The Future of OpenSolaris

jjrff writes "Phoronix has a little piece about the future (or lack thereof) of OpenSolaris. It appears based on the current support lifecycle, OpenSolaris may be going away. There is a fun thread (read: mild flameage) on a ZFS list about it."

Comments

FUD

[Anonymous Coward]

Nothing about this says OpenSolaris is going away. Just support for older versions

Re:FUD

[Michael Kristopeit]

to some decision makers, that is the same thing...

Re:FUD

[Jurily]

s/decision makers/sensationalists/

Re:FUD

[Anonymous Coward]

So you are saying you are able to get support for the current OpenSolaris version? It seems to me that is impossible.

Re:FUD

[Anonymous Coward]

OpenSolaris? It's dead, Jim.

Deal with it--suck it up and be a man. Then get on with your life. After all, it's only an operating system for God's sake. And not a very good one at that.

Re:FUD

[Spit]

If long-term usage was that important, you'd be using Solaris in the first place.

Re:FUD

[smash]

On the contrary... if you want zfs, and stability under load, there's nothing wrong with it.

Re:FUD

[mysidia]

Nonsenese.... if you install (a Supportable) version of OpenSolaris and buy support. Make sure your contract includes support for your version for a long enough duration.

You know, Microsoft doesn't support ancient EOL OSes like Windows NT or Windows XP, either.

Re:FUD

[gparent]

You know, Microsoft doesn't support ancient EOL OSes like Windows NT or Windows XP, either.

XP is supported until 2014 according to my understanding. Can you clarify?

Re:FUD

[JDmetro]

Open solaris support is stupid it currently in backwards compatible to ver 2. Now Imagine if Windows was backwards compatible to Win 3.1. Sure my old games would like it but maintaining backwards compatability is stupid and I don't play my old games anymore. Look to the future and bigger and better and pull your head out of the past. And if you can't get you head out of the past please dedicate your life to building a time machine.

Re:FUD

[WaywardGeek]

Oracle just fired many of the best and brightest programmers at Sun, because they were the most highly paid. So, if long-term is important to you, I suggest jumping ship. As one example, they fired the accessibility guru, Willie Walker. As a result, SunOS will no longer be accessible as it use to be, causing it trouble in winning government contracts. I say good riddance... SunOS was cool, but with Oracle in charge, it's time to move on to greener pastures.

Re:FUD

[mysidia]

Mainstream support for XP ended in 2009. No new features. No bugfixes. No patches except for a very small number of items deemed by MS to be critical security issues as in wormable remote code execution (almost all newly found holes in XP security-related or otherwise will never be patched).

Only pay-per-incident (very expensive) assistance and paid critical fixes.

Mainstream Support for Business and Developer products will be provided for 5 years or for 2 years after the successor product (N+1) is released, whichever is longer.

Re:FUD

[gparent]

They still release security fixes regularly for XP. New features don't really matter at this point, you can use a non-obsolete OS for that.

Re:FUD

[Anonymous Coward]

considering that every viable OS has EOS (end of support) timelines set, this is NOT news.

Re:FUD

[Digital Pizza]

Do you have some citations for this? Not being a smart-ass, I'm genuinely interested in reading more about this if it's true.

Re:FUD

[aztracker1]

They're ripping out all the accessibility already in place? talk about FUD. In any case I would agree that firing your high profile developers is a bad move.

Re:FUD

[afabbro]

Oracle just fired many of the best and brightest programmers at Sun, because they were the most highly paid.

Dear Hewlett-Packard,

Thanks for posting this unsubstantiated rumor here on Slashdot. Oh, while you're here, we'd all love to see some servers that run on something other than Itanium. Thanks.

Re:FUD

[fm6]

So, if long-term is important to you, I suggest jumping ship. As one example, they fired the accessibility guru, Willie Walker. As a result, SunOS will no longer be accessible as it use to be

Walker didn't work on SunOS accessibility. He worked on GNOME accessibility. Sun contributed to GNOME because their version of it (Java Desktop) was Sun's final stab at making Solaris a serious alternative to Windows for desktop users.

Sun lost the desktop wars years ago, but never admitted defeat. It challenged Windows with SunView, NeWS, OpenWindows, CDE, JavaOS, and finally Java Desktop, and failed each time. At the end, even Sun people had mostly given up on desktop Solaris.

Once Sun's cronyistic and politicized upper management got their golden parachutes, a lot of windmill-tilting projects were toast. Java Desktop (and Sun's participation in GNOME, which is really the same thing) was one such project. That's what this was about — it had nothing to do with how much programmers were paid.

Solaris's future, if it has one, is as a server OS. Expect to see the money that was being wasted developing a desktop nobody wanted poured into thing Solaris badly needs, like better drivers.

Re:FUD

[WaywardGeek]

I've got plenty as regards Willie Walker:

http://blog.grain-of-salt.com/index.php?itemid=394

This was already discussed in length on slashdot. As for the firing of lot's of the top-paid programmers, I can point to multiple e-mails on the accessibility lists that claim it's true, but that's not much proof. In theory, Willie was fired along with a large number of the most highly paid programmers, across Sun.

Re:FUD

[mysidia]

Extended support is not standard support, and not available under the same terms as the original support agreement. You can always attempt to negotiate with a software vendor to provide you support for an EOL product, if you're willing to pay enough extra they will do it.

Microsoft releases SOME security fixes for wormable/remote code-execution issues for free, but not other bugfixes. And they are releasing these as a good internet citizen, not to provide general support XP users.

Examples of security bugs they won't patch are remote DoS MS doesn't think will run arbitrary code: Evidence MS09-048, Frequently Asked Questions (FAQ) Related to this update:

If Windows XP is listed as an affected product, why is Microsoft not issuing an update for it? By default, Windows XP Service Pack 2, Windows XP Service Pack 3, and Windows XP Professional x64 Edition Service Pack 2 do not have a listening service configured in the client firewall and are therefore not affected by this vulnerability.

Another good example of what they won't do is release an update to fix 'daylight savings time' again if the US government changes the rules again this year or next year.

Customers who prefer to have a hotfix may be able to purchase DST updates for affected Microsoft products in Extended Support via an Extended Hotfix Support (EHS) contract. A single fee of $4,000 covers all available DST hotfixes for the current calendar year.

In other words, you have to sign up with the extended support and pay the additional fees per system for the extended support, AND in addition to that pay $4000 to get the DST fix.

At this point you are not getting standard, general, or common support for your product. You are paying/contracting Microsoft to do something for you on an individual basis.

If you pay them a big enough amount of $$ as in millions, or billions, they'd probably be willing to make DOS 5.0 or Windows 3.1 patches for you.

But that's not what we're talking about... Sun's web pages indicate the general support is ending, and Windows XP general support is over and done with.

Re:FUD

[WaywardGeek]

They're ripping out all the accessibility already in place?

No, but gnome is switching from bonobo to dbus, from gecko to webkit, and from Gnome Speech Services to speech-dispatcher. There's no one left who cares enough to make all this work in Gnome 3.0 for SunOS. Want to make any bets as to whether screen readers work when SunOS switches to Gnome 3.0?

Re:FUD

[WaywardGeek]

It's nice to hear an informed opinion on slashdot. I'm also glad to hear the rumors about Willie being fired over pay aren't true. I guess SunOS accessibility, and it's entire bid for the desktop is toast, but it might make a decent non-accessible server OS. Linux servers are accessible through the speakup kernel modules, but as a non-main-stream server OS, SunOS can't expect to have such features, can it? Certainly not if they fire the person many believe to be the world's foremost FOSS accessibility expert.

Let's just say I will fight vigorously against any deployment of non-accessible software in our enterprise, and I see no future for SunOS in companies like mine.

Re:FUD

[Digital Pizza]

I haven't been following the discussions about this; thanks for the link.

Re:FUD

[thetoadwarrior]

I personally quit using windows ever since that jerk face Ballmer and Gates quit supporting Windows 3.11.

Re:FUD

[amorsen]

All our servers are command-line only. Everything is done by ssh or the built-in serial-to-Ethernet management port. When even the BIOS is command line, why exactly do you need your kernel to talk to you? Why can't the notebook you're using to connect to the server talk to you?

I don't really see a market for server-only operating systems though. It's too convenient to run the same thing on the server and the desktop.

Re:FUD

[jgrahn]

Sun lost the desktop wars years ago, but never admitted defeat. It challenged Windows with SunView, NeWS, OpenWindows, CDE, JavaOS, and finally Java Desktop, and failed each time.

Surely SunView wasn't a challenge against Microsoft Windows. SunView was old stuff when I used it in 1990, and PCs at that time generally ran MS-DOS and Norton Commander. I suspect NeWS and OpenWindows ignored the toy computers too, but am too lazy to check the timeline.

I *do* feel that the Solaris desktops got worse and worse over time, though. Plain X11 with twm as a window manager is OK. OpenWindows was odd but tolerable. Motif sucked. CDE looks and feels like a toy, with that silly control panel at the bottom center.

Re:FUD

[fm6]

It's nice to hear an informed opinion on slashdot. I'm also glad to hear the rumors about Willie being fired over pay aren't true.

It doesn't even rate as a rumor. More like a collective ill-informed opinion.

I guess SunOS accessibility, and it's entire bid for the desktop is toast, but it might make a decent non-accessible server OS

What exactly about a server makes it "accessible"? The user never interacts with it directly — that for the client software.

Re:FUD

[WaywardGeek]

True, SunOS would be accessible if I log in through an ssh shell in Linux. Over on the speakup list, there are many blind system-admins who own hardware voice synthesizers. They attach these to the server's serial port, which allows them to hear bootup messages. For servers that don't get all the way to bringing up internet services, this is pretty useful.

I agree with you, though. I prefer to run Debian servers now because I use Ubuntu on my desktop. It's nice when they have so much in common. I'm just PO-ed that the Oracle/Sun merger is causing so much turmoil in accessibility land. SunOS was probably going to die anyway, and as that happens, clearly they are going to layoff people. I hope Oracle doesn't bring a big axe to MySQL, Java, and OpenOffice. It's also not clear how these projects make money.

Re:FUD

[fm6]

I suppose you're right. SunView and NeWS obviously predate the battle with Windows. They were part of the desktop wars though, during the pre-Microsoft period.

Re:FUD

[fm6]

I don't really see a market for server-only operating systems though. It's too convenient to run the same thing on the server and the desktop.

Why? I administer Linux boxes from Windows machines all the time. I don't see the difficulty.

Re:FUD

[WaywardGeek]

If Oracle were a bit more informative about what they're doing, we wouldn't be forming ill-informed collective opinions. This stuff is pretty important to a lot of people, and the complete silence from Oracle is a problem.

Blind sys-admins are able to slap speech synthesizers on the serial ports of Linux servers, and hear boot messages. If a server fully boots, is on the internet, running sshd, then it's not a problem. However, to hear the blind sys-admins describe it, if the machine boots that far, the problem is probably already solved.

Re:FUD

[amorsen]

They attach these to the server's serial port, which allows them to hear bootup messages. For servers that don't get all the way to bringing up internet services, this is pretty useful.

It's a bit silly to buy servers without management features these days; you can get even low end servers with serial-over-IP + remote power control. No need to mess with serial ports. Besides, this article is about Solaris. Sun hardware has ILOM.

It does suck if the Oracle/Sun merger means that the accessibility efforts in Gnome slow down.

Re:FUD

[amorsen]

The difficulty is that you don't get as used to the Unix tools if you aren't in a Unix environment. (And the same for Windows). Also, since you can run almost every type of service on any platform you want these days, there isn't much incentive to pick something unfamiliar for servers.

There might be a case for using something non-Windows even if you prefer Windows on the desktop, just because Windows servers are such a pain to administer. However, Windows is a lot better for servers than it used to be, and in 5 or 10 years it's probably quite usable. The stability problems are gone already, which certainly helps.

Re:FUD

[fm6]

I suppose I have an advantage in that I started out doing everything in Unix, then was forced to switch to DOS and Windows, then went back and forth many times over the years.

