Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
BSD Operating Systems

FreeBSD 4.6 Release Delayed 223

Posted by michael from the when-it's-ready dept.
Dan writes "Bruce A. Mah from the FreeBSD Release Engineering team announced that due to some late-breaking issues, 4.6 will be released about a week later than originally planned."
This discussion has been archived. No new comments can be posted.

FreeBSD 4.6 Release Delayed More

FreeBSD 4.6 Release Delayed

Comments Filter:

  • Sheesh! (Score:4, Funny)

    by Lethyos ( 408045 ) on Saturday June 01, 2002 @08:13AM (#3622181) Journal
    Haven't you editors heard yet? BSD is dying! Get with the program and post another Linux 2.5 patch level increment announcement. Thank you!

  • Only a week (Score:3, Insightful)

    by saphena ( 322272 ) on Saturday June 01, 2002 @08:15AM (#3622184) Homepage
    Wouldn't it be nice if all software releases were *only* a week late?
  • Just wondering -- with so many companies now getting behind Linux (maybe just because of the cute logo), what companies are building solutions with *BSD code? This is not a slam, as I use OpenBSD on my Sun boxes, but seriously, I would like to know if IBM, Compaq, Sun, etc., have any BSD solutions.

    • A lot of companies prefer *BSD to avoid a lot of complications of the GPL.
      • A lot of companies prefer *BSD to avoid a lot of complications of the GPL.


        My question, again, is who?

    • Isn't OS X based on *BSD? (FreeBSD, I think.)

      Granted, not all OS X users are hardcore UNIX hackers, but I think having Apple pushing a *BSD derivative is a great way to stir up some more interest in it.

      • Yes, MacOSX is based in part on BSDLite 4.4. Some libraries come from NetBSD, while most of he utilities stem from FreeBSD. BSD's contribution to MacOSX [apple.com] But MacOSX/Darwin is also based on Mach. Cheerfully adding to the confusion are various projects such as Fink [sf.net], which aims to port a good deal of linux software to MacOSX. Occasionally, dumb flamewars will sprout up, with one side advocating FreeBSD style ports, etc., and the other advocating a more linux-like style. I guess it depends on what systems you've used before.
    • Uhm, there's some little company, I think its called... Apple? :)
      • Apple?


        Not exactly fair to claim this as embracing a free BSD base, as OSX is not free, portable, and open-source. This is like claiming MS-DOS is based on Unix because it has files and directories.

        • This is like claiming MS-DOS is based on Unix because it has files and directories.

          Yeah, it took a couple of versions before MS-DOS got directories. :)

        • Re:Apple (Score:4, Insightful)

          by JordanH ( 75307 ) on Saturday June 01, 2002 @08:35AM (#3622234) Homepage Journal
            • Apple?

            Not exactly fair to claim this as embracing a free BSD base, as OSX is not free, portable, and open-source.

          No, but Darwin [apple.com], on which OSX is based, is free, portable and open-source. Oh, and it's based on a free BSD base (with a Mach microkernel).

          • This is like claiming MS-DOS is based on Unix because it has files and directories.

          More like claiming that Solaris is based on AT&T Unix, which it is.

          • How is Darwin 'based on BSD'? It is a Mach kernel with a BSD personality layered on top. By that argument Linux has a better claim to be 'BSD' than Mac OS X does.

            If the Darwin kernel is actually based off the 4.4BSD code then fair enough. But I haven't seen that it is. As far as I can tell, Apple took a microkernel, put a Unix compatibility layer on it and called it 'BSD' for marketing reasons.

            • Re:Apple (Score:4, Insightful)

              by JordanH ( 75307 ) on Saturday June 01, 2002 @09:06AM (#3622288) Homepage Journal
              • If the Darwin kernel is actually based off the 4.4BSD code then fair enough. But I haven't seen that it is.

              You might not see this if you don't actually look into it. Like, maybe start at that link I provided?

              From this discussion of the history of Darwin [apple.com] we read:

              Darwin also incorporates a full implementation of BSD (Berkeley Software Distribution) UNIX, welded on top of the Mach kernel.
              and
              Darwin wraps a customized version of 4.4 BSD-Lite2 kernel and userspace around Mach. It includes many of the POSIX APIs, exporting them to user-space, and abstracts Darwin's file system and networking. Darwin's BSD also provides the process model, basic security policies, and threading support for Mac OS X.

