Slashdot Mirror

>> ...As it is now, freebsd and OS X become fragmented, and some fixes in one aren't present in the other. > That is untrue. Apple has contributed to FreeBSD. Apple has even contributed code that was formerly proprietary, HFS+ (file system) code for example.

By saying that the original statement is untrue, you are effectively claiming that *all* fixes Apple made have been pushed back to BSD. I'd like to see some evidence of that, please.

The layer of Mac OS X that uses BSD, Darwin, is itself open sourced by Apple. Not only are there fixes in there, but FreeBSD can take any code at all that they think may be useful. My understanding is that the FreeBSD folks do mine Darwin. Also, some Apple employees have been granted "committer" status for the FreeBSD source.

What Apple says:
"The Darwin layer of Mac OS X comprises the kernel, drivers, and BSD portions of the system and is based primarily on open source technologies."
http://developer.apple.com/library/mac/#documentation/MacOSX/Conceptual/OSX_Technology_Overview/SystemTechnology/SystemTechnology.html

What FreeBSD says:
''Mac OS X is the latest version of the operating system for Apple Computer Inc.'s Macintosh® line. The BSD core of this operating system, Darwin, is available as a fully functional open source operating system for x86 and PPC computers. The Aqua/Quartz graphics system and many other proprietary aspects of Mac OS X remain closed-source, however. Several Darwin developers are also FreeBSD committers, and vice-versa.."
http://www.freebsd.org/doc/en_US.ISO8859-1/articles/explaining-bsd/article.html

Darwin stuff:
"Mac OS X includes a wide variety of open source software from FreeBSD, NetBSD, OpenBSD, the GNU Project, and many more projects each its own vibrant developer community."
http://www.opensource.apple.com/release/mac-os-x-1067/

Other stuff:
http://www.opensource.apple.com/release/mac-os-x-1067/

Since OSX is somewhat based on FreeBSD, wouldn't pkg_add be more suitable?

There's no need to to recompile the kernel, it's a loadable module. Even more, mount_ntfs will autoload the module automatically if necessary from an fstab entry.

Additionally, there's no need to "enable write support". The limited writing ntfs.ko supports is always on provided you didn't do a read-only mount.

http://www.freebsd.org/cgi/man.cgi?query=mount_ntfs&sektion=8

If the box is inside their perimeter, they're going to want at least some degree of control or at least monitoring for it.

If I understand correctly, you need this hole in the firewall because all your devices (e.g. iPhones) which access it, are coming from outside of the LAN (e.g. cellphone network).

These two things raise the question: why have the server in the LAN? If the server were elsewhere, would you care? Possible solution: consider hosting your server at one of these kind of places instead. Sorry you already bought a computer, but since it's yours, you can just take it home and find a new life for it.

Alternatively, maybe the box can somehow be physically in the hospital (i.e. take power from the hospital's wall sockets, have local staff show up with fire extinguishers when it smokes, and so on) but plugged into the network outside of the firewall, so that from IT's PoV it will be outside, just like the rest of the internet. Then they probably wouldn't care.

Mod parent up, since I'm losing that ability to post a better history.

http://www.freebsd.org/cgi/man.cgi?query=roff&manpath=FreeBSD+8.1-RELEASE&format=html#HISTORY

Osanna first [roff] version was written in the PDP-11 assembly language and released in 1973. Brian Kernighan joined the roff development by rewriting it in the C programming language. The C version was released in 1975.

[...]

After Osanna had died in 1977 by a heart-attack at the age of about 50, Kernighan went on with developing troff. The next milestone was to equip troff with a general interface to support more devices, the intermediate output format and the postprocessor system. This com- pleted the structure of a roff system as it is still in use today [...]

Here's just one that I noticed last time I checked in:

http://lists.freebsd.org/pipermail/freebsd-stable/2010-June/057239.html

Perhaps you could illuminate that example?

