Score: 1

      Once everything is tested OK, you sync those packages over to YOUR "stable" branch, and then that night all of your production servers will automatically get those updates. No fuss, no muss.

      You let your production servers perform automated updating? That seems pretty stupid. No matter how you tested before hand, things can still go wrong and you should still be monitoring after any update to make sure there aren't issues.

    5. Re:This is not how a sane admin uses apt. by silanea · · Score: 1

      Letting the production servers run updates automatically at a fixed time and monitoring them for issues are not exclusive activities. You can combine them.

      --
      Rudolf Hess edited Mein Kampf. He was the very first grammar nazi.
    6. Re:This is not how a sane admin uses apt. by GWRedDragon · · Score: 1

      Ah, right. That makes sense.

  33. TCP/IP Stack by rwade · · Score: 1

    ...Also, back in the day [FreeBSD] had a better tcp/ip stack [than Linux] and was generally more stable as a server platform...

    I have to agree with this assertion. In the first half of the decade, I ran Linux. I was not running a server platform; I was just running Linux for the desktop because I wasn't able to afford a copy of Windows. On dial-up internet, I noticed significant sluggishness. I was using an external serial modem at the time, so I could see that Netscape or lynx would just be sitting there waiting around with no modem lights blinking.

    I switched to FreeBSD and the problems disappeared. Everything flowed really nicely.

  34. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  35. Also legal significance. by julian67 · · Score: 1

    This is very good news imo. The BSD patent wranglings were done and dusted quite a few years ago, while allegations and legal actions against the Linux kernel are ongoing. What a great insurance policy it is to have a project working on what may become a drop-in replacement as it matures. The HURD will probably never appear in a stable usable form but in the next 6 months we'll see a familiar looking GNU/BSD that is accompanied by all the usual Debian tools and facilities. The fact that it's Debian doing this is very significant because Debian has the developer resources, the legal and ethical framework, the friendly corporate relationships, and the huge distributed community which can make it work. All they need is some money ha ha ha. And look at the vast number of Debian derived distros and projects. This will make a FreeBSD kernel easily accessible to them all, at very low investment in time and adaptation. It could attract and accelerate development, community and vendor, to both Debian and to FreeBSD. Cross pollination can produce very attractive results. On the other kind it might be a kind of heresy. Best to kill them all and let god sort it out.

    1. Re:Also legal significance. by True+Grit · · Score: 1

      The BSD patent wranglings were done and dusted quite a few years ago

      *boogle*

      That BSD 'wrangling' you refer to, e.g. the USL v. BSDi lawsuit, was about copyright & trademark, not patents.

      If you really think running a *BSD protects you from patent trolls, then you don't understand the problem with software patents. Hint: if *BSD ever becomes as popular as Linux is now, it'll have *exactly* the same problem that Linux has now.

      while allegations

      You mean FUD?

      [Yawwwn...]

      legal actions against the Linux kernel are ongoing.

      Wake me up when just *one* of those legal actions actually *succeeds*.

      [Zzzzzzz...]

    2. Re:Also legal significance. by julian67 · · Score: 1

      You're right, my terminology was incorrect, confused and confusing. I apologise for inducing such sleepiness and appreciate that you overcame your boredom and torpor to the degree that you were able to reply. I'm not sure what *boogle* means though.

      The BSD USL lawsuit was indeed about copyrighted works. So is the SCO action, so while I made a stupid and careless error the broad analogy is still good.

      As for successful legal actions against the Linux kernel, here's one (perhaps it's unique, I don't know).

      Microsoft recently pursued TomTom over a patent related to using both long and short filenames in FAT filesystems, a technique implemented in the Linux kernel (TomTom devices use the Linux kernel). TomTom caved in before it came to judgement. I'd call that a successful legal action by Microsoft (because the defendant conceded) and I also recall that a patch was forthcoming to replace this technique with another that is not covered by any (known) MS patent. See http://lkml.org/lkml/2009/6/26/313

      It does happen. The Linux kernel is no more impervious to legal issues than any other piece of collection of code. The SCO case is ongoing and because I lack prescience I'm reluctant to state that SCO definitely will lose (though I hope they do lose, and lose because their case cannot be substantiated).