              I guess seeing that much of Darwin is based on the 4.4BSD(-lite2) code, then this is "fair enough" for you.

              From what I can tell, Mach is a very bare bones kernel here, not providing a process model or networking, etc.

              • Re:Apple (Score:3, Troll)

                by Ed Avis ( 5917 )
                The web page you refer to doesn't make it clear whether Darwin is actually based on BSD, or just an implementation of the BSD process model, filesystem, and other APIs. The GNU system is designed to follow BSD Unix - does that mean that Debian should start appearing in the BSD section of Slashdot?

                Well there is the mention of 'a customized version of 4.4 BSD-Lite2 kernel'. It's not immediately obvious how to transplant a monolithic kernel to run on top of Mach, but I guess we should take Apple's word for it that you really are running a BSD system: just one that happens to be hosted on Mach in some way. Maybe the objective test is: would a developer who is familiar with 4.4BSD, or FreeBSD or NetBSD, feel at home hacking the Darwin kernel?

                Could anyone who knows more about this stuff clarify what is happening? I am assuming that JordanH is not _the_ Jordan H. :-P.
                • Well there is the mention of 'a customized version of 4.4 BSD-Lite2 kernel'. It's not immediately obvious how to transplant a monolithic kernel to run on top of Mach, but I guess we should take Apple's word for it that you really are running a BSD system

                  MACH does some VM stuff, and a little IPC, but the BSD is pretty much a full one. Don't take my word for it though, or Apple's. Go bloody download Darwin and look!

                  • Could anyone who knows more about this stuff clarify what is happening? I am assuming that JordanH is not _the_ Jordan H. :-P.

                  No, I'm not _the_ Jordan H, but I am _a_ Jordan H. See my brief bio for details [slashdot.org]. Also, see this journal entry [slashdot.org] describing a time when I tried to play off my name for a joke.

                • The web page you refer to doesn't make it clear whether Darwin is actually based on BSD, or just an implementation of the BSD process model, filesystem, and other APIs.

                  Then you may be confused by what makes an OS an OS. The original ports of linux to PPC were based on MkLinux which was a microkernel as well. Are you saying that those weren't linux? Also if you are worried that not all of the code is exactly the same as another BSD out there then every platform NetBSD runs on wouldn't be able to be called "NetBSD" in respect to one another because of the extra code it takes to port from one architecture to another.

                  If you think of Mach as your "hardware layer" then Mach becomes the platform you implement BSD on [like Lites].

                  That may not be a perfect analogy.
                  • Well SCO UnixWare (aka Caldera OpenUnix), for example, has a Linux personality to let it run many common applications. That does not qualify it as Linux. So if Darwin were simply a collection of APIs layered onto Mach, it would not count as BSD. BSD-compatible, yes, but that same claim can be made by many Unixes.

                    OTOH, if Darwin actually uses BSD code (in the same way as MkLinux uses Linux code), then it's reasonable to count it as 'BSD' while not giving that title to Linux or Solaris or Cygwin.

            • Re:Apple (Score:3, Informative)

              by be-fan ( 61476 )
              Don't be daft. Actually read about the OS-X design before talking about it! The Mach microkernel is not a full kernel in the sense that Linux is. It provides a minimum of functionality. It requires external servers to provide much of the traditional functionality of a kernel. In OS-X (as in NeXT) the external server takes the form of a monolithic BSD system server. In other words, a standard BSD monolithic kernel is altered to run as a server on top of Mach. Unlike most microkernels, OS-X puts its system server in kernel space, which eliminates many benifets of the microkernel design (ie. better protection between kernel components). In OS X, the BSD system server is based on a customized 4.4BSD-lite2 kernel with many parts (like the whole networking infrastructure) thrown in from various BSD OSs, primarly FreeBSD 3.2. In OS-X 10.2, this kernel adopts a bunch of code from FreeBSD 4.4. To satisfy your curiousity, here's the dirt from Apple itself: http://developer.apple.com/darwin/history.html. Read the WHOLE thing. I'm not going to tell your where in the page it is, because reading is good for you.
            • I think you overestimate the amount of work that's
              done in a microkernel. The BSD part of the kernel
              does much more than the Mach part. It's not a set
              of stubs on top of Mach functionality, rather it
              provides filesystems, networking, security ... aside
              from interfacing to the hardware, I don't think Mach
              does anything besides memory management and Mach IPC.
          • From what I remember, the license Apple used for Darwin only permits things to be contributed.. (eg. you can't fork it like you could Linux or *BSD, but you can give them code). If this is right, that is not open source/free software. You could look at it as Apple leeching free development without giving anything back.