It also seems that in 2GB systems, lots of hand tuning is required to avoid kernel panics but the Wiki is not kept up to date with the required information. Keeping up with the mailing list seems to be required.

A filesystem can't be considered 'stable' if it panics the kernel without a disastrous underlying hardware failure.

"Then there's also the kernel securelevel, extended attributes/ACLs, TrustedBSD/MAC, and pf/ALTQ which is far superior to iptables. BSD has really been leading Linux in the area of security--Linux is more focused on spreading GPL and getting the media wheel on your USB keyboard to work."

From http://www.freebsd.org/doc/en/books/faq/security.html#SECURELEVEL
"Securelevel is not a silver bullet; it has many known deficiencies. More often than not, it provides a false sense of security."

Linux supports extended attributes/ACLs, see: http://en.wikipedia.org/wiki/Extended_file_attributes

See http://en.wikipedia.org/wiki/Linux_Security_Modules for a Linux approximation to TrustedBSD/MAC (admittedly it does not enjoy wide spread support yet).

For pf/ALTQ see iptables/iproute2/ebtables

Looks like it's another 2-3 hours of compiling packages and the odd several hours of library/package build error resolution for you!

You can use freebsd-update to do a binary update. Also, recompilation of ports is not usually necessary in between minor upgrades (ie. 8.1 to 8.2). Of course, you may have chosen to build a custom kernel and then you need to build it manually. On my dual core CPU with 4GB RAM it takes about 10-12 minutes to build the kernel and 30-40 minutes for world. To deal with etc scripts you can use etcupdate.

Also, if you don't like this way of doing things and you are a more desktop oriented user, you can look at PC-BSD which comes with its own package system for binary packages, while still offering access to the ports system. And PC-BSD 8.2 (which is obviously based on FreeBSD 8.2) was just released, too.

Where does Linux fail where BSD succeeds?

For some people it's the licensing (BSD vs GPL). For others it is the coherence of the system (how many places hide an IP address in Red Hat?). For others, it is a question of style (BSD vs AT&T type Unix). For some, its functionality (I always liked the way the BSD _______ command worked). From some, it's the simple Joy of BSD, or the McKusick - take your pick. For some, it could be the approach taken to a particular problem taken by one of the BSDs, such as the continuous OpenBSD code audits. For some it might be a particular platform maintained as part of the main distribution. For some, it may be the continuing BSD innovations. For some it might be the counter-culture aspect BSD in the Linux world. Plenty more reasons that people could have, including: Linux - 5 letters, BSD - 3 letters. Do the math.

You could say that the only truly popular Unix desktop is Apple's Macintosh running OS X.
Mac OS X: What is BSD?

What's The Greatest Software Ever Written?

OpenBSD FreeBSD NetBSD PC BSD

FreeBSD Mall BSD Magazine

To each his own.

I've been running ZFS v28 on FreeBSD 8.2 prereleases for quite some time now. More or less weekly patches are at http://people.freebsd.org/~mm/patches/zfs/v28/. I have not had anything to report to the maintainer and from what I see, he's mostly cleaning up the code to merit further patches, the functionality is all there already.

Err... freebsd-update should be just fine: http://www.freebsd.org/releases/8.2R/announce.html

It's always amusing to see Debian fans complaining that the end users are always going for Ubuntu instead of "hey, why not choose Debian, it's the original and it's the best!" when Debian keeps making moves like this.

What moves? This is a release announcement. Do you mean that Debian should stop releasing stable versions?

It's already bad enough to think that a new Linux user would want a browser called "IceWeasel" or would understand that it's really just Firefox renamed because of some silly branding/icon tiff with the mozilla folks.

So go complain to Mozilla.

having a bunch of useful drivers removed

There are no drivers purposefully removed, FAFAIK. Are you referring to firmware, perhaps?

a nonstandard kernel!