      The argument "if *BSD ever becomes as popular as Linux is now, it'll have *exactly* the same problem that Linux has now." sounds oddly familiar. My impression is that outside of the consumer/desktop arena the BSDs are widely used and well known. They compete with the same proprietary UNIX systems as do Linux based systems, so this 'below the radar/parapet/insert_car_analogy_here' contention seems invalid.

    3. Re:Also legal significance. by True+Grit · · Score: 1

      As for successful legal actions against the Linux kernel, here's one (perhaps it's unique, I don't know).

      That wasn't against the use of the Linux kernel though, it was against the TomTom hardware device, and it involved more than just the FAT-LFN patent. The FAT-LFN patent was just one of a half dozen patents that MS sued over in that particular case.

      I'd call that a successful legal action by Microsoft (because the defendant conceded)

      The defendant conceded for the same reason many companies settle out of court, rather than fight. As everyone knows, settling out-of-court does not mean the patent is valid, and in this case, we don't even know whether the FAT-LFN patent had anything to do with TomTom's decision to settle, since it may have been one of the other mobile, GPS-related patents that they were afraid of losing on.

      The patch you linked to doesn't change the fundamental behavior of the kernel's FAT code, it only changed the *default* behavior, the allegedly infringing code and behavior is still there, and can still be used. The patch to change the default behavior is merely to make it easier for integrators who use the kernel in their products, and who don't want to risk an expensive legal fight with MS, to not have to maintain their own custom patch. It still doesn't prove that the patent itself is valid, which points to the problem with the patent system: its almost always cheaper to settle out-of-court than fight, even if you really haven't done anything wrong.

      My impression is that outside of the consumer/desktop arena the BSDs are widely used and well known

      MS's greatest concern, of course, *is* the consumer/desktop arena. They obviously have some reason to fear Linux in that arena, hence their FUD and threats, but not the *BSDs.

  36. Re:We have to put an end to their monopoly of awes by selven · · Score: 1

    No, we don't need more awesome operating systems, we need alternatives to "awesome". Try here

  37. How about the other way around? by rsax · · Score: 1

    I would love to be able to use a small Linux base installation with binary updates - for example Debian with APT. But then I want to use something like FreeBSD ports to maintain all the extra software I install. This is why I enjoy using macports. I upgrade the base OS with Apple's tools but all the juicy bits like nginx, postgresql, etc are managed by macports. If Debian had a way of making FreeBSD ports work like that then one could still get pretty decent hardware support and not have to deal with ancient versions of packages. P.S. I'm aware of NetBSD's pkgsrc and Gentoo. Neither of which feel right to me.

  38. Oh really by TangoMargarine · · Score: 1
    --
    Unity? Screw that: XFCE. Slashdot Beta? Screw that: SoylentNews. Australis? Screw that: Pale Moon. UX developers DIAF
    1. Re:Oh really by Pseudonym · · Score: 1

      Deadpan humour. It is a difficult concept.

      --
      sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
    2. Re:Oh really by TangoMargarine · · Score: 1

      And an easy excuse if nobody laughs at your joke.

      --
      Unity? Screw that: XFCE. Slashdot Beta? Screw that: SoylentNews. Australis? Screw that: Pale Moon. UX developers DIAF
    3. Re:Oh really by True+Grit · · Score: 1

      Deadpan humour. It is a difficult concept.

      Yea, and its a *really* difficult concept when it doesn't even make any sense.

      You clearly do not know the story behind *why* Linux was trademarked (you would have if you'd bothered to read the last para of that section the GP's link points to).

      The trademark was done to *ensure* Linux's continued freedom for the rest of us. Up until the point when that jackass showed up, none of the devs or users cared about the lack of a trademark.

      Alas, there is always at least one greedy bastard willing to game the system for their own profit, with utterly no concern whatsoever for anyone else's freedom.

    4. Re:Oh really by Pseudonym · · Score: 1

      Yea, and its a *really* difficult concept when it doesn't even make any sense.

      If you look back in the thread, you will discover that I was replying to someone who claimed that "software freedom doesn't include restrictions on naming". Once you get that piece of context, it should all make sense.

