FreeBSD 5.4 Released

FreeBSD 5.4 is out. Reader KFW excerpts from the announcement: "The Release Engineering Team is happy to announce the availability of FreeBSD 5.4-RELEASE, the latest release of the FreeBSD Stable development branch. Since FreeBSD 5.3-RELEASE in November 2004 we have made many improvements in functionality, stability, performance, and device driver support for some hardware, as well as dealt with known security issues and made many bugfixes." Here are the release notes.

Comments

Sorry guys

[DarkHelmet]

I'm not dead.

I'm really sorry everyone, but a story like this is just begging for it.

http://www.mwscomp.com/movies/grail/grail-02.htm

FreeBSD:
I'm not dead!
CART MASTER:
What?
CUSTOMER:
Nothing. Here's your ninepence.
FreeBSD:
I'm not dead!
CART MASTER:
'Ere. He says he's not dead!
CUSTOMER:
Yes, he is.
FreeBSD:
I'm not!
CART MASTER:
He isn't?
CUSTOMER:
Well, he will be soon. Netcraft confirms it.
FreeBSD:
I'm getting better!

Re:Sorry guys

[Anonymous Coward]

*sarcasm detector explodes*

how...

[Anonymous Coward]

Ok...So how much is FreeBSD 5.4 going to cost me?

Re:how...

[rubycodez]

Outpacing the even the steep markups on Microsoft office products in the last 3 years, 5.4 will cost you 150% as much as 5.3 did ! ! ! !

Re:how...

[td]

5.4 times as much as 1.0.

It's not dead...

[Anonymous Coward]

It's just pining for the fjords.

I mean, why do you think Linux was started in Finland?

Re:It's not dead...

[BSDFreak]

It's just pining for the fjords.

Shouldn't it be pinging for the fjords?

Better SMP support? Better MySQL performance?

[Anonymous Coward]

I hope so. SMP + MYSQL performance is horrible with the *BSD's across the board. :(

congrats

[moz25]

Congrats to the freebsd team.

I have one (uneducated) question though: they mention a number of security fixes. How long does it generally take for a fix/patch to come out on freebsd compared to linux (or the other bsd variants)? I'm considering experimenting with it, but the relative comfort of packaging systems I'm familiar with makes it sort of hard.

Re:congrats

[Anonymous Coward]

Generally the mailing list comes out with patches much quicker than other flavors of *nix. 24 hour turn around times with patches is not uncommon for FreeBSD (They pride themselves with security)

Re:congrats

[Triumph The Insult C]

you can usually measure it in hours

openbsd ... it's probably already been fixed for a few months

Re:congrats

[jwthompson2]

As fast as they are fixed, which in reality ends up being comparable to Linux, just listen on the appropriate mailing lists and follow the step-by step instructions. There are also some automated utilities in the ports collection that ease security updates. The BSD ports system will take care of most of your packaging concerns as well since it is an actively updated collection, although most require compilation from source there is the binary alternative, package, which should be easy enough for most RPM folk I would imagine.

Check out this link regarding packages and ports.

Re:congrats

[Triumph The Insult C]

"a matter of a few hours" is what i meant, but didn't type. sorry

Re:congrats

[saleenS281]

As fast if not faster than linux. Also, IIRC (don't flame me, correct me if I'm wrong) most linux variants of packaging systems were derived from BSD. As for worrying about about the packaging system... you only need two utilities, and two config files. cvsup-without-gui and portupgrade. It's literally as simple as portinstall *package you want*. And if you want it updated portupgrade *package to be updated*. You just have to keep your ports tree up to date and you'll have the most up to date versions available. It's probably one of the simplest systems I've ever seen. Also, there's portaudit which will automatically scan the security lists to see if any of your installed packages are vulnerable. And finally the "portupgrade -a" command which will update everything (although I can't say I recommend this one) but if that's your cup of tea, it makes it real simple.

Re:congrats

[drmerope]

Let me fill in some holes others left out. New releases are announced as having fixed security problems, but that is a comparison against the previous version's original ISO image only.

Security fixes are backported to earlier versions. Those versions still officially maintained have fixes backported by the security officers. Older versions tend to also get fixes but merely by the work of interested committers. Thus it isn't usual to see fixes being backported to releases as far back as 4.3.

What do I mean by backported? Users can update their /usr/src directory and rebuild. More recently a binary update service has been available.

Thus there is for example 5.3-RELEASE, and 5.3-p5.

Generally speaking, there is no need to wait for new releases to get fixes. Fixes are painlessly and automatically available almost overnight.

All of this applies to the software officially maintained by the FreeBSD system--i.e., anything in the "base system" Other software generally gets fixes in ports soon after the upstream version has a fix... but backing this is the port-audit database. port-audit is maintained by the security team and lists all the known vulnerabilities against third-party software. A cron job mails you warnings about vulnerable third-party software. The ports system warns you about vulnerable software and libraries when you attempt to install (even when a new install depends on an already installed but vulernable library.

Re:congrats

[cperciva]

24 hour turn around times with patches is not uncommon for FreeBSD

In all honesty... 24 hours is very unusual for us. I can think of one case where it happened recently, but that was when we rushed an advisory out in order to fit into the 5.4 release schedule.

A more typical time is 3 days, since we want to test carefully to make certain that a "security fix" never ends up breaking something else.

Re:congrats

[archen]

How "fast" a security fix gets out is a bit more complicated then "here is the fix". With FreeBSD the base system gets fixes out very fast, within a day or two. With Linux the fix comes out in just as fast, but how fast YOU get the fix varies a lot depending upon your distro. With FreeBSD ports (aka the third party software collection on Linux) it varies depending upon who is the port maintainer and how much testing they feel should be done - that goes for Linux too however.

Re:congrats

[jusdisgi]

A more typical time is 3 days, since we want to test carefully to make certain that a "security fix" never ends up breaking something else.

Am I the only guy that finds that really scary? I mean, I agree, we need to be pretty sure that a security fix isn't going to break other things....but my stress level on day 2 of knowing that my sshd has a remote-root-exploit would be pretty damned high. I would hope that my distro could check for breakage during the first day, before I start sucking down antacids to deal with my ulcer.

Re:congrats

[cperciva]

my stress level on day 2 of knowing that my sshd has a remote-root-exploit would be pretty damned high

This is why security issues are usually not publicly disclosed immediately. A window of a few days between informing vendors and public disclosure allows vendors to prepare and test their patches.

Obviously, if there is a publicly disclosed remote root vulnerability in sshd, FreeBSD would fix it as soon as possible.

Remember, cvsup is your friend!

[nubbie]

Using CVSup and then Rebuilding "world"

The end is near.

[Eric(b0mb)Dennis]

Sarge was frozen.

FreeBSD has risen from the grave.

It's hailing here in northern California in may.

The end is near, put on your glasses and anti-radiation suits boys, we're in for a ride.

Re:The end is near.

[grolschie]

Does this mean Duke Nukem Forever is about to be released too?

Re:The end is near.

[i.r.id10t]

That's OK. It is somebody else's problem.

Re:The end is near.

[bubkus_jones]

The glasses, they do nothing!

Re:The end is near.

[Zan Lynx]

According to this month's PC Gamer, yes.

Re:FreeBSD

[whiteranger99x]

actually, that's a Daemon :P

Re:Better SMP support? Better MySQL performance?

[Anonymous Coward]

Here's a few things from the release notes that might help with MySQL and/or SMP: A number of bugs have been fixed in the ULE scheduler. A bug in Inter-Processor Interrupt (IPI) handling, which could cause SMP systems to crash under heavy load, has been fixed. More details are contained in errata note A number of bugfixes for libpthread have been merged from HEAD. Anyone from FreeBSD know for sure if the fixes above will help bring FreeBSD up to par with Linux as far as MySQL performance on SMP machines go?

good stuff

[Herr_Nightingale]

I just wish Microsoft would pull an Apple and stick a GUI on top of it. Sigh. Longhorn would come a lot sooner (mid-2010?) if they took this route. Plus it might not suck hairy donkey balls then.

Re:good stuff

[dabigpaybackski]

I've thought the same thing on more than one occasion, but the time for MS to have made that decision was years ago, before they spent all that money on anti-Unix advertising.

It would be hard for them to talk their way out of the rhetorical position they're in, where (it is claimed) Unix is inferior/dead/too expensive.

It's too bad, because I think they would be in a stronger position had they gone the Apple route. Can you imagine how different things would be if they had released a Unix-based OS a couple of years ago? Unthinkable.