Nonstandard? Are you referring to the kfreebsd kernel? It is very much standard as released by FreeBSD. Or do you mean the firmware-split in the Linux kernel? That feature has been upstream for years. Or maybe you mean that it's a non-NTOS kernel? I guess you're right on that one, but most Free people would consider that a plus.

Listen, it's ok to do stuff like this if you're really into teh sooper 100% free as in freedom rms-approved purity, but don't subsequently go complaining when ordinary end users don't want it because it's unusable to anyone other than a free software hacker.

Listen, it's ok to use any Linux you like, but don't subsequently go complaining when ordinary distributors release a Free operating system that you woudn't use.

Oh, and regarding Ubuntu: please get back to me when Ubuntu releases a supported server variant that runs on my NAS.

It appears it's a useful feature because many applications allow commands to be embedded in documents - even ones you might not expect, like vim. From FreeBSD's pkg-message for editors/vim:

SECURITY NOTE: The VIM software has had several remote vulnerabilities
discovered within VIM's modeline support. It allowed remote attackers to
execute arbitrary code as the user running VIM. All known problems
have been fixed, but the FreeBSD Security Team advises that VIM users
use 'set nomodeline' in ~/.vimrc to avoid the possibility of trojaned
text files.

I have no experience with that stuff, nor do I own some the components necessary to test. Were I to sing or preform musical compositions to my computer I believe it might cause some type of kernel panic or hardware failure. On the other hand, it would be hard for me to imagine that recording sound doesn't work. Audio support is generally very solid on FreeBSD for output at least. I know there are users doing things like skype and webcams. Webcams are a little trickier than recording audio though so I don't have a guess one way or the other on that one.

This is the most recent update on that end.

http://www.freebsd.org/news/status/report-2010-10-2010-12.html#Webcamd

"Donald Knuth has published a book and a date has been set for the release of Duke Nukem Forever? It's all too much."

There's more yet, Freebsd is getting a new installer.
http://lists.freebsd.org/pipermail/freebsd-current/2011-January/022335.html

Will we stand the strain?

The Freebsd guys better announce it in big bold letters everywhere to cut down on the potential heart attacks as the shock may be too much for some.

Satan just asked Mary to stop by for a quick warm-up(he he) as hell just froze over.

Good God what's next????!!!!

BSD (it's not dead, after all!)

This shows a huge amount of ignorance. BSD is alive and fine, in several forms:
- FreeBSD
- NetBSD
- OpenBSD
- DragonFly BSD
These are probably the most important. Take a look at Freebsd Derivates. You'll see there are many commercial products derived from Freebsd too.

Also, there are initiatives of porting different Linux distros on top of the BSD kernel:
- Gentoo/*BSD
- Debian GNU/kFreeBSD
- Debian GNU/NetBSD (abandoned in 2002 it seems)

BSD was, is and will be alive for a long time.

FreeBSD doesn't do 'repositories', so to speak. They do ports, and then FreeBSD.

/usr/ports is just the buildtree for packages. Its shared between FreeBSD, NetBSD and OpenBSD and some other OSes.

pkg_add -r will install binaries for any port that builds properly for your version of the OS (assuming you're not using an old version that is no longer maintained).

As for the guidelines, well you should probably follow the links at the bottom of this page: http://www.freebsd.org/internal/policies.html

I think you'll find its almost entirely unlike what you say it is.

I also understand there are most certainly Debian repositories that contain non-libre software.

You might want to do a bit of reading about this stuff you are talking about.

FreeBSD doesn't do 'repositories', so to speak. They do ports, and then FreeBSD. They're conveniently independent (I suspect so that the FreeBSD project can claim superior security to everything else).

They're technically handled extremely differently; the code for the main OS is directly edited and maintained in a single consistent source tree and build system, while the ports are just extracted, patched, and built using their own individual build system.