Still, there are a lot of Linux/Unix servers out there. Sun, IBM, HP, even Dell sell them by the ton. And they're mostly administered by people whose main desktop is Windows. I've seen this first hand, at hardware manufacturers (including Sun!) and big Linux-centric data centers. There's a concept impedance there that makes going back and forth a pain (every once in a while I type "rm" when I mean "del" and vice-versa). But it's not insurmountable.

Re:FUD

[gparent]

Pretty much repeated what I knew from the start Windows wise, but from what I understood even if you wanted support from Sun you wouldn't even be able offer to pay for it.

Re:FUD

[fm6]

What, Oracle is suppose to document their reasons for every engineer they lay off? Get real.

Blind sys-admins are able to slap speech synthesizers on the serial ports of Linux servers, and hear boot messages. If a server fully boots, is on the internet, running sshd, then it's not a problem. However, to hear the blind sys-admins describe it, if the machine boots that far, the problem is probably already solved.

You seem to be saying that blind sysadmins don't need the OS to provide accessibility support. If so, why is this even an issue?

Actually, Oracle does need to provide accessibility support for all its products, or else forget about selling them in the U.S. It's just that the accessibility supported needed for the base OS is pretty low. It's the applications that run on the OS that need serious accessibility features, not the OS itself.

Re:FUD

[ishobo]

Accessibility guru? The best and the brightest for Solaris represent the kernel folks. How much more accessible can a command line be?

Re:FUD

[ishobo]

Stop it. They are called rumor and lies. Owned up to it, you are responsible for spreading them. On your comment about about not seeing a need for a sevrer OS, please find some other line of work. I am sick and tired of incompetent people such as yoruself working in this field. It didn't used to be this way. Fom the 90s onward, aeery Tom, Dick, and Harry thicks they are an expert.

Just look at your statement:

As a result, SunOS will no longer be accessible as it use to be, causing it trouble in winning government contracts. I say good riddance... SunOS was cool, but with Oracle in charge, it's time to move on to greener pastures.

You have no clue what you are talking about. Are you 12? Does your daddy know you are using his account?

Re:FUD

[WaywardGeek]

They're ripping out all the accessibility already in place?

Many attempts have been made to contact Oracle over Orca and accessibility. It was even slashdotted. This effects a lot of people, not just the Sun employee.

The speakup driver in the Linux kernel is required to use the speech-synthesis devices. SunOS doesn't work with them.

Re:FUD

[h4rr4r]

1. Why?
2. Why do you have this job?
3. Why in the hell would you do that?

Re:FUD

[WaywardGeek]

This is old ground, but I'll go over it again. The blind mostly are willing to do things for themselves in open-source land. If you were a good hacker who went blind, what would you do? Add some low-level code to the Linux kernel to drive a speech-synthesizer or Braille device, and you're off coding. That project is called speakup, and it's great for the blind. What if you were an emacs hacker? Would you write an entire desktop environment in emacs based on speech? That project is called emacspeak, and it's considered by many to be the most productive programing environment for the blind who can hear well.

What if you want to use Firefox, OpenOffice, or any of that cool new-fangled software? You'd have to edit 100 packages to begin to make that work, and you'd need cooperation from the big distros. It's too big a job for the blind, especially given how much of has to be done without working speech or Braille. Willie Walker is both the Orca and Gnome accessibility lead, and for good reason. He's the only guy I know with the clout with all the major distros to get anything done. Without him, it's going to be very difficult just keeping accessibility strong in Gnome/Linux, much less SunOS.

Re:FUD

[Anonymous Coward]

s/you/OpenSolaris Zealot/

That was easy! I salute both you and the sinking ship that you refuse to acknowledge is dead in the water. Well played.

Re:FUD

[fm6]

1. Can you be more specific?
2. This is not the place for my life story.
3. Because Linux works best for me as a server OS, and Windows works best for me as a desktop OS.

If you're interested, I'll explain answer 3. Note that "interest" is not the same as "needs to be saved from that evil demon Microsoft" I have no interest in re-fighting old flame wars.

Re:FUD

[fm6]

The post you quoted wasn't one of mine. I'll assume you meant to quote the post you replied to.

Many attempts? All you're citing is two angry posts on a blog I've never heard of.

The speakup driver in the Linux kernel is required to use the speech-synthesis devices. SunOS doesn't work with them.

And there's no reason it should. Not if you're only using it as a server. You were talking about watching the BIOS and OS startup messages over a serial port. If you're doing that, what good does speech synthesis support in the server OS do you? The place you need speech synthesis is in the terminal software you're using to monitor the serial port.

Re:FUD

[bingoUV]

I am ignorant in this field, and I have a few questions that have no been answered in this thread:

1. Why does speakup have to run in kernel mode? Is it because it has to run an external device, attached to the computer through, say, PCI/PCI-Express/PCMCIA etc.?

2. From another of your posts in this thread, I understand that Gnome is changing some of its programs and someone is needed to make the new programs accessible. But couldn't the blind use CDE? I don't like CDE because it is ugly, but that is one criterion blind people will never have. I am not sure how good CDE is for blind.

3. So Willie knows a lot about Accessibility & Gnome; and has clout with Linux distros. Gnome is the darling of corporate linux. Any linux company will employ Willie. Gnome being GPL, work Willie does for any distro will be available to OpenSolaris (unless CDDL prevents this somehow, not about licensing issues of GPL stuff shipped with OpenSolaris).

Rest is solaris kernel stuff such that the newly available accessibility features from Gnome work well with (Open)Solaris. Solaris kernel guys should be able to do this with some effort.

thanks

Re:FUD

[bingoUV]

But you can upgrade OpenSolaris for free. To upgrade XP to Vista/7, you need to pay money.

Re:FUD

[WaywardGeek]

You have been reading carefully... Speakup integration insures the blind can hear anything printed on the console. It's almost as good as seeing the boot screen. For example, if a blind sys-admin has a server with a failing hard-disk, he needs to know what messages get printed, even if boot fails. Further, speakup has become the tool of choice for blind sys-admins, so even if they are just trying to read a console, they'll miss speakup if it's not there.

I'm not familiar with CDE, which probably means the blind aren't either. It's probably not very accessible, unless it's based on gnome. KDE, for example, is not accessible at all.

I agree that any large Linux company should hire Willie. It's a mystery to me why Canonical doesn't snap him up, or RedHat, or even Novell. A lot of us are aghast at the lack of interest in accessibility from the big distros. It's like they keep waiting to see if someone else will solve the problem. Of any distro, SunOS has the best excuse for laying off accessibility guys, but the other distros are growing, and should be hiring.

The Solaris kernel guys are certainly an excellent team, but unless they decide to start hacking python scripts in user space, and 30 ancient C programs with no documentation and support, accessiblity in Gnome for SunOS will only progress due to the generosity of programers in the community. Most of us are in Linux land, and with Willie jumping ship, I would not count on SunOS accessibility in the future.

Re:FUD

[gparent]

So because the OS is free they won't accept your cash? Right.

Re:FUD

[bingoUV]

I know some people in Oracle middle management, and possibly they can arrange to accept cash from you. Not sure why you are so eager to pay, though.

Re:FUD

[gparent]

Because if you read the news, there won't be any support otherwise.

Re:FUD

[bingoUV]

If you read the news carefully, there won't be any support even if you pay. Why, then, the eagerness to pay?

Re:FUD

[gparent]

There's no eagerness, I don't know why you're bringing this up. I just said the fact that you pay or not for support should be irrelevant to the fact that you pay or not for the OS (because you brought up Vista for some reason). In either cases, I'd expect better support if I pay than if I don't.

Bugger.

[Anonymous Coward]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Re:Bugger.

[goldaryn]

"Can somebody show me something good to come from the Oracle-Sun deal? Anything?"

Cutlery related anecdotes for the next 10 billion years

Re:Bugger.

[Anonymous Coward]

"Synergy"?

Re:Bugger.

[Random Person 1372]

MySQL dying? SCNR.

Re:Bugger.

[Michael Kristopeit]

postgres?

Re:Bugger.

[segedunum]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Errrrrr, survival and preventing Sun from going bust, just off the top of my head?

Re:Bugger.

[drinkypoo]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Errrrrr, survival and preventing Sun from going bust, just off the top of my head?

Is that really good? I just met someone who now works for Oracle; they worked for a company acquired by Sun prior to the merger. Sun fired all their best SEs because they made good salaries, while there are people all over Sun (or at least, were) making big bucks for doing nothing. UltraSPARC has its uses, but mostly it is an also-ran. Solaris' claim to fame is ZFS. Under Oracle, Solaris is doomed to either fail (as it was heading towards anyway, due to Linux's increasing dominance) or to become an Oracle RDBMS engine, which is much the same thing.

Re:Bugger.

[Trepidity]

Well, Oracle offered about 3x the prevailing Sun stock price, so the Sun shareholders have done well. At least, well in relative terms--- some probably still lost money, but there was really not much else on the horizon that was looking likely to triple Sun's stock price. Before Oracle came along, the just-over-$3.00 stock was almost mocking its owners with its stock ticker of JAVA, an anachronism from the days that Sun management thought Java would somehow make them rich.

Coincidentally, for public companies, if you make a really good offer to stockholders (something >2x the current stock price usually qualifies), it's usually an offer the buyout target will find hard to refuse. That's the tradeoff you make when you IPO a company and put its ownership in the hands of the stockowning public.

Re:Bugger.

[calzakk]

What would happen to Java if Sun went bust?

Re:Bugger.

[Dog-Cow]

Unfortunately, nothing. Or IBM. It's hard to tell which would be worse.

Re:Bugger.

[durdur]

Sun was going to get acquired. The only alternative to Oracle was probably a deal with IBM. You can speculate if IBM would have been a better owner, but IMHO they'd have had many of the same corporate priorities: making money and cutting losses on things they couldn't make money on. If Sun had focused on these things earlier, rather than doing crazy stuff like spending $1B on mySQL, they might have had a chance surviving on their own.

Re:Bugger.

[Ltap]

IBM might have been better if it had kept MySQL going. It's possible that Oracle will try to sneakily phase out MySQL.

Re:Bugger.

[H0p313ss]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Errrrrr, survival and preventing Sun from going bust, just off the top of my head?

Is that really good? I just met someone who now works for Oracle; they worked for a company acquired by Sun prior to the merger. Sun fired all their best SEs because they made good salaries, while there are people all over Sun (or at least, were) making big bucks for doing nothing. UltraSPARC has its uses, but mostly it is an also-ran. Solaris' claim to fame is ZFS. Under Oracle, Solaris is doomed to either fail (as it was heading towards anyway, due to Linux's increasing dominance) or to become an Oracle RDBMS engine, which is much the same thing.

And that's been true for a decade. I'm surprised they lasted this long.

Re:Bugger.

[Anonymous Coward]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Look at the performance numbers of the Oracle Exadata V2.

Larry Ellison feels that the integration of computer components is happening at the wrong layer. He wants to recreate what IBM did under T. J. Watson Jr. and sell systems that Just Work(tm). He thinks selling people a bunch of random shit and having them treat it as a DIY kit (or hiring consultants to put the kit together) is dumb.

You call up Oracle/Sun, tell them you need something that does X with performance parameter Y, and they'll drop a box in your loading dock. If it doesn't perform you call one number and they fix it for you (instead of finger pointing).

Re:Bugger.

[Anonymous Coward]

Most enterprise users of Solaris are also running Oracle as their RDBMS. If Sun were to failscade, many of those customers would migrate to WinTel solutions where they could. In those situations, they would most likely consider SQLServer over Oracle. It would have been an overnight disaster to Oracle as much as Sun.

Thats why Oracle bought Sun.

Re:Bugger.

[thetoadwarrior]

It's possible that Oracle will try to sneakily phase out MySQL.

I hope so!

Re:Bugger.