Re:good stuff

[menkhaura]

Done and done.

Re:good stuff

[pr0c]

Herr_Nightingale (556106): I just wish Microsoft would pull an Apple and stick a GUI on top of it. Sigh. Longhorn would come a lot sooner (mid-2010?) if they took this route. Plus it might not suck hairy donkey balls then.

I've always wanted the same thing, sadly however the majority of ./ would just call Microsoft theives, etc... but its okay for Apple to do it...

I'm also waiting to see how receptive ./ is of the [potentially] new Apache project.. an implimentation of Java. Lets see if they treat it the same as mono or again hipocrits.

Re:good stuff

[jbplou]

You think they could be in a stonger position then they are now. They own the desktop market, have a good position in the server market, own the office suite market, own a decent development and business integration and back end software market(think SQL Server, .NET...). How else could they improve in position in your eyes, no wait lets not worry about your eyes lets worry about reality. Microsoft is doing fantanstic business wise, sure IBM and HP rank higher on the fortune 500 but microsof doesn't do harware or consulting. For a software only company they have dominated there chosen markets for the most part. Going to an OS based on Free-BSD would make it much easier for their competitors to run software on different platforms, it would be stupid, they need product lock-in to maintain their advantage.

Re:good stuff

[ValiantSoul]

"but its okay for Apple to do it..."

No, NeXT did it, and Apple was looking to buy either NeXT or BeOS to base OS X on. They bought NeXT which was based on BSD.

Also, anyone notice how Apple's two choices had their first third and fourth letters capital but not the second?

Re:good stuff

[IntlHarvester]

When Microsoft wrote their own kernel, *BSD was essentially obsolete, and System V UNIX was ridiclously expensive. FreeBSD is is just now adding features that were in the NT Kernel 10 years ago.

Now they've got decades worth of software designed for the very non-Unix NT kernel, so switching isn't much of an option, nor would it necessarily produce better peformance or stability.

had they gone the Apple route. Can you imagine how different things would be if they had released a Unix-based OS

Apple sells a fairly medicore Unix where all the value is proprietary API layers. I can imagine that if Microsoft had done this, they would have been tarred-and-feathered by the Unix world for Embrace-and-Extending Unix -- rather than accepted like Apple is. Lose-Lose situation.

And it wouldn't that much different either -- Windows users would still be running proprietary Win32 in orange windows and would be just as ignorant of any kernel-level features.

Re:good stuff

[netdur]

microsoft did release UNIX based OS years before they did release ms DOS, and they called it xenix, anyway... to have UNIX as OS doesn't help anything, I have IE6 running over wine on linux, lately when I start it (mainly I use IE to test my web site) IE display pop-ups with different ads every time! my homepage set to "about:blank" so the only thing I can think, I have malware running over wine on my linux box, installed trough some hole in IE, in other hand, surfing net via firefox is secure, whatever your OS is, firefox makes it secure, microsoft is in big trouble they have crapy softwares... "it just work" & "fix it later" wont work, end of microsoft has been start...
--
- ah! forgive my stupidness

Re:good stuff

[mrselfdestrukt]

User starts typing:
l
*pop*
Clippy on Unix:
It looks like you're trying to write a cross-platform, intelligent forking, self-sustained, multi-thread application. Do you need help with that?
User: Fuck off clippy. I just wanted ls!

Re:good stuff

[larry bagina]

Can you imagine how different things would be if they had released a Unix-based OS a couple of years ago? Unthinkable.

Xenix. SFU. Look into it.

Re:good stuff

[Herr_Nightingale]

Product lock-in can be done the same way Apple's managed it: proprietary GUI layer. Want to run UNIX stuff? It'll look like crap, and you have to jump through three flaming hoops to do it, install X.org, etc.
Of course Apple's got the hardware lock-in too..

Re:good stuff

[Mancat]

The NT kernel is pretty solid, and well-engineered. The problem is the crap that Microsoft likes to pile around it.

Though NT isn't very Unix in design, it's still possible for Microsoft to add a Unix sublayer, a-la Services for Unix, but more completely integrated into the OS. POSIX compliance is there, and you can run most well-written Unix software in SFU already.

If Windows has to go the Unix route, Microsoft is still capable of doing it while maintaining compatibility.

Re:good stuff

[DrSkwid]

You should go find out the path from ATT - MS Xenix - SCO

Re:good stuff

[setagllib]

It would break the backwards compatibility users expect when upgrading to 'the next Windows', which Microsoft have managed to maintain since DOS days with surprising success (and resulting bloat). The NT kernelification made DOS direct hardware access fail and some other performance hacks, but what can you do?

While I'd love to see a highly usable layer on top of a highly usable base system, I'd rather see more sensible software come out for Windows. I have begun to admit that Windows is still ahead in a lot of usability aspects, at least for desktops, and as long as you're a creative administrator you can work around its security and flexibility shortcomings. On servers, run BSD or Linux. No problem.

If Longhorn was FreeBSD + GUI it would be pretty awesome but, unless they pull of something really spectacular, would no longer be Windows-like and hence drive away a lot of customers. This is a bad thing for them. How MacOSX manages to be so backward compatible boggles the mind. Microsoft can't match that.

Re:good stuff

[mattyrobinson69]

MacOSX is backwards compatible through a subsystem (just how XP is "compatible" with some dos programs).

Windows already has a POSIX layer, and the common unix utils (google for "microsoft services for unix"). its about 250mb but at first glance looks ok.

They could do the same for w32 apps.

Re:good stuff

[jbplou]

Of course Apple's got the hardware lock-in too

More like product lock out, nobody uses Apples. Let me rephrase, very few people and almost no businesses use Mac.

Re:good stuff

[lewiz]

Says on Wikipedia that Microsoft exchanged Xenix for a 25% share in SCO. Is this still the case?

Re:good stuff

[limon.verde]

The SCO Microsoft got shares in is now known as Tarantella, Inc. From the Wikipedia article on The Santa Cruz Operation:

SCO announced on August 2, 2000 that it would sell its Server Software and Services Divisions, as well as UnixWare and OpenServer technologies, to Caldera Systems, Inc. The purchase was completed in May 2001. At that time Caldera changed its name to "Caldera International", and the remaining part of SCO, the Tarantella Division, changed its name to "Tarantella, Inc."

In August 2002 Caldera International renamed itself "The SCO Group" since the SCO UNIX products were still a strong source of revenue mainly due to the huge installation base dating back to the 1990s. It is this SCO Group, formerly known as Caldera, and not the former Santa Cruz Operation now known as Tarantella, that sued IBM in 2003 for $1 billion for allegedly "devaluing" Unix by contributing to the Unix-like Linux operating system.

Re:good stuff

[setagllib]

But it'd have to be the other way around. FreeBSD-like kernel with PECOFF and some brutal surgery to support Windows-like paths (conceivably like how WINE does it, only better, because Wine is a nightmare) and process management concepts. NTFS and FAT* would no longer be useful at all, so you'd have to either use UFS2 or develop another FS. UFS2 has ACLs which I imagine could be used to replace NTFS' admittedly useful ACL system. Sub-FS encryption and compression are ideas with little use but are still possible to replace.

But it would certainly be interesting to see anyway. A real 'BSD for WIMPs' would really be a sight. But if it doesn't match the corporate hardware support of Windows (including the native ability to use Windows drivers - but NDISulator/ProjectEvil is Good Enough For Now) it won't be useful to anyone.

Re:good stuff

[mattyrobinson69]

that was my point - they do it now with the POSIX subsystem, so they could (and bloody should) just use a freebsd kernel with a *high level* win32 subsystem, that way win32 could be phased out like msdos support was.

They dont need to port the 'drive' naming system (c:, d:, etc) as its retarded (maybe provide a *high level* emulation for it).

device drivers in win2k didn't work in xp, xp's wont work in longhorn so if they scapped blackcomb or whatever its called and replaced it with freebsd+somewin32stuff the drivers wouldn't have to be rewritten any more than they would anyway.

i believe reiser4 would be a good filesystem for microsoft to use (although i dont know anything about ufs). if they paid namesys to port reiser4 to freebsd (if its not there already)and stablise it. they would have a solid filesystem with plugin support (plus performance and reliability[atomic writes]). The plugins would be useful in reiser4 because it could then do the encryption, compression etc that NTFS supports. (reiser does support ACL's doesn't it?)