There's a slightly blurry line about some things in contrib vs ports; occasionally, things are moved to ports or moved from ports to contrib, and the code for contrib usually isn't edited that much, to facilitate updates. It could be argued that that code should go in ports and there should be "mandatory ports" rather than contrib.

I used to be an active committer in FreeBSD; I've never seen anybody say anything about using this to argue better security. All discussion about it has been about what's convenient in terms of what we do with code (and it being convenient to have a defined base system so we can deliver a consistent system).

Even then, ports don't really have 'guidelines'. "I maintain this port and I'll update it as I please, consequences be damned" seems to be the guiding message, though.

There's a Porter's handbook. There's a port manager team to clamp down on it, and before there was a port manager team there definitively Satoshi Asami (or before that Jordan Hubbard) to block bad changes.

Is there a particular port or set of ports you've had problems with? In general, I've found that a polite mail to the maintainer will take care of things - but maintainers clearly aren't perfect, and sometimes it's been necessary to bring technical points up for discussion in the mailing lists (or, if it's heated before it gets to that point, take it up in private with the port management team.)

Eivind.

From the ipsec(4) manpage for Mac OS X 10.6, history section:

The implementation described herein appeared in WIDE/KAME IPv6/IPsec stack.

The KAME stack is the same stack used in NetBSD and FreeBSD.

Even though NeXTSTEP was forked earlier from the BSD codebase than the other BSD flavors there has still been considerable sharing between it, Mac OS X, and the other BSD flavors. OpenBSD is one exception to this since it tends to be a more closed ecosystem than the other BSD variants.

Because if you look carefully you'll notice that the four colored squares leave a greek cross in the middle.
And I'm not pulling the abvious FreeBSD in.

I wonder if ZFS will continue to be released to be used in FreeBSD.

Yes -- http://lists.freebsd.org/pipermail/freebsd-fs/2010-August/009197.html

So, let's all download FreeBSD (or any of these), I'm sure we'll be doing the authorities a great favor.

Daemons

No, no! He used the backup method called "Do Nothing".
See chapter 18.12.6 from here

But: What colour should the shed be???

And it's seeded well.

http://tinyurl.com/linuxbad. Reason for http://openbsd.org/ and http://freebsd.org./

Mac OS X: http://www.apple.com/downloads/macosx/games/role_strategy/foxminesweeper.html
Debian GNU/Linux: http://packages.debian.org/lenny/xbomb
FreeBSD: ftp://ftp.internat.freebsd.org/pub/FreeBSD/ports/i386/packages-8-current/games/xdemineur-2.1.1_1.tbz
Windows Vista: http://windows.microsoft.com/en-US/windows-vista/Learn-about-Windows-games

HA!

Take out Linux or take out GNU -- what you're left with is not an operating system any more. Take out anything else, then what's left still defines an operating system. That's why it's most efficient to call it GNU/Linux, but, of course, you are free to call it whatever you want, as long as you're not being misleading. I don't think anyone has an objection to calling a system KDE/Xorg/GNU/Linux, if you have the stamina to type it then go ahead!

The difference between using two-part names and multi-part names is that every operating system needs to have a basic userland and a kernel, but all the other programs are purely optional. If a system doesn't have GNU, it will have something else in its place. Your home router will probably not be running GNU/Linux, but Busybox/Linux instead. Your phone will be running Android/Linux. The key point is that these are incompatible, different operating systems, and you will likely have problems if you want to run a program with particular system dependencies on all of these. Because this is free software, there are many possibilities of such pairs, such as FreeBSD/kFreeBSD, GNU/kFreeBSD, GNU/kSolaris, etc.

When you say "I run Linux", strictly speaking, you refer to all possible operating systems that run on the Linux kernel. Since you need a userland to make a full operating system, such a description is incomplete. And when someone asks you what system you're running, they're not asking about just the kernel, they want to know the whole thing. So *at minimum* you should say what kernel and userland it has. If you say "I run Linux" to mean "I run GNU with Linux", then, strictly speaking, you are being misleading, because Linux does not necessarily imply GNU, same as GNU does not imply Linux.