[bill_mcgonigle]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

For whom? Oracle made the purchase, so that's the only place to look for benefits. They at least got hardware that could continue to scale their database for another few years while they retool to provide a hadoop-ish backend and a whole services organization. ZFS is nice to come along for the ride, but I doubt it figured heavily in the calculation (to me that's the nicest thing Sun had, but I didn't buy Sun so that doesn't matter).

Re:Bugger.

[DesScorp]

Can somebody show me something good to come from the Oracle-Sun deal? Anything?

Are you serious? Sun was slowly dying. Solaris was on the path to extinction. There are valid reasons to no want a GPL'd operating system, so if you wanted to run a commercially supported Unix in the enterprise, you'd be down to IBM, who seemingly is more enamored with Linux than with their own AIX (not to knock Apple; OSX is a fine server product, but Apple really isn't aiming for the same market that the Microsoft's/IBM's/Sun's are). Larry Ellison seems thus far geniunely interested in not only keeping solaris and sparc alive, but growing and developing them and growing their market.

What good did it do? It keeps us from having to pick Linux or Windows or zOS. It gives business another processor platform choice besides Intel and Power. Isn't choice a good enough reason for you?

Re:Bugger.

[h4rr4r]

OSX on the server? are you nuts?
Only worse option would be windows.

OS going away, or just "contractual support"?

[Darkon]

To be honest I didn't even know they provided "contractual support" for OpenSolaris, but surely the fact that they won't support you in using it doesn't nesessarily imply that it's being canned. Maybe it'll just be an unsupported "unstable" version that you can play with before getting "real" Solaris.

Re:OS going away, or just "contractual support"?

[seifried]

"Hi boss, yeah I'd like to use OpenSolaris. .. No, we specifically can't get support for it from Sun, I mean Oracle. .. Yeah if it breaks we are totally on our own. ... Ok so I guess we're not using OpenSolaris then."

Re:OS going away, or just "contractual support"?

[Per Wigren]

As you probably are aware of, there are TONS of mission critical servers out there running Debian, CentOS, FreeBSD, OpenBSD and other "there is no company you can blame and/or sue" operating systems, just as well as they run PostgreSQL or MySQL without support contracts for their mission critical databases.

For many companies that's not a problem because they have competent server admin staff and the community support is often way better than what you'd get for money.

An unsupported "debian-testing-style" OpenSolaris would make a lot of sense for both Sun/Oracle and many users. If you want support and someone to blame, just pay for Solaris instead. This model is already proven to work great: Fedora vs RHEL (vs CentOS), openSUSE vs SUSE Linux Enterprise, PostgreSQL vs EnterpriseDB, and so on.

Re:OS going away, or just "contractual support"?

[turing_m]

"Hi boss, yeah I'd like to use OpenSolaris. .. No, we specifically can't get support for it from Sun, I mean Oracle. .. Yeah if it breaks we are totally on our own. ... Ok so I guess we're not using OpenSolaris then."

That's not really any different from Fedora, yet businesses still use Redhat.

Re:OS going away, or just "contractual support"?

[scdeimos]

For many companies that's not a problem because they have competent server admin staff and the community support is often way better than what you'd get for money.

That latter part is debatable. Community support often yields conflicting (and incorrect) answers in my experience.

Mind you, I've been waiting for Microsoft to fix Windows Search's ability to find matches in Unicode text files since NT 4.0 days (find and findstr work from the command line, why not the GUI?) so the level of support is probably no worse.

Re:OS going away, or just "contractual support"?

[drinkypoo]

"Hi boss, yeah I'd like to use OpenSolaris. .. No, we specifically can't get support for it from Sun, I mean Oracle. .. Yeah if it breaks we are totally on our own. ... Ok so I guess we're not using OpenSolaris then."

That's not really any different from Fedora, yet businesses still use Redhat.

Uh, what? Redhat is RHEL, for which support is available. Fedora is RHEL beta, and is unsupported. Solaris is supported, OpenSolaris is unsupported. So in fact, it is entirely different.

Re:OS going away, or just "contractual support"?

[jacobsm]

My management would first ask whether it costs anything to use. If the answer is no then their answer is yes. Support, just ask the ether.

Re:OS going away, or just "contractual support"?

[vlm]

there are TONS of mission critical servers out there running Debian, CentOS, FreeBSD, OpenBSD and other "there is no company you can blame and/or sue" operating systems

Absolutely false inaccurate information, at least for Debian.

As per

http://www.debian.org/consultants/

"811 Debian consultants listed in 64 countries worldwide."

Now, you can hire a consultant whom might actually moonlight as a debian developer, perhaps even the maintainer of something that is critical to you. And, as a private citizen or at least small consulting company, you could sue them when/if they screw up.

On the other hand, if you think you you can sue microsoft, and win, next time exchange falls over, you are in for a BIG surprise.

Re:OS going away, or just "contractual support"?

[Per Wigren]

FOSS is FREE only if you don't value your time.

*gasp* I value my time but I also value flexibility and independence from vendor whims.

I have an equally naive cliche for you right here: Proprietary software is only cheaper if you are incompetent.

Re:OS going away, or just "contractual support"?

[Per Wigren]

You're right of course and similar setups are available for most other Linux and BSD distributions also. I was talking about running them unsupported as in download-ISO-and-install, I should had been more clear about that.

Re:OS going away, or just "contractual support"?

[blai]

FOSS is FREE because you value your time.

Re:OS going away, or just "contractual support"?

[Anonymous Coward]

That's not really any different from Fedora, yet businesses still use Redhat.

Uh, what? Redhat is RHEL, for which support is available. Fedora is RHEL beta, and is unsupported. Solaris is supported, OpenSolaris is unsupported. So in fact, it is entirely different.

Why are you trolling Martin?

The Sun page linked to in the article summary (http://www.sun.com/software/solaris/lifecycle.xml) has a section titled "Life Cycle Model for the OpenSolaris Operating System".

The page linked by the Phoronix article lists support dates for OpenSolaris 2009.06 (the last regular release) out to 2014.

One big difference between the regular Solaris ten-years-for-every-version support and OpenSolaris support is that if a new OpenSolaris release already exists, Sun may choose to make fixes in the current version only and ask you to upgrade if you want-- i.e. backport requests are not automatically approved.

Re:OS going away, or just "contractual support"?

[vlm]

I was talking about running them unsupported as in download-ISO-and-install, I should had been more clear about that.

I'm still not seeing that. There are plenty of folks whom will gladly contract to do that.

Oddly enough, my one and only brush with consulting/contracting with Debian was doing exactly what you say can't be done. Over half a decade ago, at my day job, I downloaded a Debian ISO and installed it on a Compaq DL/360 (back when those were new). It was a very special and unusual idea for co-located hosting. I set up and supported their box, helping them use frontpage and SCP or whatever until it was working perfectly and in production, at which point I turned it over to them like any other co-locate (where we only officially supported connectivity, electrical power, cooling, physical security, etc). My support work was an appendix to the usual colocation contract.

Re:OS going away, or just "contractual support"?

[FlyingBishop]

Proprietary software is only cheaper if you are incompetent.

And then only if your vendor is competent.

Re:OS going away, or just "contractual support"?

[turing_m]

Uh, what? Redhat is RHEL, for which support is available. Fedora is RHEL beta, and is unsupported. Solaris is supported, OpenSolaris is unsupported. So in fact, it is entirely different.

My interpretation of part of the point of OpenSolaris - Sun were using it as a testing ground for concepts that would make it over to Solaris, e.g. ZFS. AFAIK it's not exactly Solaris beta, but it is at least somewhat analogous. Both Fedora and OpenSolaris are unsupported, RHEL and Solaris are supported. I don't think you'd use either Fedora or OpenSolaris if you fear cutting edge stuff breaking on you, and if you fear that sort of thing enough to want something more stable, you might also buy support contracts, which would be less costly to provide because less stuff would break (hence RHEL and Solaris and their support contracts). But the utility of OpenSolaris/Fedora in business - someone in IT could get their feet wet for free with either Fedora or OpenSolaris and then make a case to management that they want to go with RHEL or Solaris after they have done proof of concept.

I could be wrong though.

Re:OS going away, or just "contractual support"?

[smash]

Have you read the GPL? No you can't sue them for stuff under the GPL.

However, there are plenty of mission critical systems running debian, etc...

Re:OS going away, or just "contractual support"?

[Elektroschock]

OpenSolaris is far too big to fail. It has a large user base.

You know even FreeBSD is still supported, OpenSolaris will be, too.

Unlike Linux OpenSolaris stands for real quality and maturity.

Re:OS going away, or just "contractual support"?

[jedidiah]

This is funny because before there was Linux there was the GNU toolchain being installed on top of SunOS.

FOSS is what you used if you valued your time.

Linux only gained traction because Sun gave x86 no respect.

Re:OS going away, or just "contractual support"?

[maxwell demon]

OpenSolaris is far too big to fail.

You mean, it will get a government bailout?

Re:OS going away, or just "contractual support"?

[vlm]

Have you read the GPL? No you can't sue them for stuff under the GPL.

Oh of course you can sue them. Anyone can sue anyone else for anything, and if they have more money, they'll win.

Lets say you sign a contract to "maintain our webserver in an operational status 99% of the time with no security breaches".

If your dude fails, you either get a contractually declared SLA payoff, or if you screwed up and did not include that in the contract, you sue them for malpractice or whatever. The license doesn't matter, except that open source stuff is better and more secure so there is a much higher likelihood that it'll actually work.

Business people don't care what license GCC and apache are under, they know you can't sue a big software company, so why bother considering that whole tactic? But, you can sue contractors and outsourced support personnel.

As if anyone could sue big corps like MS or Sun for providing bad software... Laugh.

Re:OS going away, or just "contractual support"?

[Anonymous Coward]

And then only if your vendor is also incompetent.

Here, fixed it for you. A competent vendor serving an incompetent client means "charge the sucker as much as he can pay!".

So:
* if you are competent, and your vendor is incompetent, better go with FLOSS.
* if you are competent, and your vendor is also competent, FLOSS is cheaper.
* if you are incompetent and your vendor is also incompetent, Propietary is the way.
* if you are incompetent and your is competent, stop paying through your nose and look for another vendor.

Re:OS going away, or just "contractual support"?

[JasterBobaMereel]

My time is expensive so I don't use Proprietary software, I want software where the support I need is available now, and free ...or if it really is a complex product (because it needs to be) support I pay for and actually is helpful

Most Proprietary products come with support as an afterthought, if at all, or charge a fortune for it if it is any good ...

Re:OS going away, or just "contractual support"?

[Elektroschock]

Indeed it makes sense to support certain software infrastructure elements by the government.

Operating systems and office productivity software are examples. Open solaris and Open Office could be developed and supported by the military.

Re:OS going away, or just "contractual support"?

[BrokenHalo]

Linux only gained traction because Sun gave x86 no respect.

I think Linux would have gained traction regardless. Sun has always had its diehards, often for very good reasons, but there are some serious players working with Linux (e.g. IBM and Cray, to name just two off the top of my head) who have seen enough value in that architecture to pursue it.

Re:OS going away, or just "contractual support"?

[Just Some Guy]

Have you read the GPL? No you can't sue them for stuff under the GPL.

What? I can think of a lot of ways you could sue over GPLed software. For example, if I hire you to add a given feature to a GPLed program, pay you, then you fail to deliver, then you bet I can take you to court (and win). The dispute is not with the software, but with your failure to perform your contractual duties. The legal protection that the GPL offers is this: if I release software under the GPL, and some random stranger has problems with it, then they can't sue me because it didn't meet their needs. If that random stranger becomes a paid customer, then they most certainly can.

Same with the Debian support contractors mentioned here. If you pay them to ensure a given system meets specific performance and feature requirements, then they're legally obligated to meet those goals. They don't get to say "well, it was the fault of this GPLed library - so sorry!" and walk away. That's the risk that you're paying the contractors to assume.

Re:OS going away, or just "contractual support"?