Re:good stuff

[setagllib]

Reiser4 goes against the grain of nix-centric file systems by fragmenting like NTFS and FAT* do. I don't know what solution has been developed for this but it definitely means there's a shortcoming in the design.

I respect UFS for being very well balanced: it scales okay, it is reliable even after crashes (with SoftUpdates anyway), it performs well on most tasks even without clever journalling, and the lack of journal means it is suitable for even very small volumes (whereas ReiserFS seems to carry a universal 50 meg journal, making it useless for smaller media). It also happens to be supported very well in every BSD and to a lesser extent in Linux, with a limited Windows driver available. Although I'm all for journalling file systems, I don't recommend relying on them like Linux often seems to do (this is a problem on lower-end systems which need reliable storage but can't afford the overhead of journalling).

Anyway, you present the points of high level emulation, but it's not really as easy as that: if people run in to performance troubles they'll think the new OS sucks, especially if they're gamesr. But WINE has been remarkably successful at preserving performance, so maybe it's still possible. It's cleanliness that should matter here, and stability - but if MS was to re-wrap their own layer around a BSD base there shouldn't be a problem. Heck, some of their own subsystems started out as BSD copy-n-hack jobs.

Re:good stuff

[losinggeneration]

QUOTE: have a good position in the server market

around 20% isn't too bad, but considering a few years ago they had almost 35% of the market they seem to be loosing their footing again for the server market. (I'm just going off of netcraft which like any statistic isn't perfect by any means.) Anyways, for them to improve to me is to not continually lose market-share and not make as buggy of products which they lock people into (a prett good market strategy. But only for so long until people get fed up with it.)

Re:good stuff

[jbplou]

Netcraft stats are meaningless, what percentage of servers are exposed to the Interent, not a very high number I don't know the percent but only web, dns, and mail servers are exposed for the most part. You need to look at sales numbers to find out, I don't have these but I imagine Netcraft numbers are high because Linux and BSD are good for web servers but not really used for alot of internal applications by companies

Re:good stuff

[losinggeneration]

Looking at sales is meaningless since if you have a talented admin they'll either make their own, or download one for customization ;-)

Re:good stuff

[jbplou]

The majority of businesses uses comercial OS's be it Windows, Solaris, AIX, HP-UX, or one of the Linux's. Your crazy to think people roll thier own. I bet if you just add up Windows, Solaris, OS/400, and AIX you got over 60% of the servers out there.

Torrents are your friends:

[rkrabath]

##### Disk One #####

##### Disk Two #####

Of course, in their infinate wisdom, the coders of slashdot have decided to make my life difficult with their damn lameness filters

Re:Torrents are your friends:

[phpCypher]

I wish CVS could integrate with bittorrent somehow ... until then ::seeding::

Re:Torrents are your friends:

[acidos]

If you buy a set from BSD Mall they now come with both 2 CDs and a DVD that has all sorts of extra packages on it, for the same price that just the 2 CDs would cost you in the past.

Re:Torrents are your friends:

[smokeslikeapoet]

For those of us with AMD64 systems try this one:
http://people.freebsd.org/~kensmith/5.4-torrent/5. 4-RELEASE-amd64-all.torrent
I'm going to test it's stability against my gentoo and ubuntu amd64 solutions. I'm currently happy with neither. Not necessarily because of the distros themselves. But from the lack of 64-bit support from certain apps.

Re:Torrents are your friends:

[alexhs]

Huh ? But you don't dare to install FreeBSD from a CD set !

You network-install it, right from the three floppies

Re:Torrents are your friends:

[cerberusss]

If you buy a set from BSD Mall

While you're there, check out the plush daemon. Your girlfriends will love it :D

Re:Torrents are your friends:

[TCM]

While you're there, check out the plush daemon. Your girlfriends will love it :D

Do they make hand puppets now as well?

*runs*

Fifty Dollars.

[Anonymous Coward]

I'll come over to your house and install it for free. If you want me to shower first, that costs extra.

Re:Fifty Dollars.

[kevcol]

RMS- is that you?

Re:Fifty Dollars.

[ModMeFlamebait]

I'll come over to your house and install it for free. If you want me to shower first, that costs extra.

In his house? Eww.

Re:Fifty Dollars.

[essdodson]

I'll come over to your house and install it for free. If you want me to shower first, that costs extra.

Must be a linux user, we freebsd users shower.

Bloody Thieves!!!

[The Ancients]

Software is released that someone has obviously put a lot of time and effort into, and within five minutes of the story on /. someone's posted torrent links.

What is it with you people???

You make me sick.

Re:Bloody Thieves!!!

[discogravy]

oh, don't feel too bad -- the site that has the .torrents for download is slashdotted, so it's not like there are as many folks helping the download happen as there could be. :)

Re:Bloody Thieves!!!

[rkrabath]

I know you're modded funny, but I'm not sure you're actually kidding.

My view: They have released it for free, and provided links on their site to get it easily, again for free. I might as well encourage people to use BT so the fine developers don't have to pay as much for our free downloads.

Of course people should buy the CDs. But failing that, at least keep the bandwidth bills down.

VIA CLE266/VT8235 USB support

[green pizza]

No mention of it in the release notes, I wonder if USB finally works properly on the VIA CLE266 / VT8235 chipset. That's the only thing that keeps me on Linux.

Re:VIA CLE266/VT8235 USB support

[Markos]

Forgive my ignorance but...

What exactly is the issue? Is it a lack of support, or existing bugs that make it unusable?

Situations like this make me think that it'd be nice if there was some sort of compatiblity goal so that drivers could be somewhat crossplatform, or at least easily ported acrosss the various unix's via some sort of shared api or something.

I'm sure there are 101 reasons why this isn't possible, but it's nice to dream :)

Re:VIA CLE266/VT8235 USB support

[Zemplar]

"No mention of it in the release notes, I wonder if USB finally works properly on the VIA CLE266 / VT8235 chipset. That's the only thing that keeps me on Linux."

Does any VIA chipset actually work properly?

Re:VIA CLE266/VT8235 USB support

[Len]

Does any VIA chipset actually work properly?

I never had a problem with my KT333 board.

Of course, it was running Windows.

Re:VIA CLE266/VT8235 USB support

[infiniti99]

I'm sure there are 101 reasons why this isn't possible

Well it shouldn't be impossible, just improbable. :) Considering the Linux kernel often breaks compatibility with its own drivers between minor versions, I don't think we have a prayer for cross-kernel driver compatibility.

Re:VIA CLE266/VT8235 USB support

[arved]

I never noticed a problem with the CLE266 using ukbd(4) & ums(4). I tried umass(4) and axe(4) for a short time, and they seem to work too.

Re:VIA CLE266/VT8235 USB support

[cant_get_a_good_nick]

There was a commercial supported attempt at this, the Uniform Driver Interface project. It was an environment loaded into kernel space which provided APIs for memory management, interrupts, memory mapped I/O, and all of the normal stuff you'd expect for a driver. There were no synchronization primitives, your code segment was to be run until you yielded and would not be interrupted. as such you had responsibility to compartmentalize your code into smaller chunks, with the benefit of having someone else deal with all the synchronization stuff. It was somewhat message passing, a comparison to DragonFly BSD probably wouldn't be inappropriate.

Cool parts: device drivers are source code compatible on all platforms, no code changes at all. Binary compatible for platforms that share an ABI (Can compile adriver on Linux x86 and have it work for UnixWare x86).

But it never really went anywhere outside of Caldera. That website looks pretty much what it looked like in year 2000. Most of the issues we dealt with were non-technical. One big one being the implosion of Caldera (which became NewSCO). Caldera wanted this environment to make it easier for people to move device drivers from OpenServer (which had a lot of device drivers but was essentially a dead codebase) to UnixWare (which was newer, SVR4.2, then became SVR5).

There was/is a Linux environment, but RMS hates the fact that it's easy to have binary only drivers, making it less likely to release source. Obviously Microsoft wouldn't use it since everyone makes drivers for windows, and that's their competitive advantage. I forgot why the BSDs never used it. I don't think it was licensing, we used a BSD style license.

Re:VIA CLE266/VT8235 USB support

[Xophmeister]

Works for me... using umass and all uhci, ohci and ehci. I can probably remove some of those, but I don't really know what I'm doing ;) My thumbdrive clocks in at 40Mbps; seems a tad underpar, but before 5.4, it was 1Mbps, so I'm easy!