So, you see, it's not just about giving credit, it's also because Linux is not an operating system, and it's because us geeks want to communicate precisely.

I suspect that part of the problem with Windows is that MS doesn't enforce the kind of cleanliness and neatness of code that some of the competitors do. Granted all OSes have some cruft and scariness in places, but most of the ones that are known for stability have long since shaped up and enforced something along the lines of style I'm sure that most modern projects of any size and reliability have something similar to work from. The more uniform the style is and the better the adherence the easier it is to find bugs that might be hiding security problems.

The other thing is that it's very difficult to get a look at the Windows source code legitimately, without being paid to work on it. Whereas with Linux or *BSD if you have a bug you've got the option of fixing it yourself or if you don't have the time or expertise you can usually find somebody who's willing to do so for a price. Frequently is thrilled to get to fix the problem on somebody else's dime.

Beyond that, backwards compatibility has to be careful considered and engineered otherwise you can easily end up in the situation where a vulnerability exists due to legacy code or the model itself is prone to exploitation.

Let me quote from the FreeBSD manual: http://www.freebsd.org/doc/handbook/boot-blocks.html

"Conceptually the first and second stages are part of the same program, on the same area of the disk. Because of space constraints they have been split into two, but you would always install them together. They are copied from the combined file /boot/boot by the installer or bsdlabel (see below).

They are located outside file systems, in the first track of the boot slice, starting with the first sector. This is where boot0, or any other boot manager, expects to find a program to run which will continue the boot process. The number of sectors used is easily determined from the size of /boot/boot."

The FreeBSD booting mechanism on the x86 does the SAME thing as GRUB. It can also use LILO for booting.

Your comment is not warranted. A Windows dual-boot with such a "bad program" could possibly damage a FreeBSD boot loader as well.

Well that's a load of bull if I ever heard one. Your entire post needs a giant [citation needed].

I use FreeBSD ZFS at $work and it works great. Combine it with HAST and CARP and you have a pair of redundant application servers with instant data replication and failover. All on FreeBSD, all very, very usable.

The FreeBSD ports system is awesome IMO, I never understood why some people gets all worked up over it. There are advantages and disadvantages in all package management systems, but I've found FreeBSD ports (with Portmaster being my tool of preference) to be flexible and nice to work with. Maybe it's been a while since you used FreeBSD ? If so, try out Portmaster :)

Saying that the FreeBSD project as a whole is in decline is pretty strange considering the size of their "Quarterly Status Report" filled with all kinds of exciting things. See for example the latest one: http://www.freebsd.org/news/status/ - or the one before that.

Saying that the USB or AHCI subsystems hasn't worked for years is also, obviously, very wrong. For example, the AHCI driver works very well for my SATA drives, and it includes SATA hotplug support and whatnot. The USB system has worked fine for everything I've used it for - from 4.x through to 8.1 today.

The only thing I'll grant you is that deduplication and a couple of other nice ZFS features are not yet in FreeBSD, and they would indeed be nice to have. But saying ZFS on FreeBSD is useless because of this fact is a bit harsh don't you think ?

You may be happy with Solaris but please stop spreading FUD, check your facts, and have a nice day.

ps. sorry for AC, forgot my login somewhere, been a while since I last posted.

From the handbook it suggests you drop down to single user mode (see #4): http://www.freebsd.org/doc/handbook/makeworld.html pfft, I usually skip that step and just shutdown anything important (dbs, mail, www, etc); but could get all wacky. Anyways, now days you can get IPMI on a lot of servers/motherboards. I like SuperMico, nice solid boards and IPMI that works "good enough". HP also has nice ilo cards (but more costly to get an HP server).

I assume he means "firewalls" by "FW". Seriously, you can't even bother to spell out "firewall" in a presentation?