[SocietyoftheFist]

I know you want to believe that but you don't have the first clue about the majority of big business in this world if you really do. Since you think that TPB was worth of hero status I will chalk you up as another clueless open source supporter unwilling to admit that big business has taken Linux farther than it ever would have gone on its own. Want to see what no corporate support gets you? Take a look at the BSD camp. Most Linux kernel development comes from corporations now because there is good money in supporting it. The mantra of only incompetent staff needing support is woefully ignorant. Ever heard of risk management?

Re:OS going away, or just "contractual support"?

[tbuskey]

FOSS is FREE only if you don't value your time.

I spend lots of time debugging non-FOSS app nstallation scripts. The apps I deal with are often ports from Windows that make lots of eroneous assumptions about a Solaris or Linux box. The security settings are often wrong. They'll use a custom installation script instead of the OS standard packages. When the next version of the app comes out, I have to do it again.

Once the app is installed, I have to install & maintain a license server. When users can't run the app, I have to check the license server to see if it's running. Or if more people are trying to run it vs the number of licenses. If the licenses are in use, I can tell the user to wait a bit until someone else finishes. Or figure out how to kick a user off if he's gone on vacation. I have to balance the cost of the licenses vs the usage. I have to write my own tools or buy expensive ones to track that. If I need to move my server to new hardware, I have to reinstall every license after providing the vendor with information from my new server. Some vendors charge $$ for the rehosting, but most do not.

Some apps have lots of information on Google. Many only have the vendor docs which usually assume every just works. Or they require a custom viewer or search.

With FOSS, someone has usually created a package, rpm, etc for my OS. Or they use the standard, freely available configure and install tools (GNU configure, perl Makefile.PL, Ruby gem, etc). I'd love to see vendors use these install tools instead of writing their own. Just so I don't have to learn yet another tool.

FOSS doesn't have license servers. That frees up major parts of my time, speeds up users, etc. If I need to renew support contracts, the apps are not going to stop working because I'm a few days late. Most non-FOSS will extend a license when you call to review, but you get a stoppage from when the license expires until you get the new license from the vendor.

Most FOSS gets documented and put on the web. I can google others experiences in blogs, wikis and forums in addition to the official docs and web sites. Most of them will point out issues and workarounds that are not in the offical documentation.

If the app does what is needed and is not FOSS, I'll buy it. But there's no way a non-FOSS app using a license server is easier to support then a FOSS app. I'm not talking buggy apps. Those exist on both sides.

Re:OS going away, or just "contractual support"?

[rubycodez]

nope, a business would just apply for and get a stack of as many free (as in beer and lunch) Solaris licenses as it wanted for testing. Individuals can do it too, I have licenses to run Solaris 9, 10 and it cost $0 and you can download as DVD or CD ISO.

Re:OS going away, or just "contractual support"?

[BitZtream]

FOSS is FREE only if you don't value your time.

I have an equally naive cliche for you right here: Proprietary software is only cheaper if you are incompetent.

Yes, we got it, you're both idiots who have no idea what you're talking about and just spouting bullshit, we got it, lets move on.

Re:OS going away, or just "contractual support"?

[BlackSnake112]

How many of those companies are in the top 10, top 50, top 100 companies worldwide?

Re:OS going away, or just "contractual support"?

[grub]

FOSS is FREE only if you don't value your time.

Yep, everyone knows that Windows installs itself, configures itself, patches itself, etc. with no valuable user time needed.

Re:OS going away, or just "contractual support"?

[Per Wigren]

I don't know how many companies there are worldwide, but for the argument let's just say that there are 5 million. That means that 0.002% of the companies are in the top 100. There is a whole world of small-to-medium sized companies out there doing just fine. The millions-per-minute, or even thousands-per-minute servers are extremely rare.

Re:OS going away, or just "contractual support"?

[Per Wigren]

What the heck are you talking about? I very much admit that big business support is fantastic for Linux and Open Source!

Everyone seem so obsessed with these huge multinational corporations with thousands of servers spread out all over the world. I was talking about relatively small companies (the vast majority of companies) that has a small but competent IT staff, adminstrating servers that can affort to have 99,99% or even 99,9% uptime instead of 99,999%.

For the vast majority of server it doesn't make economical sense to pay the premium 99,999% uptime would cost. For these servers a community supported free operating system like OpenSolaris/BSD/Linux on low-to-midlevel hardware can be perfect!

Re:OS going away, or just "contractual support"?

[thetoadwarrior]

You would go for Solaris instead rather than being a tight wad and going for the freebie.

Re:OS going away, or just "contractual support"?

[metamatic]

Want to see what no corporate support gets you? Take a look at the BSD camp.

You mean like the 75 million systems running BSD derivative OS X? Not exactly a lack of corporate support or a big failure.

Or do you mean the BSD code improvements contributed by Apple, like the Zeroconf support, the self-defragmenting code for the filesystem, and so on?

Re:OS going away, or just "contractual support"?

[MikeBabcock]

First, who cares? Second, in terms of top-n anythings, supercomputing has an awful lot of nobody-to-sue OS usage.

Wait, you honestly believe that most companies run only proprietary systems?

If you asked me (a database programmer and consultant), I'd point out that only the small fries actually believe Windows is in their best interest. Most of the medium to large sized companies have competent IT staff who want to make things work on a decent budget and would rather spend the money on hardware than OS licensing.

Re:OS going away, or just "contractual support"?

[jackspenn]

Proprietary software is only cheaper if you are incompetent

And it isn't cheaper if you're using Apple's software. It is neither free as in beer nor free as in speech.

Re:OS going away, or just "contractual support"?

[Anonymous Coward]

For many companies that's not a problem because they have competent server admin staff and the community support is often way better than what you'd get for money.

Nope. Most organizations that host Debian/fBSD/CentOs, etc... not only have admins, but development staff, testing/QA org, SCM org, open source group, SEs, use case experts, etc.. basically a full internal software department mirroring a s/w company. All the Fortune 500 companies have an organization of that type. I only see 'just a good sysadmin' when you're just running a webserver and maybe outsourcing e-com (e.g. Visa) payments...

As for mission critical open source systems, name one that uses out of the box Linux, like out of the box Windows XP or OSX? Anyone running a mission critical app that has ubuntu updates turned ON and connected to the Internet? Likely not. Most orgs will have their internal development group customize it such that it will be a 1-way compatibility and likely break on apt-get updates--or in other words a full-custom app/system.

I'm waiting for the day when someone says: the rocket blew up, or the power plant shutdown, my stock trades failed, or the Visa server's lost data and say, "well it was because 'someone' put a new feature in the latest Debian build and we can't get a hold of him to figure out the conflict". Yes forget about suing, time lost, money lost, and time wasted in fixing the problems yourself cause the author has left the project.

Re:OS going away, or just "contractual support"?

[seifried]

My experience is Solaris shops tend to be a bit more.. ahem.. corporate in nature. I feel badly for Sun admins, Solaris is a pain to modernize, OpenSolaris was a heck a nice solution to that.

Re:OS going away, or just "contractual support"?

[Anonymous Coward]

I realize that but nobody, including the FreeBSD camp, says there are 75 million BSD users as a result of this. I use a Mac, got tired of making Linux work. I would always run in to one piece of hardware with non-existent or poor support so I said, hello Apple.

Re:OS going away, or just "contractual support"?

[Anonymous Coward]

You really believe that is a enormous pool of talented competent admins running all these servers? I've known shitty open source admins that could have used a support contract. You were originally arguing that it just takes a competent admin and acted as if every Windows admin is incompetent. I've done administration on a wide variety of UNIX platforms and Windows as well and a competent admin knows his platform and how to get what he needs done. It has been my experience that most admins that really knock on Windows don't have much actual experience with it. Of course with that being said, I prefer UNIX, but a competent Windows admin can keep a Windows system secure and running. And computing is all about the apps and there are things that only run on Windows.

Fork?

[migla]

Even if Oracle ditches Opensolaris all together, shouldn't the community keep going and shouldn't third party companies fill the hole left in the market with regards to support?

Or is this a question of reality not working out as the theory? Does that mean that, in a similar vein, Monty was right (and Eben was wrong) ranting and going to the EU about the fate of MySQL in the hands of Oracle?

(I don't know. I don't mean to imply anything. Just asking, sincerely.)

Re:Fork?

[Anonymous Coward]

There is no doubt that the acquisition of Sun by Oracle has an impact on the MySql product. Sun financed the development of the product for financial benefits and it is not in Oracles interests to do so. That means while MySql will not die and it will be forked it may or may not survive or thrive as well as it had Sun remained independent and in control of it under a different path. Obviously Sun had financial issues it needed to work out and being bought by Oracle was probably not the best route for anybody involved except the executives involved.

Re:Fork?

[Anonymous Coward]

The OpenSolaris community is awfully small. It really can't do very well to be honest. Couple dozen active people are not enough to keep an entire operating system afloat and meaningful.

Re:Fork?

[DennisZeMenace]

How big is that community really ? And what percentage of that community is actually made out of Sun employees ?

Re:Fork?

[c6gunner]

Even if Oracle ditches Opensolaris all together, shouldn't the community keep going

I doubt that OpenSolaris has enough of a following. If businesses ditch it due to a lack of support, it's unlikely that there will be enough of a "community" left to prop it up.

Personally, while I use OpenSolaris myself, I'd be more than happy to ditch it if the BTRFS project lives up to the hype. As far as I can see, ZFS is the only reason to prefer OpenSolaris over Linux for personal use, and I know that a significant percentage of the non-business users feel the same way.

and shouldn't third party companies fill the hole left in the market with regards to support?

I believe some already do. NexentaOS is built on OpenSolaris, and they at least provide support for their own products. I'm pretty sure they offer support contracts for OpenSolaris in general.

Re:Fork?

[Elektroschock]

Solaris is VERY common in the enterprises. Freebsd shows what a community driven Operating System may achieve.

Re:Fork?

[jedidiah]

One cool thing that could happen due to all of this nonsense is that newer Solaris only features can be liberated and added to Linux.

Oracle has a lot less motivation to keep ZFS away from Linux than Sun did.

Re:Fork?

[shentino]

When you're bleeding and drowning, you learn to expect sharks to show up.

Re:Fork?

[BrokenHalo]

Oracle has a lot less motivation to keep ZFS away from Linux than Sun did.

Hmmm. The guys at Oracle might be very nice people, but do you see them having a sufficiently wide streak of altruism to change their CDDL licence for ZFS to GPL? I wish I did.

Re:Fork?

[M. Baranczak]

Forget altruism. Oracle is heavily invested in Linux, so it's in their interest to help Linux development. Also, I don't see how freeing ZFS would actually cost them anything. So I'd say there's a fair chance that it will happen.

Obligatory Netcraft

[Anonymous Coward]

Breaking: Netcraft now confirms OpenSolaris is dying.

Re:Obligatory Netcraft

[smash]

Yet ZFS will live on, in BSD. ohwai...

Another "dead unix" for the collection.

[psergiu]

A/UX
IRIX
Unicos
Xenix
Ultrix
OSF/1

soon: OpenSolaris
and if Larry Ellison has a bad dream: Solaris

:-(

Re:Another "dead unix" for the collection.

[segedunum]

When IBM decices to cut maintenance and development costs on AIX, which they're already showing signs of doing, you can add that to your list.

Re:Another "dead unix" for the collection.

[tadas]

A/UX IRIX Unicos Xenix Ultrix OSF/1 soon: OpenSolaris and if Larry Ellison has a bad dream: Solaris :-(

But has Netcraft confirmed it?

Re:Another "dead unix" for the collection.

[jeffstar]

HP-UX didn't make your list?! I would have thought the writing was on the wall for it too, as well as everything HP except for printer ink.

Re:Another "dead unix" for the collection.

[DesScorp]

soon: OpenSolaris
and if Larry Ellison has a bad dream: Solaris

:-(

As far as Ellison is concerned, OpenSolaris probably doesn't equal Solaris. A lot of people thought that GPL'ing Solaris was a stupid marketing stunt that wasn't going to make Sun any money. Ellison has repeatedly stated that he wants to pump money into both Solaris and Sparc. The difference is that he's not going to fool around with the low end. He's going to refocus Sun/Sparc on the high end, which is where you can make a decent business case for a Sparc CPU like the T2. None of Sun's lower end processors... stuff like the UltraSparc III line... can possibly compete.