Let's just hope I can get X to work properly with the CLE266...

Help promote their new torrent option,

[Anonymous Coward]

Help promote their new torrent option, seed it for a bit me me and the 5 others doing it currently.

http://people.freebsd.org/~kensmith/5.4-torrent/

if you can, join the all seeds ; )

Re:Help promote their new torrent option,

[KozmoStevnNaut]

Where are the ones for Alpha, I wonder?

Mostly I want to see if 5.4 can disklabel my Alphaserver 1000A disks. 5.3, 5.2.1, 5.2 can't even boot, and 5.1 can't label the disks :-(

Free BSD

[a3217055]

Congrats Well awaited, will install and give it a try. Sorry not top of the line hardware... But then what about Debian, Debian is like dreamer in high school. J/K But BSD is well welcomed, I run BSD on my laptop but after some stand offs it is one of the most nicest systems I have used. But I always ask this to the Linux guys at my compnay ( ps I also run linux ) why did linux get the market it has now and not BSD ? Even thought BSD has a lot of cooler things . . . PS Apple OSX is not BSD, it is a lot like your lil'sister who gets involved with the wrong type of guy in the adult industry.

Re:Free BSD

[dmaxwell]

But I always ask this to the Linux guys at my compnay ( ps I also run linux ) why did linux get the market it has now and not BSD ?

Just plain marketing for one. *BSD can and probably is better by any number of measures. "Better" doesn't always equate to "sexier".

The other reason is that GPL can be more business friendly than the BSD license. The trick here is that the GPL is picky about which businesses it is friends with. For strategic reasons, a company like IBM can open something up but place the contribution under the GPL. It is perfectly free from an end user point of view but will require re-implementation on the part of a competitor who wishes to use knowledge from the code in question. This takes nothing away from scenarios where the BSD license is more "business friendly". Personally, I find the "moral" arguments around all of this induce finger drumming. If the choices were BSD or nothing or GPL or nothing then I expect we'd see much less funding of interesting projects by business.

Re:Free BSD

[menkhaura]

Apple OSX is not BSD, it is a lot like your lil'sister who gets involved with the wrong type of guy in the adult industry.

I couldn't have said better! Mod me offtopic if you will, but please stop saying MacOS X == BSD!

Re:Free BSD

[vendull]

why did linux get the market it has now and not BSD ?

There are a number of reasons for this. In the beginning, Linux has better support for low end hardware (e.g.: 386's). FreeBSD was geared only towards the server market at that time. For example, FreeBSD only supported SCSI drives for a while while Linux supported IDE drives.

Another factor was that Linux development was happening in public, through Usenet, while FreeBSD development was happening on private mailing lists. This was before the web had really taken off, and Usenet was quite popular.

In addition, the USL lawsuit against BSDI and the Regeants of the University of California was in full swing when Linux was starting to gain momentum. Lots of folks were afraid that if the lawsuit went USL's way, that it would be very bad news for the BSD variants.

There is a paper that expands on some of these details here: http://www.freebsddiary.org/linux.php. I didn't write it, but it is an interesting read.

Re:Free BSD

[Brandybuck]

The other reason is that GPL can be more business friendly than the BSD license.

That doesn't explain the pre-commercialization days of Linux. Is the GPL really more business friendly than the BSD license for a one man firm ten years ago? Hardly! They weren't worried about proprietary companies "stealing" their shell scripts because too many other one-man Linux outfits were "stealing" it instead!

Instead Linux's popularity can be attributed to two other things, in my opinion. First, BSD got bushwhacked by AT&T/USL. Second, Linux required integrating hundreds of different "parts", which resulted in commercial and semi-commercial firms providing that service, which in turn resulted in *marketing*.

Re:Free BSD

[diegocgteleline.es]

Just plain marketing for one. *BSD can and probably is better by any number of measures. "Better" doesn't always equate to "sexier".

So, uh, do you have numbers around? AFAIK even some freebsd hackers admit that linux is kicking their asses in more than one field. It's not strange to see post from freebsd users benchmarking databases etc. against suse/redhat and getting better numbers with them. And let's not start talking about features.

So no, Freebsd is not "just better" and linux is not just about marketing. Sorry.

(It certainly is in some places like which fast forwading routing packets, but saying that freebsd is "probably better by any number of measures" and "linux is about marketing" having docens of paid IBM/SGI/redhat/suse programmers around making it fast and reliable...oh my)

Re:Free BSD

[dmaxwell]

I'm a Linux user myself; but let's give props where props are due. One measure in which most any BSD is better is integration. BSD has been maintained as a coherent system since before Linux has even existed. Their userland has a bit less evolution and tad more design in it. The init scripts are arguably better due to their relative simplicity. As for features that BSD lacks, that can be a feature as well. Simplicity often =='s robustness. The individual flavors also have their own merits. There is OpenBSD's well known penchant for correctness and security. NetBSD runs on even more arches than Debian.

I'll also point out that the BSD's tend to be more predictable in their quality from release to release. There have been some real brown paper bag kernel releases and distros like RedHat and Mandrake have pulled boners on their own.

I'll bet a real BSD fanboy could probably think of a few more.

Re:Free BSD

[Thunderbear]

I think it was because the Linux license was more attractive to the large number of coders who came online at that time. You were guaranteed that everybody benefitted from yours and others work.

The social implications of this are in my eyes very important.

Additionally Linus (IIRC) did a lot of work to get the user submitted patches in the kernel very quickly. This was an instant positive feedback, which also is very important for you to feel part of the developmental process, and to encourage you to do it again.

For some reason the BSD's did not get that synergic effect - perhaps because they had already demonstrated that they would fork and that work had to be replicated in 2-3 source trees.

Re:Free BSD

[MrHanky]

FreeBSD's base is well integrated, I'll give you that. But I've used several versions, and for some reason GDM never worked in a fresh install, no matter how I did it, from packages or ports. Some modifications would some times fix it, but once I would be logged in as root no matter which user's account and password I logged in with. I've never seen this level of brokenness in a common package in any of the Linux distros I've used since 1999.

Don't get me wrong. I like FreeBSD, and I've used it quite a bit. I just don't feel its integration is superior to Debian's when you consider other packages than what you find in base.

Re:Free BSD

[Trix]

But I always ask this to the Linux guys at my compnay ( ps I also run linux ) why did linux get the market it has now and not BSD ?

Probably, at least in part, because of the USL v. BSD lawsuit in the early nineties.

Most people don't seem to realize that the BSDs (NetBSD & FreeBSD, anyway) and the Linux kernel both got started at about the same time. (Yes, both are predated by the GNU project, but that's another posting entirely).

The net is rife with stories about the Berkeley folks getting sued by AT&T/UNIX System Laboratory. I think that did a lot to hamstring BSD in the eyes of the general public - - even though the products of the BSD team, like TCP/IP, have enjoyed widespread acceptance.

The lawsuit was kind of analagous to the way that SCO is trying frame its case against IBM; "You're using our code without our say so."

In the BSD case, it turned out that AT&T et al were using more BSD code than BSD had AT&T code.

Re:Free BSD

[top_down]

No, Linux attracted those companies because it was big and growing fast already. And the lawsuit is just a convenient story to blame someone else for the failure to win the battle for the user. It wasn't that important.

Imo the big problem of freebsd was (is?) that it is an OS by specialists for specialists (and specialist wanna bees). These people were not interested in marketing and helping out newbies they were focused on building the best OS available.

The early Linux community was very different in this respect. Many had only just switched from some commercial OS and were Unix newbies themselves. They were in a better position to recruit new users and did so enthusiastically, aiming for 'world domination' rather than technical excellence.

Re:Free BSD

[someonehasmyname]

I'd stop saying it if it were completely true.

While Apple's OS X has a very NextStepish Mach MicroKernel, the entire userland is from FreeBSD 5.

I'm sitting at a Tiger workstation right now, and if I open a terminal window and type "man ls" I will get the _exact_ same man page a FreeBSD 5 system shows. All the userland utils in OS X understand the same switches as the FreeBSD userland utils. They are the same.

It's like "GNU/Linux" really. Linux is Linux, but it heavily relies on GNU userland utils, so RMS wants it to be called "GNU/Linux".

At least the BSD developers aren't assholes, screaming about calling it "BSD/OS X".