...depends if he's a FreeBSD fan or not. :)

FreeBSD 8.1 is now only a year behind OpenSolaris in terms of its ZFS support; it's compatible with pools created with OpenSolaris 2009.06. If you assume that freely available innovation from Oracle's side is going to stagnate, it shouldn't take long until they've caught up with the latest of the open-source ZFS releases. At that point it will be FreeBSD vs. paid Oracle Solaris as the presumed only way to get advancements. Since I don't actually care about the non-free crap that Oracle peddles, I expect a near future point where the FreeBSD version could be the only interesting ZFS release to me.

Except that not all of the FreeBSD world prefers a compile-the-world approach, which is why there's a freebsd-update utility now in base for binary updates of the base OS.

How do you count premiere? Features? How about code quality? Compare these two for a trivial example:

FreeBSD cp.c: view
Coreutils cp.c: view

The latter is embarrassing and the person should be ashamed to call himself a programmer. And this is, by far, one of the better-written GNU parts. I have long felt that the FreeBSD tools are better suited to being paired with Linux than the GNU tools are, as they both (FreeBSD & Linux) maintain similar coding standards, and the FreeBSD tools are better documented and undeniably more secure & bug-free.

They just released FreeBSD 8.1with G5 support and with a bootloader that can boot from a ZFS file system. So make it into a power hungry ZFS file server.

Real operating systems have real jails.

zfs works in 7.X, just 7.0 -> 7.2 it wasn't considered production ready.

zfs + iscsi integration doesn't work in 8 and didn't work in 7 either because iscsi target support doesn't exist in the base system like OpenSolaris. Running iscsi targets is possible and stable with some implementations from the ports tree however(I assume you tried /usr/ports/net/iscsi-target since you are so down on FreeBSD iscsi target support, use /usr/ports/net/istgt). You don't get the ease of use as in Opensolaris, but a sparse file on a compressed ZFS filesystem makes a fine iscsi target. A normal file on a non-compressed filesystem can also be easily encrypted with GELI if you so desire. GELI would work on sparse file on a compressed FS too, but there's no point in that method.

http://people.freebsd.org/~rse/iscsi/iscsi.txt

Or they both legitimately got it from BSD, or linux got it from the standard.

HISTORY

The ELF header files made their appearance in FreeBSD 2.2.6. ELF in
itself first appeared in AT&T System V UNIX. The ELF format is an
adopted standard.

We know from the AT&T settlement that there's a lot of BSD in AT&T Unix, and that even some of the non-BSD AT&T stuff simply isn't protectable by copyright.

Seriously though, MAC OS X has its roots with the BSD's of the world (Mach kernel http://developer.apple.com/macosx/architecture/) and one need only visit any of the number of SSD inquiries from the FreeBSD community to get an idea of how Mac OS X would respond at a kernel level: http://forums.freebsd.org/showthread.php?t=1169 for example.

Granted, there are some filesystem differences between the other BSD flavors and MAC OS X, but the architectures are very similar and one can draw the same conclusion as this parent review did:

MAC OS X can run quite well on SSD drives, it just hasn't been sanctioned by Big Brother Jobs yet. Remember too, Apple offered 128GB SSD's in their Mac Book Pro's as far back as 2008, so it's nothing new.

If you don't mind voiding your Apple warranty, I'd say go for it!

FreeBSD is currently moving (slowly) towards Clang/LLVM with their ClangBSD subproject. They've just imported clang into the base system of FreeBSD (-CURRENT), and clangbsd is on its way.

Note that FreeBSD has just imported clang to FreeBSD head.

That nerdy little BSD kid just keeps on going, and going, and going, and going. I think maybe she's getting some of the Energizer Bunny.

Okey, I'm calling your bullshit.

Look at http://wiki.freebsd.org/SummerOfCode2009Projects and you will find a decent percentage of female names.

Let me start some speculation here.