      I tacitly assumed that it was common knowledge that Linux was trademarked for perfectly good reasons, and that everybody would pick up on this. I apologise for misjudging the company I keep.

      --
      sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
    5. Re:Oh really by True+Grit · · Score: 1

      I tacitly assumed that it was common knowledge that Linux was trademarked for perfectly good reasons, and that everybody would pick up on this.

      And you still don't get why the trademark happened. No one within the FOSS community was interested in the name until an outsider tried to abuse the trademark system and restrict everyone else's freedom to use Linux.

      So the OP's original phrase is still perfectly valid: those truly interested in freedom of software aren't especially concerned with names either.

      I apologise for misjudging the company I keep.

      Yea, some of us not only know what happened, but also *why* it happened. We're not all as ignorant of the history as you are.

  39. Re:That's...nice by Anonymous Coward · · Score: 0

    Compared to your puny distro?

    What, Fedora?

  40. Perfect Timing!! by teraom · · Score: 1

    Call it timing? Snow leopard and Windows 7 are fresh on the shelves!! [ubuntu 9.10 is due too]

  41. Re:I bet the HURD team is turning in their graves. by petermgreen · · Score: 1

    Hurd seems to run though last I tried it (in virtualbox) I got some strange errors (tar repeatedly claiming that files changed while it was reading them when trying to do a package build)

    --
    note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  42. Re:Why not Mach / L4 by Neil · · Score: 1

    You have it backwards, surely? First the "work needs to be done", then if the port becomes consistently release quality then perhaps it will get incorporated into the official release mechanism like other first-class architectures.

  43. Or putting it another way by ClosedSource · · Score: 1

    "GNU appears to be a complete operating system (although not finished)."

    GNU appears to be a finished operating system (although not complete).

  44. It's all about the code, baby by ClosedSource · · Score: 1

    Isn't it great the F/OSS doesn't suffer from all the political BS you find in proprietary software development.

    1. Re:It's all about the code, baby by DragonWriter · · Score: 1

      Isn't it great the F/OSS doesn't suffer from all the political BS you find in proprietary software development.

      Yes, because the completely different political BS around F/OSS doesn't usually stop you from being able to legally use the software.

  45. Re:I bet the HURD team is turning in their graves. by marcosdumay · · Score: 1

    HURD runs over L4, and there are already testing releases of it that run on virtualizaed hardware. It doesn't have a lot of drivers, so it isn't really usefull yet, but it is quite out of the achitect's heads aready.

    Also, it is available on Debian testing, if you want to try the kernel.

  46. Wholesale different bland shoes From(HTTP://www.tn by Anonymous Coward · · Score: 0

    Http://www.tntshoes.com
        The shoes We can supply all kinds of shoes with different styles, You will find what your like.
    Features:
    1) Size for men US8-12 UK7-11(some US13/UK12)
    Size for women US5-8/10
    2) Packing: 1pr/box, 12prs/carton, 12prs/style/color(original box and retro card)
    3) Many designs and colors available
    4) Delivery can be prompt shipping
    5) We accept paypal +++aaa quality .

    OUR WEBSITE:
                                                            YAHOO:shoppertrade@yahoo.com.cn

                                                                    MSN:shoppertrade@hotmail.com

                                                                        Http://www.tntshoes.com
       

  47. Wholesale Jordan Small kid shoes by Anonymous Coward · · Score: 0

    Http://www.tntshoes.com

    Our products are of good quality but reasonable price. And the price depends on the quantity. The more order, the better price.
    AS a professional exporting company dealing with sandals etc. We have wide and good business relations with many factories here. OUR COMPANY accept small trial orders, even drop shipping business is welcome here. We have good business relations with express company here, and we have got the stocks here in our house ware, so we can ship the shoes immediately once we got the confirmation from you. Kindly let us know your interest and we will send you photo samples and quotations for you to study.

    OUR WEBSITE:
                                                                Http://www.tntshoes.com

                                                            YAHOO:shoppertrade@yahoo.com.cn

                                                                    MSN:shoppertrade@hotmail.com