Re:Free BSD

[MrHanky]

Is that so? After googling around a bit and visiting sites like Secunia and Securityfocus, the most recently found vulnerability in GDM seemed to be fixed in October 2003. XFree86 and X.org have had loads of problems since then.

tail -f *log

[_Sharp'r_]

From the release notes:

"The -f option of tail(1) utility now supports more than one file at a time."

That enhancement alone is worthy of upgrading!

Re:tail -f *log

[Anonymous Coward]

That enhancement alone is worthy of upgrading!

Never heard of xtail? It was released in 1989 and does exactly that.

Re:I hope it's better than 5.3

[bcrowell]

I was running FreeBSD as my desktop OS for quite a while (from the 4.x through the beginning of 5.x), and am still running it on my server. Recently I've switched my desktop machines to plain Debian and Ubuntu. At the risk of being too factual and objective :-), here's what I see as the pros and cons:

Good things about FreeBSD:

1. Variety is good. If we wanted a monoculture, we'd be running Windows. It's cool that we have an OS that's under a non-viral license, as well as one that's GPL'd.
2. It's a single, consistent system. You can get a book (or actually, several books) describing FreeBSD, and get a description of exactly the system you're using. With Linux, you have, e.g., O'Reilly's Running Linux, which says, "If you're running Red Hat, then..., but if you're running Debian, ... etc."
3. If you want something rock-solid to run a server on, and you don't need support for the latest hardware, FreeBSD is great.

Reasons to prefer Linux:

1. If you want support for the latest hardware, you either need to run Linux, or FreeBSD 5.x, and FreeBSD 5.x is somewhat flaky.
2. The FreeBSD ports system is not all it's cracked up to be. Stuff is constantly breaking. The desktop apps just aren't maintained carefully enough (not surprising, since FreeBSD is not a major desktop OS). After a cvsup, you get left wit a system in a state where you can't upgrade one piece of software without breaking a lot of other software. Portupgrade is a disaster -- I've never seen a better way to bork a system than to unleash portupgrade on it.

Commercial flavours of unix maybe.

[Some Random Username]

Netbsd and openbsd are just as fast as freebsd with the fixes, and so are most linux distros. Its really only commercial unix vendors that are slow with the fixes.

GUI to desktop

[guildsolutions]

I still contend that all BSD and linux developers need to work together on making a stable, solid, well functional GUI like OS X for the X86 arch, so that it can compete against windows on a larger scale for the desktop position. Without forcing microsofts hand with a good, stable well working GUI... Its never going to happen.

Re:GUI to desktop

[LittleLebowskiUrbanA]

And that is why you fail.... Seriously, that's not the goal. OSS is your way, the way you want it. Not to replace some software company

Re:GUI to desktop

[rubycodez]

maybe that's what GNUStep will be, give it a couple more years

Re:GUI to desktop

[Aldric]

Not all programmers are good GUI programmers. I'd say leave it to those who are best at it.

Re:GUI to desktop

[mickyflynn]

Even Linus has stated numerous times, for instance, in his essay for "Open Sources" that taking on Microsoft was not the goal. Hell, Linux wasn't even meant to replace UNIX. It's all just a happy accident.

BSD wasn't meant to replace anything. It was a project by a school to do stuff which got turned into a personal project to replace that in order that it could continue unhindred.

I really dont understand what this obsession is with "replacing" microsoft. Most people really don't care. Nothing is stopping you from doing what you're doing. Why try and stop them?

I fear the "hacker spirit" of learning and invention has been replaced by a generation of people who use Linux just to be "l3t0 k3wl" and hate on Microsoft instead of trying to make the software better, do neat things, and learn how the computer works. Its sad really. It also show a lack of disconnect with the real world in which real work must be done. Not everyone can sit around and hack up their system and play on the computer all day. Most people dont care. Windows 2000 Pro is actually not that bad. I have no problems with it, myself. I can do my work and then have fun. In high school (graduated '02), I spent more time with my computer in parts and recompiling and tweeking shit than I did in a word processor and working. My grades suffered accordingly. Of course, AbiWord was nothing to write home about then, OO didn't exist and StarOffic wasn't "Free" as in no money unitl I was a senior, iirc.

But still, its a little irrational and vitriolic, all this "lets conqurer MS!!" not that I didn't engage in it myself when I was like, 14.

Might be a stupid question, but...

[stanthegoomba]

what is the easiest way to upgrade FreeBSD? I installed 5.3 a couple months ago and I want to know how to safely update to 5.4 without downloading and burning another CD.

Re:Might be a stupid question, but...

[menkhaura]

Create a standard-supfile with the following content:

*default host=cvsup2.FreeBSD.org
*default base=/usr
*default prefix=/usr
*default release=cvs tag=RELENG_5_4
*default delete use-rel-suffix
*default compress
src-all

(I like to put it under /etc). Then you can run the following command:

cvsup -g -L 2 /path/to/your/standard-supfile
Go make some coffee while your sources are synchronized, then read the Handbook to learn how to build the beast.

Re:Might be a stupid question, but...

[sn00ker]

cvsup.
pkg_add -rv cvsup-without-gui, then take a look at the example supfiles in /usr/share/samples/cvsup
Once you've brought the system up-to-date (if you just want to go to 5.4, set the release tag to RELENG_5_4), follow the instructions in the Handbook on building the world.

That handbook section covers all the stuff I've mentioned above. The Handbook is your friend.

Re:Might be a stupid question, but...

[SpinJaunt]

there is some interesting stuff in /usr/share/examples:

% ls /usr/share/examples
BSD_daemon bootforth drivers ipfw mdoc pf printing smbfs
FreeBSD_version cvs etc isdn meteor portal scsi_target startslip
IPv6 cvsup find_interface kld netgraph ppi ses sunrpc
atm dialog ibcs2 libdialog nwclient ppp slattach tcsh
bc diskless ipfilter libvgl perfmon pppd sliplogin worm

Look in /usr/share/examples/cvsup/ and there should be some "templates" or use as-is SUPfiles. Oh and watch stable-supfile, it's 5-STABLE, not 5.4-STABLE "Patch/Fix branch.

Re:Might be a stupid question, but...

[Anonymous Coward]

Or find the fastest ;)

http://fastest-cvsup.sourceforge.net/

5.4 Dedication

[Anonymous Coward]

The FreeBSD 5.4 Release is dedicated to the memory of Cameron Grant. Cameron was an active FreeBSD Developer and principal architect of the sound driver subsystem despite his physical handicap. His is a superb example of human spirit dominating over adversity. Cameron was an inspiration to those who met him; he will be fondly remembered and sorely missed.

http://www.freebsd.org/releases/5.4R/announce.html

Re:5.4 Dedication

[KozmoStevnNaut]

Here's to you, Cameron. I didn't know you, but I'm sure you were a great guy.

I'm having a beer for all the people who have overcome their physical and mental handicaps and done everything they could to function normally in society, as hard as it might have been. And also for their families who helped them through everything and never stopped loving.

Requiem for the FUD

[Anonymous Coward]

... facts are facts. ;)

FreeBSD:
FreeBSD, Stealth-Growth Open Source Project (Jun 2004)
"FreeBSD has dramatically increased its market penetration over the last year."
Nearly 2.5 Million Active Sites running FreeBSD (Jun 2004)
"[FreeBSD] has secured a strong foothold with the hosting community and continues to grow, gaining over a million hostnames and half a million active sites since July 2003."
What's New in the FreeBSD Network Stack (Sep 2004)
"FreeBSD can now route 1Mpps on a 2.8GHz Xeon whilst Linux can't do much more than 100kpps."

NetBSD:
NetBSD, for When Portability and Stability Matter (Oct 2004)
NetBSD sets Internet2 Land Speed World Record (May 2004)
NetBSD again sets Internet2 Land Speed World Record (Sep 2004)

OpenBSD:
OpenBSD Widens Its Scope (Nov 2004)
Review: OpenBSD 3.6 shows steady improvement (Nov 2004)
OpenSSH (OpenBSD subproject) has become a de facto Internet standard.

*BSD in general:
Deep study: The world's safest computing environment (Nov 2004)
"The world's safest and most secure 24/7 online computing environment - operating system plus applications - is proving to be the Open Source platform of BSD (Berkeley Software Distribution) and the Mac OS X based on Darwin."
BSD Success Stories (O'Reilly, 2004) (pdf) ~ from Onlamp BSD DevCenter
"The BSDs - FreeBSD, OpenBSD, NetBSD, Darwin, and others - have earned a reputation for stability, security, performance, and ease of administration."
..and last but not least, we have the cutest mascot as well - undisputedly. ;)

--
Being able to read *other people's* source code is a nice thing, not a 'fundamental freedom'.