Apple took a bold move when they dumped their outdated and technologically inferior Mac OS and rather than starting from scratch they used open source (FreeBSD) at the heart of their new OS.

Another bold move when the Apple browser Safari was again based on an open source core (KHTML).

So I predict this new iPhone version will make the same bold and logical move and be based on ..... Android 2.2!

Go Apple! :) /me ducks for cover

The problem is Microsoft has consistently failed us all. This catastrophe must be squarely laid at their feet. Windows powers over 90 percent of the world's desktops therefore the solution must start there. On the desktop. MS has had decades to rectify this untenable state of affairs. They have so far proven themselves grossly incompetent. The only solution and the one I discovered is to quite simply switch to an alternative. Any alternative. Otherwise, Windows monoculture will be the downfall of us all.

I'll enthusiastically second this. I have been running FreeNAS solidly for 2 months now. I have configured 5 x 1.5TB drives in RAID-Z2 configuration. That gives me 4.01Tb of storage, allowing for two simultaneous drive failures without data loss. My final build uses an ECS 945GCD-M(1.0) Atom 330 Micro ATX board with an Intel 1Gb PCI card; I had a lot of trouble getting the network working with a pure intel board (which did NOT have an Intel network controller on it, GRR). While a highly regarded SATA controller worked with the original Intel MoBo, it didn't with my current one, but a cheaper one does. I'm using a 400MW 80 Plus PSU, and Kill-a-Watt says I pull about 75W total while running, motherboard reports that it's running at about 30C. I have 4Gb of RAM, of which maybe 30% gets used, and CPU seems to be around 10-20% generally. Discounting the hardware that didn't work, it comes to about $1000. I also invested in $130 of UPS (which brought it through a 5 second mini-blackout), and a Gb switch. I get sustained 30-40MB/sec, both to/from Vista and Ubuntu systems over a wired Cat-5e home network; far more than needed for use as a HTPC.

ZFS is stunning. I was sure I was missing something when I set it up but - it's - just - that - easy. It's like encountering a bullet train after spending two decades using hand-drawn sledges to get around. Copy-on-write, self-healing, snapshots as easy as sneezing. RAID configuration that would fit into a Twitter message. Hot spares, automatic re-silvering when adding or removing disks. It's about 4 tech levels above what I'm conversant in, which does make me nervous; I have not tried to recover from a drive failure yet. I'd also like to move the OS off the ancient 30Gb boot drive and onto a flash disk, but want to make sure my tertiary backups (mostly external USB drives) are *really* up-to-date. FreeNAS allows export and import of configuration XML files, so hopefully that will be relatively easy.

I did learn that while you don't need to explicitly format the drives, if you have used them for a prior ZFS system you should wipe them before reusing them. I lost three weekends of my life to trying to configure OpenSolaris (the time would have been better spent getting femur piercings). In the process, I briefly had a four drive ZFS zpool. When I tried to build a pool with those plus one more in FreeNAS, Bad Things Happened. I had to use DBAN to clear off the drives, after which everything went fine. I earlier tried FreeBSD, but it refused to boot from a USB CD with the EliteGroup motherboard. Ah, and I did need to modify vm.kmem_size and vm.kmem_size_max, I think both to 4G. That can be done from the FreeNAS config page (99% of FreeNAS management is done from a webpage, similar to router configuration).

http://www.freebsd.org/where.html

ZFS is buggy and unreliable:

http://www.mouldy.org/zfs-the-final-straw
http://opensolaris.org/jive/thread.jspa?messageID=473421&tstart=0
http://mail.opensolaris.org/pipermail/zfs-discuss/2008-January/016042.html
http://forums.freebsd.org/showthread.php?t=7682
http://markmail.org/message/4prch5ydto6nxquf

I'm interested in Tux3 and (particularly) Hammer though.

http://www.ntecs.de/blog/articles/2008/01/17/zfs-vs-hammerfs/