Sparc is still popular in Japan, and considering Ellison's fondness for Japan and its industrial culture, don't be terribly surprised if a deal is struck with Fujitstu to eventually produce Sun's hardware for them, with Oracle/Sun basically just doing most of the design and development work. If you ARE going to commit yourself to Sun/Sparc, then spinning off Fujitsu's semiconductor business and merging it with Oracle/Sun makes a lot of sense; a melding of Silicon Valley marketing/software/design with Japanese hardware industry. Since the T1 and T2's designs were GPL'd by the previous failed regime... yes, Schwartz... we're probably going to see a new CPU design in the coming years that will decidedly not be open sourced... and again, focused almost completely on the business high end.

Re:Another "dead unix" for the collection.

[Colin Smith]

DG/UX

I wanted to like OpenSolaris but...

[Viol8]

... I tried installing the latest version (as of november) on my year old laptop (ok , not its natural enviroment but if they want to compete with linux...) and it looked nice.

However, it didn't detect:

the wifi adaptor
the ethernet adaptor
the sound ship

wifi and sound I can just about live without , but no ethernet is a show stopper. If I can't access any networks then its little use for any real work. On the same laptop even years old opensuse 10.2 which I installed as a temporary OS when I first bought it saw the ethernet adaptor so we're not talking some brand new chip fresh out the design foundry.

So while I wish opensolaris all the best I think its going to remain a niche OS even if it doesn't get the chop.

Re:I wanted to like OpenSolaris but...

[c6gunner]

However, it didn't detect ...

Of course it didn't. It's not a desktop OS, even if it does have a purdy interface. Go check the hardware compatibility list - it's pretty friggin' small.

When I put together my home file-server, I made damn sure to check the HCL before purchasing any hardware. Even after doing that, I still had an issue with the on-board LAN chipset - had to compile a different set of drivers in order to stop it from dropping the connection every 5 minutes. OpenSolaris is a great server OS, but it's just silly to expect it to be compatible with some random laptop.

Re:I wanted to like OpenSolaris but...

[segedunum]

Even after doing that, I still had an issue with the on-board LAN chipset - had to compile a different set of drivers in order to stop it from dropping the connection every 5 minutes. OpenSolaris is a great server OS, but it's just silly to expect it to be compatible with some random laptop.

Comparable driver support with Linux, whether for laptops or for servers, is one of the things that has put Solaris on the slide over the past ten years.

Re:I wanted to like OpenSolaris but...

[Viol8]

"but it's just silly to expect it to be compatible with some random laptop"

I don't think so. Server farms will go with proper supported Solaris and yes they will check the HCL first. The freebie option is for other people who want to try out solaris and who will have all sorts of random desktop and laptop configurations. If opensolaris doesn't support much hardware then who exactly is it aimed at?

Re:I wanted to like OpenSolaris but...

[Kjella]

Of course it didn't. It's not a desktop OS (...) I made damn sure to check the HCL

Sorry, but this has nothing with being a desktop OS and everything about that in 2010 people expect to drop in an OS and have it work on any reasonable hardware. Maybe not the supported, recommended configuration but they still expect it to work. Windows and Linux runs on almost any server, Solaris/AIX/OS X Server etc. only come on their supported hardware and everything else is fail. You see exactly the same on desktops except there Windwos runs on almost any desktop, OS X only comes on its supported hardware and Linux is fail. It's just as unfair when people expect to plug in any USB printer or gadget in Linux and have it work, but the world isn't fair.

Re:I wanted to like OpenSolaris but...

[c6gunner]

If opensolaris doesn't support much hardware then who exactly is it aimed at?

Small business users, companies like Nexenta which produce their own server hardware/software products, and tech-savvy individuals looking for a home-server solution.

It's not exactly a huge market, but it is a niche (niches?) that needs to be filled. OpenSolaris is currently the best solution for projects such as mine. The ability to build a redundant array with automatic data corruption detection and a simple yet powerful snapshot functionality is what sold me on it. Nothing else on the market can do that, and the solutions which come close would have cost a lot more.

Re:I wanted to like OpenSolaris but...

[c6gunner]

You see exactly the same on desktops except there Windwos runs on almost any desktop, OS X only comes on its supported hardware and Linux is fail.

Linux is fail?

I have an 8 gig USB stick that boots into a fully-installed version of Ubuntu. I have used it on at least 6 completely different desktop computers, 4 laptops, and 2 netbooks. Each time I plug it into a different computer it boots and detects the new hardware without a problem. Out of all of those systems, the only exception has been an Asus EEPC on which the wireless card wasn't detected.

I'm not sure how you can consider that "fail". I've yet to see anyone do something comparable with windows.

That's not to say that linux isn't without it's problems - I still use windows as my primary OS - but it certainly does run on just about anything you can throw at it. The only thing keeping me from switching to it permanently is a problem with ATI graphics cards - mine works great ... but causes a memory leak which forces me to reboot every few days (annoying, but not an issue for users who actually like to turn off their computers).

It's just as unfair when people expect to plug in any USB printer or gadget in Linux and have it work, but the world isn't fair.

All of my USB gadgets work just fine on Linux. It even detects my old wifi usb dongle, which windows doesn't.

Re:I wanted to like OpenSolaris but...

[Anonymous Coward]

The kernel is fairly compatible with the BSDs and OpenBSD at least has superb 100% free support for most documented and many undocumented devices. Especially for laptops. When I looked into OpenSolaris it seemed that they were trying to follow BSD driver development.

However, let's face it. Nobody cares about Solaris enough to spend time porting drivers.

It has a horrible base system, and the alternative is using GNU tools. If it had been more consistent and attractive to Unix fans, like BSD derivatives are, it would still have a decent following. As it is, even Linux beats it in elegance and internal consistency.

A pair of features are coveted by some people, but as vocal as they are, they just want to transplant them to their favorite system. They don't care if the donor dies. Especially if he has an incompatible blood type.

Re:I wanted to like OpenSolaris but...

[crispi]

Yeah - same problem - of about nine or ten systems that I've tried it on (up to snv_133), all of them have at least one hardware problem.

eg from my memory

NIC drivers (Broadcom, Even Intel)
W/LAN drivers (Atheros for instance)
Display driver support (not just VESA!)
HW RAID drivers (Compaq, Promise)
AHCI drivers (including NCQ and hot plug support (slated to fix in snv 135)
AMD PhenomII support (fixed now since snv 126)

and I've had issues with the install (eg installation from USB CDROM)

However, saying all this, the journey is worthwhile - some features really are fantastic - especially together:

ZFS + snapshots + dedupe + Virtualbox VMs.

YMMV

Re:I wanted to like OpenSolaris but...

[Anonymous Coward]

I still had an issue with the on-board LAN chipset - had to compile a different set of drivers

You Linux people have been made soft by the likes of Ubuntu. Compiling a driver to fix a problem? Just another day for a slackware admin, or someone equally competent.

Re:I wanted to like OpenSolaris but...

[smash]

Opensolaris works *just fine* in a vmware virtual machine. Which includes workstation (testing, playtime), ESX clusters, etc.

And virtualisation is a big deal. Who cares what hardware the OS supports, so long as it can run under a hypervisor, which supports your actual hardware?

Re:I wanted to like OpenSolaris but...

[smash]

My win7 netboot/install image works on all the hardware i throw it at. Customising it was a piece of piss, too.

Re:I wanted to like OpenSolaris but...

[c6gunner]

Not the same thing. I've been dicking around with such images since BartPE first came out. It can be useful for certain things, but it's nothing like having a complete OS running from flash.

Re:I wanted to like OpenSolaris but...

[jedidiah]

There's really no good excuse for an x86 Unix to not have robust ETHERNET support.

This sort of nonsense is what kept Solaris x86 on the sidelines to begin with.

Not supporting the desktop frills is one thing, but ethernet is pretty fundemental to a Unix.

Re:I wanted to like OpenSolaris but...

[jedidiah]

It's not Ubuntu.

Even something like SLES will "pamper" you. It's about the level of hardware support, not how fancy the decorations are on the package manager.

Slackware won out over Solaris x86 "back in the day" over this very problem. Although then it was a bit more fundemental. Solaris x86 was SCSI only.

Re:I wanted to like OpenSolaris but...

[bill_mcgonigle]

There's really no good excuse for an x86 Unix to not have robust ETHERNET support.

Of course it has great ethernet support. But that's a layer above the NIC driver. And when a random Taiwanese chip vendor does a new rev, a compatible NIC driver doesn't fall out of the sky for any OS, it takes work, which can be easy or hard depending on the vendor.

Oh, for hardware vendors to support a handful of the most common non-Microsoft OS's.

Re:I wanted to like OpenSolaris but...

[bingoUV]

I agree driver support for OpenSolaris is not that great. While it has the disadvantage of lower driver availability over Linux, it also has a big advantage: kernel ABI stability. So a driver, once it is created for OpenSolaris, will keep working for a long time. Unlike the linux drivers which can break during any kernel update, and the driver needs to be actively maintained so that it keeps working on latest kernel versions.

One example that I see: Cisco VPN. On every new distro release, I have to find out the linux kernel version and figure out whether Cisco VPN will work with it or not. No such uncertainty with OpenSolaris, it just works.

We still use OSF/1

[Viol8]

A bloody good version of Unix it is too. 64 bit from the start back in the early 90s when PC manufacturers and Microsoft were still wetting their trousers about moving to proper 32 bit.

The alpha CPU - what a missed opportunity. Perhaps in some ideal world in an alternate reality people woke up to what a dogs dinner x86 is and the alpha chip had as much development effort put into it. I wonder what apps would be possible on a 2010 alpha chip that is still pie in the sky for x86?

Re:We still use OSF/1

[ubersoldat2k7]

I wonder what apps would be possible on a 2010 alpha chip

Alphlash? AlFlash? Flash for Alpha? Meh! They probably wouldn't support it

Re:We still use OSF/1

[HuguesT]

Actually most of what was very good in the alpha chip went to AMD and their Athlon64 chip. For a while they were even pin-compatible. Now Intel has the upper hand again, with no up and coming competitor on the horizon, except maybe IBM/POWER one day.

Re:We still use OSF/1

[LWATCDR]

There is ARM on the low end. The new ARM chips are getting faster and faster and compete well with the Atom right now for speed and blow them away for power consumption.
I would love to see an OMAP4 or Tegra 2 ITX board with enough sims slots for a few gigs of ram and a few SATA connectors.
Would make a great little NAS, SAN, or even small database server.

Re:We still use OSF/1

[tbuskey]

ARM

No, really. It's beat Intel in phones, mp3 players (ipod). It's coming on Netbooks.

From the bottom where Intel came from to beat the others.

Re:We still use OSF/1

[BlackSnake112]

I thought DEC's deal with Intel was what screwed the Alpha CPU. And that the PENT II was a CPU that had a RISC core with a CISC interpreter. Which was why Intel needed to pull the plug on the Alpha chips.

Side note I do remember running NT 4 on Alpha chips. It ran faster on the slower Alpha CPU then the higher clocked Intel CPUs.

Re:We still use OSF/1

[Anonymous Coward]

Actually most of what was very good in the alpha chip went to AMD and their Athlon64 chip. For a while they were even pin-compatible. Now Intel has the upper hand again, with no up and coming competitor on the horizon, except maybe IBM/POWER one day.

Heh, if you're relying on IBM to provide a *VIABLE* competition/alternative to ANYTHING, you're most definitely screwed. Even though IBM may be technologically capable of building the product, they'd only end up handing it to the OS/2 Marketing Department.

Re:We still use OSF/1

[machine321]

For a while they were even pin-compatible.

Wait, so you're saying the Alpha 21064 and some Athlon64 share the same pinout? Or do you just mean that they both have pins?