            Hopefully I'm wrong on this....
              • From what I remember, the license Apple used for Darwin only permits things to be contributed.. (eg. you can't fork it like you could Linux or *BSD, but you can give them code). If this is right, that is not open source/free software. You could look at it as Apple leeching free development without giving anything back.

                Hopefully I'm wrong on this....

              I think you're wrong, as I don't see the restrictions you are referring to in The Apple Public Source License [apple.com], which covers Darwin. But, the language is legalese so maybe I'm missing it.

              I note that the Open Source Initiative [opensource.org] includes The Apple Public Source License [opensource.org] on their list of approved licenses [opensource.org].

              • Well, I just read the entire thing, and you're right :-) Maybe something good will come of darwin yet..
          • Actually the latest in the Darwin CVS is not so much a microkernel anymore. According to the latest reports, the kernel is now a more FreeBSD 4.4 smooshed together with Mach for a more monolithic architecture. This was to overcome the inherent performance issues with micro-kernel design.
              • Actually the latest in the Darwin CVS is not so much a microkernel anymore. According to the latest reports, the kernel is now a more FreeBSD 4.4 smooshed together with Mach for a more monolithic architecture. This was to overcome the inherent performance issues with micro-kernel design.

              That's a familiar story. I recall hearing that the OSF/1 developers, which was originally based on Mach, did exactly the same kinds of things to get acceptable performance.

              The only OSF/1 system to see much use was from Digital (now Compaq, I mean HP), with their Unix for the Alpha, originally called DEC OSF/1, then Digital UNIX and then Digital Tru64 UNIX, then Compaq Tru64 UNIX and now, pant, pant, I hear that another name change is in the works under HP.

              • The only OSF/1 system to see much use was [...] DEC OSF/1, then Digital UNIX and then Digital Tru64 UNIX, then Compaq Tru64 UNIX and now, pant, pant, I hear that another name change is in the works under HP.


                hmmm ... like ... um ... Garbage Fill ?

                IIRC, it's supposed to be scrapped in favour of world's ugliest surviving Unix, aka HPUX.

        • Darwin has lots of BSD code in it, and it's open source and portable (will work on X86). The only part that's not is Aqua.
    • Well, right off the top of my head I can think of three that havn't been mentioned: Niksun NetVCR (a really sweet piece of kit), Juniper routers, and the Cybernet NetMAX. LOTS of people with embedded solutions that require a bit more oomph than your normal embedded OS can provide use FreeBSD. From a corporate point of view, the FreeBSD has a very favorable license and a conservative release schedule that helps insure a stable OS for your embedded project. Also, it doesn't hurt that the FreeBSD source tree is in CVS, and you can maintain a branch relativly painlessly without having your proprietary changes merged back in the main branch (although some companies merge their changes anyway, look at vinum).
    • OpenBSD has good commercial support, due to the support of the hardware crypto folks.

      Of course, being security companies, they don't talk much about it.
    • I believe the Nokia/Checkpoint platform is BSD based...
    • Compaq sells hardware to Yahoo, which is a
      FreeBSD shop.

      The Nokia Firewall-1 implementation is based on
      a modified FreeBSD.

      IBM's InterJet router-toaster is based on FreeBSD.

    • Apple has a BSD solution. It's called OS X [apple.com].

  • Not too bad (Score:2, Insightful)

    by jquirke ( 473496 )
    He summed it up well.

    Good to see how the quality of the release takes precedence over any deadlines. That's the way it should be. I'd rather have FreeBSD 4.6 a month late than have a buggy one now.

    • Re:Not too bad (Score:2, Funny)

      by vchoy ( 134429 )
      He summed it up well.
      Good to see how the quality of the release takes precedence over any deadlines. That's the way it should be. I'd rather have FreeBSD 4.6 a month late than have a buggy one now.