Re:Requiem for the FUD

[monkeyman_67156]

Mod grandparent down. This is the same guy that always posts in every BSD story.

stable development?

[jbellis]

which is it? :)

Re:I hope it's better than 5.3

[menkhaura]

FreeBSD 5 is now the stable branch, and 4.x became legacy.

Please enlighten me, how is FreeBSD 5.x flaky?

The ports system works almost perfectly for me (FreeBSD is my main desktop and it is on one of my test servers). Keeping curent is simply a matter of make update && portupgrade -a every morning (I don't even run cvsup manually), with the occasional hiccup that is easily solvable by yourself or with a little help from TFM, freebsd-questions@lists.freebsd.org, or freebsd-ports@lists.freebsd.org.

Re:BSD is dying 5.3 released

[Anonymous Coward]

Exactly. Get some throw-away e-mail accounts and register some slashdot accounts with them. Then make a bunch of posts pandering to whatever views are the most popular, thus building up your mod points. Note that you don't have to really mean what you write, you're just trying to build up points. Be sure to spread your points out among your various accounts. Kind of like building up a nuclear stockpile or patent portfolio, you can then use these points to blast anyone you later disagree with or just don't like don't like. Its great fun, but AC's aren't allowed to play!

Re:Better SMP support? Better MySQL performance?

[mi]

Anyone from FreeBSD know for sure if the fixes above will help bring FreeBSD up to par with Linux as far as MySQL performance on SMP machines go?

Probably not, because certain very busy people have -- once again -- forgotten to turn off INVARIANTS in the threading libraries' Makefile :-(

I doubt, many benchmarkers will bother turning these off on their systems and recompiling libthr/libc_r ...

Re:BSD is dying 5.3 released

[skingers6894]

You're a funny lad.

I don't think Mod points are as powerful as a "nuclear stockpile" but you may have a point in terms of the "patent portfolio"

I'd mod you up to "amusing" if I could...

Re:I hope it's better than 5.3

[saleenS281]

I have a feeling you haven't touched 5.x since it's first iterations. Granted, the VERY FIRST 5.x releases were QUITE flaky. But since 5.3 was released it's been nothing but stable for me.

As a desktop O/S, I'm sure you did have issues, but then, that was never it's intention.

What exactly have you had portupgrade break? I've never had a single problem with it. If nothing else it's helped to take care of my dependency woe's.

Does SATA work right now? on SIS965L southbridge

[pkphilip]

Does anyone know / has anyone tested if FreeBSD 5.4 supports the SIS965L southbridge chipset ?

And also, how is the support for SIS190 Ethernet? Sorry about posting these questions here, but I haven't received much response in the kernel mailing lists (atleast not on Linux).

Re:I hope it's better than 5.3

[sp0rk173]

If you want support for the latest hardware, you either need to run Linux, or FreeBSD 5.x, and FreeBSD 5.x is somewhat flaky.

Not true. Device drivers are usually backported unless they depend on some system difference between 4.x and 5.x. I've never noticed any hardware incompatibilities between versions and i've used both extensively.

The next one is a doosy...

The FreeBSD ports system is not all it's cracked up to be. Stuff is constantly breaking.

I honestly have not encounted a break in any major apps in ports in the past 3 years. It's evolved a lot since you last used it, i guess.

The desktop apps just aren't maintained carefully enough (not surprising, since FreeBSD is not a major desktop OS). After a cvsup, you get left wit a system in a state where you can't upgrade one piece of software without breaking a lot of other software. Portupgrade is a disaster -- I've never seen a better way to bork a system than to unleash portupgrade on it.

No, no no. Not true. I had a production system with apache, php, postgresql, gnome, KDE, etc installed (it was a workstation/light-use webserver for a lab i was working in). I installed it at 4.5, last time i touched it it was at 4.11, all ports upgraded (using cvsup and portupgrade), only one install point. After being a FreeBSD user for about a year. If I can do it, in a production environment, without any break in's or security issues, anyone can. My webserver here at home has been running 5 since 5.2.1, same deal - all things installed from ports, only one point of install, all upgraded by cvsup and portupgrade. No problems. Then there's my workstation, it runs Gentoo, Windows, Solaris and FreeBSD 5.3. FBSD has been installed since 5.3 first made -RELEASE, runs gnome 2.10 (which hit ports before it hit portage, ~1 week after official release). Only one install point, constantly updated using cvsup and portupgrade. Gentoo? Great little distro, but i've installed it at least 3 or 4 separate times due to major breakages or just aggrivation with portage. I don't hold it against portage, it's just still maturing.

Your report couldn't be further from my experience. Ever since i started running freebsd back four years ago i've been able to keep an up-to-date, stable system without much difficulty.

Re:Does SATA work right now? on SIS965L southbridg

[menkhaura]

I have no idea, but you may try FreeSBIE, a FreeBSD on a Live-CD.

So what's 5.4 like for 4.x users?

[gkitty]

I'm running FreeBSD 4.11 on my servers and laptop. To this point I've avoided 5.x due to rumours of a performance downgrade for my single processor systems. (Well that & pragmatism, I am slow to change good systems.)

I have no great dissatisfactions with my 4.11 systems, but could make use of cardbus support on the laptop, and if 5.x's threading could improve the performance of samba, mySQL, postgreSQL, maybe java and apache (my primary server loads) that would be a win.

Is 5.4 ready for primetime for satisfied 4.x users? What are the real world performance implications at this time in terms of memory usage, I/O throughput, performance on my favorite server apps and as a GUI workstation? Is it as solid as 4.x has always been?

Re:So what's 5.4 like for 4.x users?

[kaiwai]

Personally, I'd wait till 6.0 comes out; they're moving to time based releases, and hopefully by then, they would have optimised it enough to provide a decent enough performance improvement over 4.x to make it a worthwhile move.

On the side issue, however, it would be nice if they actually updated their website, inreference to the smp, kse and busdma status; I mean, not to sound whiny and flamebaitish, but it'd be nice to see a status of how things are *ACTUALLY* progressing *NOW* rather than from up to 12months ago, in regards to the kse status page (since it was last updated).

Re:So what's 5.4 like for 4.x users?

[chrysalis]

If you like FreeBSD 4.x, there's a better upgrade path: DragonFlyBSD. You will probably get better performance out of it and as long as you stick with the RELEASE or PREVIEW tags, DragonFlyBSD is extremely stable.

Re:So what's 5.4 like for 4.x users?

[dinivin]

Yes, the DragonFly base system is extremely stable. Just don't try and build anything big from ports or pkgsrc, cause it'll probably fail.

Re:So what's 5.4 like for 4.x users?

[setagllib]

With minimal/no tinkering I could get a full Apache2+MySQL+mod_php rig up from pkgsrc in DragonFly BSD. X is another matter, but it's been done.

There are worse problems than that though. I recall having mysterious behavior (also seen on mailing lists) when trying to forward things to a local FTP proxy, which is the only way to have transparent FTP NATting with PF (and IPFW/natd just didn't work at all, but I might have just missed something: it's been years since I last used it). So it has some caveats as a gateway, but if you're willing to work around them it's a great system.

Personally I'm waiting until all of the other important work is done first, finally revealing the power of their SMP and VFS implementations and so on. We could either have a strong contender for Linux' position of "does everything fast enough without being too complicated", or a depressing failure (which is more likely to be from lack of software support than any developer issues: there's little point running DFly if the package manager issue isn't resolved).

Re:So what's 5.4 like for 4.x users?

[Brandybuck]

I've had absolutely no problems with 5.2 and 5.3. 5.0 and 5.1 were a bit flaky, but that's long gone. There is no performance downgrade. Stop listening to the DragonFly fudsters, FreeBSD 5.x is fast and stable.

Re:I hope it's better than 5.3

[jadavis]

I am about to make my first attempt at a preliminary freebsd box that will turn production eventually (webserver).

Can someone please set me in the right direction with some basic guidelines like:

(1) What version should I use? I assume 5.4 is too new, but 4.x is too old, right?

(2) Are the defaults too conservative? If so, how should I tune it for webserving (i.e. what are some sane settings)?

(3) How well does SMP work? Is it pretty much out-of-the-box, or do I need to mess around with it?

(4) Are there any linux -> freebsd gotchas?

(5) Is there a linux -> freebsd guide?