Re:We still use OSF/1

[TheRaven64]

AMD licensed the Alpha (EV6) interconnect from Digital. I'm not sure if the Althlon was ever mechanically compatible with the Alpha, but they used the same electrical signals to communicate with each other and with the the north bridge. There was some talk, before Intel came out with the Itanium hype, of producing motherboards that could take either chip, so that the Alpha could benefit from the same economies of scale as x86 chips. More recently, this idea was revived with IBM and others adopting Hypertransport.

Re:We still use OSF/1

[MikeBabcock]

Hypertransport itself, heavily used by AMD was an Alpha technology in the first place as well as I recall. Very neat stuff.

Re:We still use OSF/1

[Anonymous Coward]

But they can't run x86 code, it's only a matter o time for Intel to catch up, the same that did with the Athlon.

Re:We still use OSF/1

[LWATCDR]

I am not sure that the x86 ISA can match the ARM in the speed to power use race for the same price.
Also who cares if it runs x86 code. I am talking about servers here not a desktop. I guess you could run a SAN off Windows but their really isn't a good reason to. BSD and Linux are both very good at running SANs, Apache, MySQL, PostgreSQL, SQLLite, and Firebird. An Omap3, Omap4, and or Tegra2 will be more than powerful enough for a lot of tasks. They are powerful enough to run a PBX, Database server, or Intranet server, or SAN in many small to medium businesses.
None of these tasks require x86 compatibility.
For the desktop you could use such devices as clients. You could run Browsers, OpenOffice, email clients, and any of a number of other applications "nativity" and then use RDS to run any Windows Apps you may need. The only down side right now is the lack of Flash for these devices but even that may change soon.
For some places this would be a great solution.
Not everywhere for sure but I can see it working.
Also since your Internet access would be through a Unix like OS and on a none X86 ISA the none targeted security risks are a lot lower since your setup will not share many of the same exploits as the masses.

Blatant Troll

[Anonymous Coward]

Only mild flamage? You see this is why I prefer Linux!

Hardware/apps

[cbuosi]

I, like others, tested OpenSolaris and the 2 main problems that i saw where, 1) lack of support for fancy/new hardware. 2) not so many native programs as Linux/BSD. I think that OpenSolaris will live forever, but not as a OS, its bests features (ZFS, others) will be incorporated in linux/free bsd/ others)

Re:Hardware/apps

[hedrick]

Again, Solaris and Opensolaris are not currently desktop/laptop OS's. For a server you buy appropriate hardware.

Actually hardware support for desktops is acceptable. It's laptops that are weakest. You really have to choose your laptop carefully. But I can understand that this wouldn't be a priority for developers.

At any rate, the original posting is FUD. It's true that there is concern in some of the Opensolaris forums, because Oracle hasn't said anything about Opensolaris. But there's no particular reason to believe there will be trouble. The article that the posting points to says nothing that would imply problems for Opensolaris. To avoid developing Opensolaris, Oracle would have to come up with some other way to develop and test new Solaris features. The approach that would cause problems for the Opensolaris community would be to close the process. I wouldn't think having less testing of new features would in their interest, but we'll see. Oracle has a number of current opportunities to shoot themselves in the foot with Sun-related issues. This is one of them.

Re:Hardware/apps

[HuguesT]

ZFS will probably have to be reimplemented somehow to go on Linux. We'll have to wait for ext5 or 6 to get a reasonable subset of ZFS feature list.

Re:Hardware/apps

[drinkypoo]

I think that OpenSolaris will live forever, but not as a OS, its bests features (ZFS, others) will be incorporated in linux/free bsd/ others)

They won't be incorporated into Linux unless Oracle changes the license. Sun chose the license for OpenSolaris specifically to make it incompatible with Linux's license, in order to prevent Linux from gaining a further competitive advantage over Solaris, mostly in the form of ZFS. The best features might be replicated in Linux. Of course, BSD already has ZFS, because their license is compatible... Which of course has stolen some users from Linux.

Re:Hardware/apps

[mikechant]

ZFS will probably have to be reimplemented somehow to go on Linux. We'll have to wait for ext5 or 6 to get a reasonable subset of ZFS feature list.

Sort of. There will probably never be an ext5, ext4 will be stabilised at some point. The future 'standard Linux filesystem' with ZFS features is intended to be BTRFS and it's well on the way (in the mainline Linux kernel but not ready for general use just yet).

http://btrfs.wiki.kernel.org/index.php/Main_Page

However, the fact that Oracle is the priciple driver behind BTRFS, but now owns Sun and thus could GPL ZFS does obviously cast some doubt on the future of both - although they can both carry on with non-Oracle devlopers, it will obviously be very important which one Oracle throws its weight behind (they're surely unlikely to give them both equal resources).

Re:Hardware/apps

[turing_m]

However, the fact that Oracle is the priciple driver behind BTRFS, but now owns Sun and thus could GPL ZFS does obviously cast some doubt on the future of both - although they can both carry on with non-Oracle devlopers, it will obviously be very important which one Oracle throws its weight behind (they're surely unlikely to give them both equal resources).

I'm curious - what do you see as the advantage to Oracle of making ZFS GPL?

Re:Hardware/apps

[bingoUV]

Not the GP poster, but...

Oracle is a "Linux company" in more ways than one. Linux is the principal development platform for major Oracle products. Oracle (as in Oracle Enterprise Linux) does Linux support (similar to RHEL), though I am not sure whether Oracle's contract is cheaper/better than RedHat's. Oracle also sells hardware/software combination and sends a representative to install and configure it according to the buyer's preference. Before Sun acquisition, the only operating system for this used to be, you guessed it, Oracle Enterprise Linux (On HP hardware). After Sun acquisition, it is Sun hardware but most likely the OS will still be Linux. Most medium-end customers of Oracle are more comfortable with Linux. Extremely high-end customers, though, prefer Solaris. Oracle can view this situation in 2 ways:

1. Any value addition to Linux, is a good thing for Oracle. Since Oracle employs ZFS developers, the customers who want to use ZFS will be more comfortable buying the full Oracle solution hoping, not without reason, for a better support than RedHat, Novell etc.

2. On the other hand, Oracle can consider Linux to be a competitor to Solaris. If ZFS is not GPLed, medium-end customers who currently use Linux will be pushed to upgrade to Solaris if they want to use ZFS.

I see more sense in 1, but 2 is also not entirely without merit.

Article Doesn't Quite Say it, But Not Suprised

[segedunum]

The article doesn't quite say it, and it doesn't have the smoking gun of "We're canning OpenSolaris", but that end of life page for OpenSolaris looks pretty damn final to me and there is little room for interpretation.

I wouldn't be surprised if Open Solaris went the journey. The whole point of it was to arrest the slide of Solaris in the face of Linux, in particular, and so that Sun could tell everyone that Solaris was open and just like Linux. Unfortunately, OpenSolaris has contributed little, if anything, to Solaris. There's no community of developers apart from those Sun sanctioned and things like Solaris's driver support is still a long way behind where Linux is. Development still hasn't been opened and there is no public repository development model. Sun, or Oracle now, is bankrolling it with none of the cost savings you would expect from such a project.

One can only hope that Oracle won't follow the same 'strategy' that Sun have followed for the past ten years, because it got Sun into trouble and it'll cost Oracle rather a lot of money if they get it wrong. However, they look as if they're doing swift about-turns on that and a statement of their future intent is clear when you go to www.sun.com - it redirects straight to www.oracle.com.

Re:Article Doesn't Quite Say it, But Not Suprised

[Anonymous Coward]

The article doesn't quite say it, and it doesn't have the smoking gun of "We're canning OpenSolaris", but that end of life page for OpenSolaris looks pretty damn final to me and there is little room for interpretation.

The Oracle page lays out a software support policy for OpenSolaris releases and, following the policy, specifies end-of-support dates for existing releases. Oracle generally does not talk about specific release or support dates for future versions of software.

Given those facts, what on the page makes you think that there won't be another OpenSolaris version? What on the page is different from the end-of-support date pages for the Oracle RDBMS?

I detect a whiff of speculative FUD coming from both articles.

Re:Article Doesn't Quite Say it, But Not Suprised

[cpghost]

Development still hasn't been opened and there is no public repository development model.

ON sources (i.e. the base sources + kernel, something like FreeBSD's /usr/src tree) are under Mercurial though, and one can easily clone the repository. So they're moving in the right direction.

Re:Article Doesn't Quite Say it, But Not Suprised

[Anonymous Coward]

Hopefully Oracle has enough experience with its Unbreakable Linux and how the whole RH/Fedora ecosystem works, that it can use that knowledge to make a Solaris/OpenSolaris ecosystem work too.

It's that or kiss Solaris good bye... OpenSolaris is pretty much its last chance to get mainstream. But Sun hasn't been doing a good job at opening it and now there are a lot of questions. If Oracle is commited to something like RH/Fedora, even then it will take a while. Development of OpenSolaris is way too closed... and if you search around, you can find people complaining Sun didn't benefit much from open sourcing it. Of course, half assed project it was so far, how could it benefit fully.

Re:Article Doesn't Quite Say it, But Not Suprised

[Anonymous Coward]

but that end of life page for OpenSolaris looks pretty damn final to me and there is little room for interpretation.

Are you mad?

"This describes the OpenSolaris End Of Service Life timeline for OpenSolaris OS releases."

"Sun provides contractual support on the OpenSolaris OS for up to five years from the product's first General Availability (GA) date as follows."

It's not an EOL page for OpenSolaris, it's the EOL schedule for OpenSolaris RELEASES. There IS little room for interpretation, how the fuck did you get this so twisted?

Re:Article Doesn't Quite Say it, But Not Suprised

[segedunum]

The Oracle page lays out a software support policy for OpenSolaris releases and, following the policy, specifies end-of-support dates for existing releases.

Yes it does, because they're obliged to do so to meet support obligations. That's what usually happens when a product is getting end of lifed and being wound up. They have to lay out what happens to their obligations.

Given those facts, what on the page makes you think that there won't be another OpenSolaris version? What on the page is different from the end-of-support date pages for the Oracle RDBMS?

Because this wasn't already on the cards at Sun. This is part of an Oracle review of Sun's products and software, and that's what they've decided. As the article summary says, compare that with the Sun schedule. If you also read the mailing list conversation you'll see that any mention of OpenSolaris has been yanked from the support pages, which means they will support existing versions under their obligations but there will be no new support commitments.

Re:Article Doesn't Quite Say it, But Not Suprised

[segedunum]

It's not an EOL page for OpenSolaris, it's the EOL schedule for OpenSolaris RELEASES. There IS little room for interpretation, how the fuck did you get this so twisted?

Because I read the articles and the mailing list conversation. This is an EOL without any new support subscriptions or commitments. That means nothing new will happen.

Well this sucks...

[jedirock]

I was planning to build a file server using OpenSolaris in the coming weeks, but I may have to rethink that now.

Anyone know a good place to get access to ZFS in another place? Would BSD or FUSE on Linux be better?

Re:Well this sucks...

[Architect_sasyr]

So long as it's FreeBSD 8 and not 7, a properly tuned and setup ZFS install is a breeze to put together. It took me maybe 20 minutes to kernel tune mine (i386 chipset and less than the recommended RAM in it at the time) and it's got good stability on a 1.7TB raidz unit. YMMV, but I wouldn't stick 7 back on another box again. I've no comment on FUSE.

Re:Well this sucks...

[turing_m]

Anyone know a good place to get access to ZFS in another place? Would BSD or FUSE on Linux be better?

FreeBSD - ZFS is no longer in experimental status as of version 8.0. I haven't heard anyone recommend FUSE on Linux. As far as other BSDs go, I know that at least OpenBSD has no plans to include it at this stage.

http://kerneltrap.org/mailarchive/openbsd-misc/2009/1/17/4750954 - But that was over a year ago.