      Code release: Free BSD: 7 days delay - Secure + very rare security patches
      MS: 7 months late (only binaries) and then... urmm...SP1 +reboot+SP2+reboot +SP2a...oh stuff it...http://www.windowsupdate.com
      BSD wins

      Marketing: Free BSD: 4.6 + /. MS: 2000/XP + $Marketing$
      Afterall 2000 is 434.78260869565217391304347826087 times better than 4.6
      My calculator refuses to give me a comparision for XP/4.6 ....always displays '0'.
      You be the Judge of this one...

    • I'm not sure if we can all say the same thing about Mozilla. :-)

  • A Good Thing (Score:3, Insightful)

    by grokBoy ( 582119 ) on Saturday June 01, 2002 @08:23AM (#3622205)
    If only major software vendors delayed code to iron out the bugs, rather than shipping it on a date set twelve months ago regardless of the bug count.

    I'm sure that having a stable DHCP installation is going to be important to all the cable modem users out there running FreeBSD, so this is clearly A Good Thing.

    • "If only major software vendors delayed code to iron out the bugs, "

      And if only major hardware vendors delayed iron to code out the bugs....

      graspee

    • I used to work at VERITAS Software [veritas.com]. They routinely delayed release dates on products if quality wasn't good enough.

      There is a problem with this, though. Customers made business plans based upon planned release dates. When those release dates slipped, the customer's plans were upset. This could leave them in awkward situations because they couldn't do certain things without features present in newer versions of the software. As a consequence, we always dreaded a slipping of the release date because it would provoke great annoyance from the customers.

      Customers want it right and on-time. If they have to pick between those two, they will select to have it right, but will not be happy about it!
      • This is the difference between 'consumer' software and 'professional' software - we've come to expect x new features and x new bugs with every release of say, MS Word, and they are probably not showstoppers. Plus the vendor is probably planning a big advertising campaign so the software has to be ready.

        But on the flipside, people who need software like Veritas are using it predominately in mission-critical environments, and would probably agree that quality outweighs release timeliness. Its all very well getting it right on schedule, but if you then lose your entire storage network due to a few bugs, you'll wish it had been delayed in the first place.

        In fact, I'd hypothesize that most non-free software that ships on time is driven purely by commercialism and a desire to get to market before the competition, rather than a real desire to keep the user base happy.

      • In large businesses the situation is different. Veritas will release Netbackup 4.5 but that doesn't mean everyone running a backup envornment will upgrade immediately. With a couple million dollars worth of equipment and million dollar contracts you don't try something as soon as it comes out. You put it in your lab for several months and hammer it. You don't depend on another company telling you that the new version is much better even if they have a good reputation like Veritas.

        I believe that this is one of the main reasons for the easy adaptation to a Windows enviroment in large businesses. If everyone uses Office 2000 and XP comes out they test the new product for months to ensure that no difficulties are encountered. When upgrading an OS (we all use 2000, not XP) - every single application the company uses must be tested in the new environment.

        Gettign to the point now. Stability and compatibility are the two most important things in the enterprise enviroment. Release dates that vary by a couple weeks for even a couple months won't really affect large scale customers since they will have it in testing for months.

        The Canadian military is currently using Windows 95 and Office97. For the past couple years they have had Windows 2000 and Office 2000 under testing. They will eventually transition to the new system. They wouldn't however be upset it the release of Windows 2000 or Office 2000 wasa delayed by a couple months since they spend years testing software and hardware before being used.

        Release dates affect mostly home cnmsumers. The people upset that Office XP isn't out right now. The people that will buy it and install it on their main machine - if it break they rebuild from scratch.

        On another note, I think this may be a reason that a Microsoft solution is used more often than a linux solution. Microsoft has much less updates. This is bad in the respect of needing security updates, patch fixes and other such things but it is good in the fact that not much changes over time. Does anyone here have a linux system from 1995 that they use on a daily basis and haven't updated since?

        That's all for now...
    • DHCP problems? Never had them. Maybe i'm just lucky?
    • Thats great and all but will someone wake me up when I can use cardbus network cards on my laptop with ANY of the bsds, straight out of the box (meaning not having to build my own kernel to support it)

      I could not do this as of a month or two ago.

      siri
      • Wow, your life must be a living hell... having to type in the 2 or three commands required to build a new kernel. Or is it that you are having trouble with vi?

  • Shoot (Score:3, Funny)

    by Greenrider ( 451799 ) on Saturday June 01, 2002 @08:28AM (#3622220)
    Well, there goes my weekend. Leave it to Slashdot to be the bearer of bad news.