(6) One of the primary reasons I'm trying to run the webserver on freebsd is security. Linux is fine, so long as you upgrade the kernel every release (which seems stupid to me), but I'm looking for an OS that I can forget about for a while longer. Can I expect to see some improvement in this area?

I can find stuff like "how to use ports" and all that, and I've run a basic freebsd os before with all the software installed. I just need something in between to tell me what to do to make a webserver (a real one with real settings).

Re:I hope it's better than 5.3

[evilviper]

If you want support for the latest hardware, you either need to run Linux, or FreeBSD 5.x, and FreeBSD 5.x is somewhat flaky.

FreeBSD 5.x is a bit flakey compared to v4.x, but plenty stable when compared with Linux. Not trying to flame Linux users out there, but modern releases of Linux (ie. not Debian-stable) are far more unstable than FreeBSD 5.x.

The FreeBSD ports system is not all it's cracked up to be. Stuff is constantly breaking.

This is really ridiculous. I've practically never heard anyone say that ports has broken on them. I get the feeling you're just subtly trolling.

Re:I hope it's better than 5.3

[bark]

I can answer a few of those questions. I've been using freebsd in a production webserver / mysql server environment since 4.9

1) Use 5.4. The 5.x tree is stable now, to use anything older than 5.3 is to invite yourself to mysterious flakeness, as other posters have said, earlier releases of 5.x such as 5.1 and 5.2 are not really rock solid. But 5.4 is pure gold.

2) The defaults are not too conservative, because the system self tunes itself according to the usage of the machine. Beware of tuning the system yourself through the sysctl's. The same with Linux, you wouldn't want to tweak the system until you are comfortable with it enough to know what to tweak.

6) There is a binary upgrade tool for patching security errors, but as people mentioned before, the "cvsup", then "buildworld" process makes sliced butter of this aspect of upgrading to patch security errors. Basically, the source to every single binary in the base system is included in the release, and updated together, so you can rebuild your entire system with all the security fixes included whenever you want.

FreeBSD Installation

[alexhs]

I currently run FreeBSD 5.3 on the home server. "It just works".
I suppose you could go with 5.4. I will do, soon.
If you have broadband, just download 3 floppies, then install from the network. Quick'n'easy.

Firewall is disabled by default. To enable it, you will need to recompile a kernel (you need to install with sources). Note that you need a firewall if you intend to do some sort of NAT/masquerading/diverting.
Be warned that this isn't straightforward as in Linux : you need to manually change configuration files. Some "howtos" here and here

SMP works well (but wasn't tuned for SMT/HT not a long time ago). Don't know if it is working out-of-the-box, I don't need it on a pentium 166 MMMX, but as you're going to compile a kernel anyway...

Re:FreeBSD Installation

[jaseuk]

>Firewall is disabled by default. To enable it, you will need to recompile a kernel (you need to install with sources).

You don't need to recompile the kernel. pf & ipfw are available as modules.

I suggest you have a look at /boot/defaults/loader.conf.

ipfw_load="YES"
pf_load="YES" etc.

or

kldload pf
kldload ipfw

There are very few reasons why you'd need to build a new kernel in most situations.

Its a shame that this isn't made very clear in the handbook. The only situation where you might want to recompile a kernel is for SMP, but using freebsd-update you can just download a precompiled one.

Most of the docs tell you how to recompile the kernel to enable a feature and then as an after thought mention the kernel modules.

Really all those kernel compiles should be stripped from the handbook. As its completely unnessassary and scares off users.

Jason

Heh

[Anonymous Coward]

I've been running 5.4 since Friday, when the source was tagged. cvsup is wonderful. :)

Know your weather!

[LFS.Morpheus]

Please see hail versus sleet.

Quick summary:
* Hail is ice pellets produced by strong thunderstorms, and is most likely in the summer (as that's when strong thunderstorms are most likely to occur).
* Sleet is re-frozen precipitation, caused by snow that has been melted and re-frozen on its way down.

While it is quite likely you were indeed experiencing hail, not sleet, hail is not uncommon in May if a strong front passes through.

2/3 isn't bad though, I'd give a 70% forecast that the end is indeed near. In the meantime, be on the lookout for hail...

Re:I hope it's better than 5.3

[DrSkwid]

I've never noticed any hardware incompatibilities between versions and i've used both extensively.

Not extensively enough to try Bluetooth on 4.x

I had a horrible situation where I had to dual boot for my Wavelan (4.3) and my Bluetooth (5.1)

Does it have decent ext2 support yet?

[Viol8]

I dual boot a 4.6 install with my linux install. My home dir is in ext2 format and I'm always worried when working under BSD that something in my home dir will corrupt (since the BSD developer(s) warn their ext2 driver is not 100% kosher) so I tend to do only small amounts of work in it. Does anyone know if the ext2/ext3 support is now rock solid or is that still on their to-do list?

Re:Does it have decent ext2 support yet?

[Brandybuck]

Nah, it's still meant for read-only use. Make a proper UFS2 partition for your home partition. This means you can't share one home between two operating systems though.

Re:I hope it's better than 5.3

[kan0r]

I recently installed 5.3-RELEASE. The *freshly* installed system didn't unmount filesystems (ext2 and ufs) when performing a shutdown or reboot. Fsck on every single startup until I worked around it: I unmount the filesystems now "by hand" in rc.shutdown. And also, portupgrade broke my X. I could of course fix it, but it was far from a clean upgrade. Since BSD is not a real desktop system, thats not too bad of course, but still...

Re:Better SMP support? Better MySQL performance?

[Anonymous Coward]

I benchmarked the code from CVS about 2 weeks ago, and although the gap is narrower, its still much better to use linux for mysql.

Re:I hope it's better than 5.3

[Just Some Guy]

If you want support for the latest hardware, you either need to run Linux, or FreeBSD 5.x, and FreeBSD 5.x is somewhat flaky.

For the record, BSD supported USB before Linux did. While you're correct in that Linux is likely to have at least partial support for new or odd hardware before FreeBSD, it's certainly not always the case.

The desktop apps just aren't maintained carefully enough (not surprising, since FreeBSD is not a major desktop OS).

Sorry, but I definitely can't agree with that point. I've been using FreeBSD as a desktop OS for years without significant differences from the equivalent Linux setup. It is, however, critically important to read the /usr/ports/UPDATING file and follow the directions in it before randomly upgrading stuff. The biggest problem with portupgrade seems to be that it works well enough that a lot of people expect it to perform miracles.

MOD THE PARENT DOWN

[jack_csk]

What a terrible parent post! If you really can't download the ISO copy and burn it to CD by any mean, here is the place that you shop:
FreeBSD Mall
At least they DO support the FreeBSD development community financially.

Avoid CVSUP on a server

[Billly Gates]

Otherwise you are syncing FreeBSD to -current.

THere is a stable cvs tree but it does not include security fixes. At least thats what i saw. Also in the FBSD 4.x series I saw several ports downgraded for some bizaare reasons. Why I dont know

I broke my system several times from cvsing up

Re:Avoid CVSUP on a server

[kkenn]

Wrong on both counts:

1) The stable branch does include security fixes

2) The ports collection is not branched, so there's no possibility for "several ports downgraded" in the "4.x series". The only situation in which ports are downgraded is if there are serious problems with the newer version, and a reversion to the previous version is a net gain.

Miniinst iso

[eventhorizon5]

Is there a miniinst ISO image for release 5.4? (it's the network install image). 5.3 had one, but there doesn't seem to be one available.

-eventhorizon

Re:I hope it's better than 5.3

[wirelessbuzzers]

I'm going to pitch in with GP on this one. I've been running FreeBSD 5.3 on a file/web/printserver for a while. I'll upgrade to 5.4 now, but this summer when I have more time, I'm going to switch to Linux (probably Debian or Ubuntu).

Why? Debian has a better package manager. I won't have to deal with portupgrade. I won't have to compile to get a package made this month; on a slow machine this is a big win. I won't have to wait 2 months for security fixes (cough MySQL). And FreeBSD has been stable, but not overwhelmingly so. The machine still crashes once a month or so, and half the time hangs during a reboot. Certain of the daemons die at random and thus have to be babysat by cronjobs or runscripts.

There are some things I'll miss though. For one, POSIX ACLs are built into FFS. For another, snapshots: FFS snapshots are much simpler and more efficient than LVM snapshots (any word on when someone's getting NetApp-style snapshots though?)