At the moment I'm learning FreeBSD over OpenSolaris because I want ZFS, FreeBSD is fully free and open source, FreeBSD looks to have a wider array of ports, which should be easy to install, even though with the LiveCD of OpenSolaris it boots up straight to X. On a production server or maybe even workstation, I think the choice would be down to FreeBSD versus Solaris, rather than OpenSolaris. Someone correct me if I'm wrong. Solaris does have a lot of nice features though, FMA (fault management architecture - lets you know when something has gone kaput and what to do about it.) And FreeBSD will lag in terms of the version of ZFS it supports. Deduplication looks to be a pretty cool feature - if you copy some data to another part of the HDD, and then you leave it a bit and your hoarding nature kicks in and you don't know whether you can delete it or not - no fear, ZFS will recognize the data as the same, only store it in one place (unless modified of course) and so there is no benefit to deleting the copy other than being a neat freak.

I'm presently wrestling with setting up FreeBSD on wireless. After that I have to get X set up. It would be nice if FreeBSD had version specific handbooks ala PostgreSQL, but they don't. So it's a combination of man pages, handbook, googling, etc to get me where I want to go. It's a bit of a contrast to Ubuntu which I set up on another box in the space of about an hour, including updates. Unmetered FOSS mirrors on ISPs kick ass!

Anyway, I suspect that the user base of FreeBSD will grow by leaps and bounds when people realize the advantages of ZFS and don't want to wait for BTRFS or whatever the results of this meeting might be: http://blogs.sun.com/bonwick/entry/casablanca

Re:Well this sucks...

[drinkypoo]

FUSE involves overhead. You don't want to use it when you're trying to achieve maximum performance.

Re:Well this sucks...

[Anonymous Coward]

ZFS has been ported to FreeBSD 2 years ago. It's quite stable there by now and there are people using FreeBSD with ZFS on mission critical servers. Big ones too.

Just, a few features are missing, since FreeBSD does not have bits for it in kernel(sharesmb and shareiscsi properties do nothing in FreeBSD since the kernel does not have native suport for serving those protocols, you'll have to installa samba and create widnows shares the old way).

Re:Well this sucks...

[jedirock]

Not too worried about performance right now. It'll essentially be a NAS over 1Gbit/s, so as long as it's faster than that.

Re:Well this sucks...

[moosesocks]

ZFS itself isn't exactly a performance screamer, although it offers some of the best volume management and redundancy options in the business. A few features (such as the ability to grow a raidz pool) are still waiting on the sidelines, although they are for the most part very competitive with what you'd get from an enterprise-grade RAID controller.

Of course, if you want high performance and redundancy, you'll still need to resort to one of those expensive RAID controllers, and something like iSCSI, which allows for graceful and transparent failover in the event of a hardware failure in the array itself.

Re:Well this sucks...

[0100010001010011]

Latest FreeBSD supports ZFS on root. Which is nothing short of awesome.

I haven't tried yet, but I want to pull one of the hard drives and see if it still boots (it should).

I have 2x2TB as / and 5x1.5TB in RAIDZ2 for my multimedia archive. Irreplaceable family photos are on both (and offsite backed up.)

It's taken me a while to get used to the "BSD way" over the "Linux Way" and still frequent google with "Freebesd how to X"

Re:Well this sucks...

[Anonymous Coward]

Try PcBSD, you'll like the install process and the MacOS-like installation packages (self-contained). It's based on FreeBSD.

Re:Well this sucks...

[cbhacking]

Not sure if the project is still alive or not, but might the DesktopBSD bits be useful here? FreeBSD with a pre-installed/configured X server and KDE. I tried it once, and rather liked it, but it's kernel was a full version behind the actual FreeBSD codebase (6.something when FreeBSD was at 7.2 or somesuch). There was a beta version using 7.x that I never got around to trying, and this was all several years ago, but at the time it was a pretty nice setup, and according to their site you could simply install their userland components on top of a base FreeBSD system if you wanted to, and get what was essentially the result of using their nice graphical installer.

Re:Well this sucks...

[Anonymous Coward]

FWIW, as someone who has invested somewhat heavily in the past in 3rd party filesystems and Linux... Be prepared to stop using ZFS under FreeBSD.

They can't take it away but unless the community fully takes it over and essentially forks it from Sun, there will possibly be a time when it simply won't be the filesystem you think it is. I had some large GNU/Linux systems with XFS and a couple systems with JFS and at one point I even had some ReiserFS systems (which was sort of supported) IBM and SGI seem to have taken the "port it and it's done" approach and keeping their code up to snuff has fallen on to the community which just isn't that interested, a couple SGI guys kind of kept fighting the good fight for a while but it looks like it has petered out, IBM simply dumped it over the fence. There were routine crashes when I'd update to a kernel provided by a distribution maker. Then there was also the aspect of which rescue discs supported JFS or XFS, many seem to but make sure they work before you bank on it, the whole idea in the first place was to have robust filesystems to keep your data safe but if you suffer a crash and your rescue discs won't access it then what do you do? Ubuntu supports JFS as an option and I've seen it crash when loading the module and then attempting to mount a filesystem. Then there have been some shifts in greater Linux that have lagged on these systems, SELinux requires extended attributes which weren't supported and then not supported in some configurations for a while. (If I'm not mistaken, XFS had xattr but then XFS on LVM didn't for a period of time, for who knows what reasons.. seems a very odd limitation) After a few years it just felt like I was on an island. Then there were the barrier issues when the IO subsystem was redone, these supposedly "faster" filesystems had some not so pathological cases where they'd slow way way down, it was fixed with a configuration change but it was just one more thing you had to do differently..

Don't get me wrong, for certain types of servers or something it might be a fine way to go. The filesystems work and I never did lose data (except to reiserfs) but it sure felt like I was getting to a point where it could happen. I cut everything over to ext3 and haven't really thought about it much since then. On BSD, if ZFS doesn't become the filesystem of choice, then if Solaris closes source and maybe 3 revs of BSD later, it could easily be an island. ZFS does have the pooling and such so there are some things that are potentially very useful and unique to it but I never needed the things XFS could do that Ext3 couldn't and even now the limits of ext3 and ext4 are still well beyond what just about everyone will need and btrfs should be good by then. I remember when FFS didn't get replaced with a logging filesystem because they added soft updates, if those guys have their own plans to make ffs2 or something and they are closer to the core BSD team then ZFS will be an island.

Re:Well this sucks...

[Anonymous Coward]

Version specific handbooks: http://docs.freebsd.org/doc/

Re:Well this sucks...

[domatic]

I wouldn't deploy a FUSEed filesystem as a server filesystem. I'm glad to have the option in case I need to use a live-cd or something to retrieve files. I can't see it being good for much more than rescue purposes.

Re:Well this sucks...

[Colonel Fahlt]

The porting of ZFS to NetBSD is in progress.

http://wiki.netbsd.org/users/haad/porting_zfs/

NetBSD is also seeing work done on integrating DTrace, incidentally.

http://mail-index.netbsd.org/tech-kern/2010/02/15/msg007333.html

I'm not very familiar with DragonFlyBSD, but they have a different approach called HAMMER.

http://kerneltrap.org/DragonFlyBSD/HAMMER_Filesystem_Design

Re:Well this sucks...

[Trixter]

5x1.5TB in RAIDZ2

That wasn't very efficient. You could have had 14% more space with 50% less CPU if you'd gone with raidz. raidz2 was intended for 6 drives or more, not 5 or less.

Re:Well this sucks...

[0100010001010011]

Understandable. I'm just paranoid. This is going to be holding irreplaceable family photos and movies once I get to digitizing. I guess my biggest fear is to have one drive fail and while I'm restoring another one go.

Now I can have 2x drives fail. And my offsite backup site implode.

Re:Well this sucks...

[Anonymous Coward]

And Dtrace, whatever happened to that ..?

Re:Well this sucks...

[TheRaven64]

HAMMER isn't really directly comparable to ZFS. They are both modern filesystem designs, but they have very, very, different sets of design constraints and expected use cases. HAMMER will probably be ported back to FreeBSD at some point (Matt Dillon wrote some docs explaining how to, it just needs someone to bother doing it), but there aren't many situations when it would be difficult to pick whether ZFS or HAMMER is more appropriate.

Re:Well this sucks...

[TheRaven64]

Deduplication looks to be a pretty cool feature - if you copy some data to another part of the HDD, and then you leave it a bit and your hoarding nature kicks in and you don't know whether you can delete it or not - no fear, ZFS will recognize the data as the same, only store it in one place (unless modified of course) and so there is no benefit to deleting the copy other than being a neat freak.

Actually, ZFS does that without deduplication. It's a copy-on-write filesystem. When you copy a file, it creates a link. When you modify one copy, it duplicates the modified blocks and updates one of the copies to reference the new blocks. Deduplication allows it to combine identical blocks from different sources. For example, if you do periodic snapshots from a dozen workstations onto your ZFS server, dedup will let it store only one copy of the core OS files, rather than one copy per client per snapshot. For a home user, it's not a huge win, but it might still be useful if you've got some spare CPU power.

Re:Well this sucks...

[turing_m]

On BSD, if ZFS doesn't become the filesystem of choice, then if Solaris closes source and maybe 3 revs of BSD later, it could easily be an island. ZFS does have the pooling and such so there are some things that are potentially very useful and unique to it but I never needed the things XFS could do that Ext3 couldn't and even now the limits of ext3 and ext4 are still well beyond what just about everyone will need and btrfs should be good by then. I remember when FFS didn't get replaced with a logging filesystem because they added soft updates, if those guys have their own plans to make ffs2 or something and they are closer to the core BSD team then ZFS will be an island.

Thanks for your comment. Frankly the license does worry me, though I have to make a choice. My gut feeling was that FreeBSD is the largest BSD and so if they felt that the license risk was sufficiently low to put the sort of work into it that they have, then that would be good enough for me. It's not like they haven't been through legal battles before - BSD was given a baptism of legal fire. (Mind you, I'd certainly prefer it if OpenBSD was porting ZFS.)

My other gut feeling is much stronger, a conviction. And that is this: the solution to silent data corruption on the HDD (or SSD) that ZFS provides is a killer app. Unlike XFS, it has the potential to make also-rans of other file systems without equivalent features (and those with equivalent features but without sufficient testing). If you are running a server (which is the focus of FreeBSD) and you care about data integrity, ZFS is a no-brainer. RAIDZ will allow you to sleep at night, especially with inexpensive hardware. For much the same reasons, I insist on PostgreSQL over MySQL because it has a reputation for not treating my data like junk. And now with ZFS I have the opportunity to build my castle on a foundation of granite. The developers of FreeBSD must be equally aware of the advantages, considering that they first ported it over 3 years ago whereas it took me until a couple months ago to even realize what the hype was about. So I'd be very surprised if FreeBSD let ZFS become an island.

Re:Well this sucks...

[turing_m]

Try PcBSD, you'll like the install process and the MacOS-like installation packages (self-contained). It's based on FreeBSD.

I notice that it's now based on the production release instead of the release candidate. I might give it a whirl. My thoughts were that I'd be better off getting to grips with FreeBSD because of the benefits of the larger community.

Re:Well this sucks...

[turing_m]

Actually, ZFS does that without deduplication. It's a copy-on-write filesystem. When you copy a file, it creates a link. When you modify one copy, it duplicates the modified blocks and updates one of the copies to reference the new blocks. Deduplication allows it to combine identical blocks from different sources.

Thanks for the heads up, I did not know that. Dedup should still be useful if you are migrating redundant data from another filesystem rather than making fresh copies. For backups, I have to wonder how many identical blocks there are going to be in two zip files containing largely the same data, since I was under the impression that if the data changes then the files might be compressed a different way and so have different blocks. I hope I'm wrong.

Re:Well this sucks...

[hotfireball]

But keep in mind that FreeBSD is a slowest OS. And probably the most conservative and really old at ecosystem. And also has completely f*cked up release time schedule (even Warner Losh himself agreed on that). As well as quality is quite shitty for other things than just a network router. And threads there are mostly crap.