    ...oh wait...no FreeBSD? I thought they said no free LSD.
    • ...oh wait...no FreeBSD? I thought they said no free LSD.

      You know how the saying goes... there were two good things that came out of Berkeley. =)

  • Lack of nerds? (Score:1, Interesting)

    by Anonymous Coward
    It seems to me that what we have is a lack of nerds here. Everytime we see a FreeBSD topic, 50% of all posts are pure junk.
    Now it seems to me that we mostly have wannabe nerds here. I'd say that if you truly are a geek, you would not be afraid to install another OS. In fact it would be something you would thing was fun. "Hey another OS/version, lets check it out just for the fun of it".
    Personally I like to install all kind of OS, yes I even have tried to install M$ servers just to check them out, not that I would like to run one. Playing with all kinds of operating systems and making it run of strange hardware is fun to me.

    But maybe the people truly into *BSD might not post here at all because all the responses to their posts most likely will be met with a useless reply.
    It's not that I don't like Slashdot, but why are the view and posts so narrowminded? It seems to me that the topics that gets the most posts(besides from M$ bashing), are the ones where you can apply some kind of generic opinion and without any deep insight into the subject. But maybe this is how it always has been and it's just me who are wrong.

    • The thing is that all REALLY HARDCORE computer nerds are inside their boxes, putting together new hardware components, programming drivers for them, doing hardcore kernel hacking etc. They don't have time to waste on slashdot!


      Yes, it is really tiring that EVERY BSD article is flamed by the exact same stupid responses. Yes Linux does have a larger support group, and it was more user friendly at an earlier stage, but that doesn't mean it is worth less as an OS. No matter what anyone thinks of any operating system, I am going to use whatever I think works best and most efficiently for ME on my machine. Currently that is FreeBSD. If a major OS release comes out on the x86 platform, I usually give it a go. Ignorance will never help, it might only stop me from finding something I find valuable.

      • I am going to use whatever I think works best and most efficiently for ME on my machine ME is anything but best and efficient. ;)
  • Suppose an operating system that isn't geek-chic announced that their OS was going to be a tiny bit late. Slash-dotters would be standing on their heads to be the first person that posts a "serves 'em right" message.
    As much as I don't like M$ do you think that they delayed their OS's because they wanted to or that they didn't think they were stabel enough?

  • OpenBSD (Score:2, Insightful)

    by Anonymous Coward
    This is one of the many reasons why I prefer OpenBSD to FreeBSD. OpenBSD is ALWAYS due either December 1st or June 1st. Now, today would've been the official release date of OpenBSD 3.1, but it was officially released 2 weeks ago!! This is the only big project I can think of that does not delay its release. Linux 2.4 was late by a year, FreeBSD 4.5 was late by a couple of weeks, 4.6 will be late by at least one week, FreeBSD 5.0 was delayed by 14 months, etc. The thing with OpenBSD: they don't do revolutionnary released like Microsoft. Each new version contains many security and bug fuxes, new application, new hardwares code and a couple of new features (e.g: openssh, pf, pfauth). Maybe some other projects should take example, not only on OpenBSD's commitement for security, but also its commitement to respect release schedule.
    • If I wanted OpenBSD, I would install OpenBSD. I Don't, I want FreeBSD. I want it NOW - so I have installed 4.5. I still runs fine. No problems there. Maybe, if I get round to it, I'll install 5.6 another day. What's the problem.

      Then again, Windows 95 still runs as good as it ever did (I take that back - windows95 DOES NOT run as well as it used to, it crashes even more often than it used to on my dual boot machine that runs FreeBSD without crashing at all). And Upgrading Win95 requires spending money. With FreeBSD, you get to upgrade for nothing.

  • both userland and kernel PPP caused kernel panics for me as soon as i tried to pass data. I don't know why, but I had to rebuild 4.5-RELEASE. I'm glad they're fixing some issues. I'm sure that's one of them. Everything else was working great though...atleast that i noticed.
  • I saw this the other day - and was about to submit it. But then I figured - who cares? It's just one week. This is a BSD section thing at best - certainly not front page.
  • What is this "release" you refer to? I just make buildworld whenever I feel like it.

Slashdot Top Deals

I've noticed several design suggestions in your code.

Close