Simultaneously and independently, the Harvard Computer Society, of which I'm a member, is making a similar switch. They're switching to Debian for easier maintainability, and for grsec and/or systrace. HCS allows student groups to run scripts on their servers, which causes them to get hacked every once in a while (like when the phpbb exploits were 4ll t3h r4g3), and with grsec they should be better able to limit the damage. The main things they'll miss are pf (which they might get a hardware firewall to replace) and dump -L (much easier than LVM snapshots).

correction, 2.0, not 1.0!

[bluGill]

FreeBSD 1.0 cannot be run unless you have a Unix license. I'm not sure what this would cost you, but SCO is selling licenses to Linux users for $699.00, so my guess is about that. However you need to ask SCO, as they are the only ones legally selling such a license.

For Freebsd 2.0 the requirement of a Unix license was eliminated (there were only 7 files to re-implement).

Re:correction, 2.0, not 1.0!

[bovinewasteproduct]

FreeBSD 1.0 cannot be run unless you have a Unix license. I'm not sure what this would cost you, but SCO is selling licenses to Linux users for $699.00, so my guess is about that. However you need to ask SCO, as they are the only ones legally selling such a license.

For Freebsd 2.0 the requirement of a Unix license was eliminated (there were only 7 files to re-implement).

I belive that requirement is no longer valid. It was based on the licensing of V7/32V Unix which was released by Caldera in January 2002. A later release put it under the original BSD license. Here is a Groklaw article talking about the way SCO tried to later say it was on for the 16bit code and non-commericial.

Since the bits of 1.X that were tainted are now release under a BSD license, well..... Thats why you can once again get FreeBSD 1.X if you look around enough.

BWP

Re:Damn

[n0dez]

It's worth upgrading to 5.4-RELEASE -- LINUX: Linux Is Not UniX

making your own world

[n0dez]

If you already cvsup'ed your sources,
make buildworld
make buildkernel
make installkernel
reboot

boot in single user mode, then
mergemaster -p
make installworld
mergemaster
reboot

Voila, you should be running 5.4-RELEASE at this point :)

Re:Alpha?

[adiposity]

Not dropped, relegated to tier 2.

See here.

Re:I hope it's better than 5.3

[wirelessbuzzers]

Ah, didn't know about the systrace port, thanks.

As for binary packages, yes, I know that it's possible to patch them, but ports updates faster when there's say a PHP vuln, so we still have to compile it. We usually update due to security vulns and so time is important.

And while I'm not a ports expert, neither I nor the HCS admins install system stuff via make, and so fixing broken dependencies in the package database is pretty irritating. Maybe there's a way to avoid this problem, but it's not in the Handbook.

Re:I hope it's better than 5.3

[M1FCJ]

I recently installed DragonFly on a laptop and a very old desktop. It had been some time since I attempted a BSD as a desktop and I was quite impressed with its speed (laptop had 48MB ram and Pentium 150 CPU and the desktop had 128MB ram and Pentium 200). None of the modern Linux distributions were as fast and responsive on these boxes as the DragonFly is.

Taking I've been using commercial Unix and Linux flavours for quite some time now, can you explain if I should ditch these two DragonFly's and switch to FreeBSD 5.4? I am seriously thinking of installing my home web server on a DragonFly machine because of its performance on old hardware and simplicity. What would be the pros and cons?

(Moderators: This is a serious question, not a troll so bugger off).

Re:I hope it's better than 5.3

[molnarcs]

If you read the ext2 notes, it is not supported very well. Nevertheless, I have never had any issues with it - except for dirty filesystems after boot. One of them devs sent a patch (Michael Nottebrock) - download it here. - that will unmount your ext2 filesystem before running the rest of shutdown procedure.

Re:I hope it's better than 5.3

[Ratman96]

Then we won't have to use that text-script kludge of an installer that hasn't seemingly been updated since the 2.x era. I really look forward to the day when both graphics AND audio can work on a FreeBSD system without a PhD/ guru being hired to set them up. They already do work. At least for me with a cmipci-based soundcard and a nvidia geforce4 for graphics. It could be your hardware combination: some stuff will just not work. Same luck for Linux and Windows though. The installer is acknowledged as way too old even for the FreeBSD team, but it still works. OTOH i've never been able to install Debian, i just can't figure the installer. Now i use Gentoo, it didn't even have an installer, go figure. I remember when i first tried Linux, when the 2.4 kernel series began, getting sound out of it was a real mess. I didn't know if i had to get OSS or what! I managed to make it work after three weeks or so. Now, with kernel 2.6, it works like a charm! Perhaps it was just my ignorance back then... or bad luck, some hardware just won't work.

Re:I hope it's better than 5.3

[evilviper]

I won't have to compile to get a package made this month; on a slow machine this is a big win.

Either you don't know what you're talking about, or this is a troll. FreeBSD has binary packages for almost all of it's ports available. The only exceptions being those that can't be distributed for legal reasons.

They're switching to Debian for easier maintainability, and for grsec and/or systrace.

Systrace support in Linux is considered unstable. If you wanted systrace (which is a very good program), you'd be switching to OpenBSD/NetBSD.

The main things they'll miss are pf

Bingo, another reason to be switching to OpenBSD rather than Debian.

Re:I hope it's better than 5.3

[menkhaura]

I've heard good things about DragonFly, but I've never used it, so I cannot compare it to vanilla FreeBSD. Maybe you can try it for yourself? All I can say is, I'm impressed with this new release of FreeBSD, they did an awesome job with the new scheduler (meaning, this baby is fast!); it is worth a try.

Re:I hope it's better than 5.3

[kan0r]

Thanks alot for the information. My problem was not only related to ext2. The ufs partitions/slices didn't dismount clean either. I think I am just gonna upgrade to 5.4-RELEASE and see if that solves my problem.
<opinion>
I appriciate the hard work of the FreeBSD developers, but nevertheless I think for a -RELEASE version of FreeBSD, this is somewhat poor.
</opinion>

faculties

[CarpetShark]

So how much is FreeBSD 5.4 going to cost me?

Depends. How much did your hardware cost, and how much do you value your sanity? ;)

No, actually, FreeBSD was pretty sweet last time I tried it. I'd be all in favour of it, if it was ported to more archs.

Re:I hope it's better than 5.3

[suitepotato]

(note that I have been using BSD systems for some 8 years but it's been about a year or two since I had to install one from scratch; I just remember that the install worked a lot better back then)

A lot of things worked better back then but Levitra doesn't do anything for ease of BSD installation and configuration.

It's like Hotel California sometimes. "You can ifconfig any time you like, but you get no ip..." (guitar solo)

Re:I hope it's better than 5.3

[demon]

Well, not that it's an _overwhelming_ thing, but the 2.6 kernel supports POSIX ACLs out of the box with XFS, JFS, ext2/3 and possibly ReiserFS (but I don't trust it further than I can throw it). Just 'apt-get install acl' for the ACL tools, and use POSIX ACLs to your heart's content. I have been using 2.6 kernels on every server I'm running Debian on now - I'm using Debian testing (sarge) on them, due to hardware support, and it's been quite stable. /me crosses his fingers for sarge going stable RSN...

Re:I hope it's better than 5.3

[niteice]

From a UI design standpoint, sysinstall is roughly on the level of Program Manager. It does its needed tasks, but not all that well.

BTW, I'm also a high school freshman, and I find that OpenBSD kicks FreeBSD's ass.

Yes

[toadlife]

link

Download the 'bootonly' ISO image.

Installation Bug, Anyone Else Seen It?

[devphaeton]

Starting with (iirc 5.1.x) I began to see an issue when installing via FTP (using the floppies). While downloading 'base' it would get to about 46% and fail with an:

"Fatal error: Invalid realloc size of 0! - PRESS ANY KEY TO REBOOT"

message. There are a few google references (some people see it at 54%, some at 63%) and there was once a bug report on it. The bug report seems to have vanished, but when doing a test install on an unused computer, i saw the same thing. (I do the test install, because you are essentially left with an unuseable system when this fails).

I can download the full ISO and install it from that with no problems. I just prefer to do it the other way. (iirc the mini-inst iso fails with the above message as well).

Anyone else see this? It doesn't appear to be very common and i've not seen any conclusive reasoning as to what is causing it, just some random speculation from a few different google hits, some may be right, or may not.

thanks.

p.s.- top marks for the FreeBSD team though. This doesn't negate what a great achievement 5.x is turning out to be.