OpenSolaris has *valid* open source license, so your statement "I want fully free and opensource, hence FreeBSD" -- is sounds more like a FUD. :-)

YMMV.

Re:Well this sucks...

[Anonymous Coward]

have you gotten to the bit yet that says FreeBSD zfs is lagging 2 years behind opensolaris version?

Re:Well this sucks...

[TheRaven64]

You might get better results from using a compressed ZFS volume and dedup than from a zip. As I recall, the start of a zip file will always be the same if the first files into it are the same, but everything after a modification in the input stream will be different, so you won't gain anything from dedup.

If you're just backing up a single machine, your best option is to rsync your filesystem to a compressed ZFS volume in a cron job, then snapshot the ZFS volume on that machine. This will give you incremental backup, history, and so on. Turning on dedup might give you a small extra boost, but probably not much unless you are making lots of redundant copies. If the machine you are backing up is also using ZFS, then you zfs send / receive instead of rsync.

Re:Well this sucks...

[turing_m]

But keep in mind that FreeBSD is a slowest OS. And probably the most conservative and really old at ecosystem. And also has completely f*cked up release time schedule (even Warner Losh himself agreed on that). As well as quality is quite shitty for other things than just a network router. And threads there are mostly crap.

I'm not sure how relevant that Phoronix comparison is because the filesystem used was UFS, not ZFS. However, unless we are talking about Gcrypt (which I have never used), the slow performance of FreeBSD is acceptable for my purposes. For my purposes, data integrity trumps speed. If I were the type to prioritize raw speed over correctness I would have flunked out of engineering school through not double checking my figures, chosen MySQL over PostgreSQL, etc. Conservative for an ecosystem is good too. Maybe old goes along with that, though for example PostgreSQL has 8.4 in ports from what I understand, which is recent. Maybe the release schedule will bug me, maybe not. Release schedule? It's not something that really worries me, as long as the operating system does what I want and the ports (or repos, or whatever) are comprehensive and up to date. What exact quality is bad about FreeBSD? To me, that sounds like FUD. ;)

OpenSolaris has *valid* open source license, so your statement "I want fully free and opensource, hence FreeBSD" -- is sounds more like a FUD. :-)

Maybe you are right. Other people from Sun have said that, and maybe it is valid (maybe the FSF is the source of the FUD, though I trust them well enough, anecdotes from Theo about RMS holding up an airplane notwithstanding). e.g. http://www.mail-archive.com/opensolaris-discuss@opensolaris.org/msg13877.html I would like to see what direction Larry Ellison takes things before extending full trust to Oracle and what Sun is now that Oracle holds the whip hand.

At the same time, since something like ZFS is so groundbreaking, I would like to support the creation of it financially once whatever I do with it has some financial returns. It seems wrong that Sun develop something like ZFS, release it FOSS, and not benefit financially when there were certainly costs involved.

Re:Well this sucks...

[hotfireball]

in fact, CDDL has better legal defense. Bruce Perens still has a hard time to convince FSF to get this resolved. Hence why fat companies biting open source and trying to steal property. :-(

ZFS: well, it was always funny with these FreeBSD guys... :-) I was asking usually: look, if you need ZFS and that's the main point, so why do the heck you need to run it on FreeBSD through solaris kernel mapping, instead to run it natively on actually Solaris itself? The answer is pretty clear: religion. The thing is, that they rather would like to eat outdated ZFS with way lower version (and more bugs + FreeBSD bugs) and cry it hangs their FreeBSD, but they still continue presenting how wonderfull FreeBSD is, using Macs or Windows/PowerPoint or Linux... :-) I mean... why FreeBSD if there is Solaris anyway?

Shortly, if you're not doing a router or a firewall, then simply I see zero merits messing up with that really old stuff, that has irregular releases. Itself, OS is not any bad, but rather I'd say good. However, supporting it is quite a mess. Besides, Yahoo really-really wanted to go with Java. But they went with mess like PHP (really crying) only due to the one single reason: their infrastructure is entirely on FreeBSD since long time. And Java VM's threads on FreeBSD are really that bad, because they're linked to OS's ground. If Yahoo would move to at least Linux a decade ago, then you would never see any PHP on Yahoo projects. So one more thing to think: if you picking up FreeBSD, would it render to a severe side-effects like in Yahoo case?

Anyway, good luck in what you're doing. ;-)

Re:Well this sucks...

[hotfireball]

I would like to see what direction Larry Ellison takes things before extending full trust to Oracle and what Sun is now that Oracle holds the whip hand.

Oracle will build their Solaris on top of OpenSolaris. It is just Fedora/RHEL analogy. You should be fine with OpenSolaris if you hacking or developing or fine supporting things yourself. Or, if you like, you can pay $$$ for Oracle's Solaris in case you move your stuff to the enterprise and need papers/support for CYA.

Please keep in mind that taking too much risk on your shoulders only because it is emotional moves to follow FreeBSD -- not really safe. With Solaris you can quickly convert your stuff to defended set by getting an official support. In FreeBSD case?.. Who supports it with a very big name to make sure your CEO won't get a heart attack?..

That's what I am talking about: tech stuff is fun, but you also need to think about covering your ass to sleep fine during nights. ;-)

while we're here, what about linux zfs

[drfireman]

So far as I can tell, zfs is the only piece of opensolaris that's exciting enough to make anyone want to install if if they'd otherwise want to install a linux distribution. With that in mind, could someone post an authoritative update on the supposedly intractable licensing issues that prevent ZFS from being incorporated into the linux kernel? Is it still hopeless?

Re:while we're here, what about linux zfs

[mr_da3m0n]

Well it's fairly simple. OpenSolaris is licenced under the CDDL, which is incompatible with the GPL, which is the license the Linux kernel is released under. Nothing "supposed" there, it's a fact.

It is, however, compatible with the BSD license, which is why FreeBSD has ZFS support now.

Re:while we're here, what about linux zfs

[Mr Thinly Sliced]

Well according to the wikipedia info (not sure how up to date it is) - the problem is that Sun chose (on purpose) an open license (CDDL) that makes distributing a derivative work of it and GPL software illegal.

Even a clean room implementation may have issues due to patents.

You can apparently try to run it in userspace (that's the FUSE stuff the other posters are talking about) but that's a messy solution for sure.

Chances are we'll have a production btrfs before we get an in-kernel ZFS implementation.

Re:while we're here, what about linux zfs

[Jahava]

It would be nice if, should it intend to pull the plug on OpenSolaris support, Oracle would do a GPL release of ZFS ... or maybe the entire OpenSolaris operating system (kernel, userspace, etc.).

It seems like a huge waste to lose a huge resilient code base like that to obscurity. Plus, Linux is definitely one of Oracle's strategic technologies.

Re:while we're here, what about linux zfs

[Trepidity]

I believe this Wikipedia summary is as good as an update as anyone has of the progress and likelihood of future progress. An alternative is FreeBSD 8 (released Nov. 2009), which includes ZFS as an officially supported feature for the first time.

Re:while we're here, what about linux zfs

[Paul Jakma]

You've got the direction wrong. The CDDL is compatible with both those licences. I.e. you won't violate the CDDL by porting it to BSD or GPL works and distributing it. The GPL and BSD licences are not compatible with the CDDL though, in the sense that the CDDL is more restrictive than those licences. Note that FreeBSD ZFS is *not* in FreeBSD core (and never will be?) precisely because of it this, last I checked. Also, Linux could easily have the *same* level of support, if someone wished. Nothing in the CDDL would prevent it.

A lot of people see Sun as having deliberately chosen this licensing situation. Whether that's really true or not, who knows - however it does seem to be the general perception.

Re:while we're here, what about linux zfs

[Anonymous Coward]

almost exactly the same as the license issues that stop FAT32 being in the kernel. microsofts implementation of FAT32 was not GPL compatable, so the code could not be copy'n'pasted into Linux. Some one had to sit down and write a GPL implementation of FAT32. of course there is no promise that MS wont say you infringed their patents.

Re:while we're here, what about linux zfs

[Anonymous Coward]

isn't it funny that there are bits of freeBSD that are not BSD licenced. So if you though you could take freeBSD and use it for whatever you want, you have to check the licence of every component.

Re:while we're here, what about linux zfs

[Alcoholic Synonymous]

Note that FreeBSD ZFS is *not* in FreeBSD core (and never will be?) precisely because of it this, last I checked.

It's not in the core... but it is in base. FreeBSD ships with full support for ZFS (since 7.0) and only requires zfs_enable="YES" in /etc/rc.conf.

If you are using FreeBSD in a device and don't want or cannot use ZFS, there are several settings (WITHOUT_ZFS, WITHOUT_CDDL, WITHOUT_OPENSOLARIS) and such that can be dropped into /etc/src.conf to omit these bits completely from your build.

Sources for ZFS and other bits of non-BSD licensed software (that may be redistributed) are found under src/contrib and src/sys/contrib, where they can be easily segregated from the "pure" BSD bits.

Re:while we're here, what about linux zfs

[Rysc]

There are some other nice things about solaris/opensolaris. Linux distributions could learn a lot about integration from looking at how solaris solves some of its problems. Not that they're perfect, but compared to Linux "throw everything in a big pile" distributions it's better.

And of course there's DTrace. too.

Re:while we're here, what about linux zfs

[Ant P.]

One of the major problems is that ZFS was designed as a huge blob of interdependent code, or in other words the complete opposite of the layered Linux VFS design. Even if they hadn't intentionally gone with an incompatible license it would still be a nightmare to port into the kernel, and that's effort that could be spent doing far more productive things (Btrfs, Tux3, Reiser4 etc.)

Re:while we're here, what about linux zfs

[Anonymous Coward]

Now I remember why I stopped posting to Slashdot. The mods.

Re:while we're here, what about linux zfs

[cbhacking]

How would this work with Linux? The reason it works with *BSD is that the BSD license doesn't exclude linking with anything. You can, quite legally, compile a bunch of random .o files that showed up on your doorstep on a CD-ROM into either kernel. However, even if they came with a license that says "Copyright 2010 Somebody Or'Other. You're allowed to do anything you want with these files and give them to anybody, but I'll never give you the source!" you couldn't legally redistribute your compiled Linux kernel; because the kernel is now linked against non-GPL code, making copies of it would constitute copyright infringement.

On the other hand, with BSD you can quite legally redistribute the resulting kernel binary. Nothing in the BSD license prohibits redistribution due to any linked code, or the lack of source availability, or much of anything else (falsely claiming you wrote the code violates the BSD license, but not much else does). Now, the FreeBSD developers don't *want* to compile bits that are under a more restrictive license into their pure kernel, but just as with Linux, you can compile a loadable kernel module. Unlike with Linux, they can legally redistribute said kernel module, so they do. It may not be part of the core install, but it's on the CD.

Re:while we're here, what about linux zfs

[butlerm]

ZFS was designed as a huge blob of interdependent code

Not true, any more than the claim that Linux is. ZFS has internal layers and an architecture that is as sane as any system out there.

Re:while we're here, what about linux zfs

[Paul Jakma]

you couldn't legally redistribute your compiled Linux kernel; because the kernel is now linked against non-GPL code, making copies of it would constitute copyright infringement.

How so? People have distributed Linux with non-GPL bits many times in the past (AFS, Ubuntu provide the completely proprietary NVidia drivers even). The CDDL ZFS bits clearly wouldn't be infringing on the GPL. If you distribute the glue between ZFS and the Linux VFS As GPL, then that itself doesn't infringe either. We've had a decade or more of kernel hackers being quite happy to tolerate binary modules (even Harald Welte doesn't seem to sue anyone for shipping binary modules, only if they don't supply source to the GPL bits he has copyright in - AFAIK).

So who is going to sue whom exactly, and why?

Re:while we're here, what about linux zfs

[ceswiedler]

Of course it does, but that's different from having RAID or VFS layers which are designed to work with filesystems other than ZFS. The advantage Sun has was that they could basically make those layers interdependent, while Linux needs to worry about keeping the layers independent since other filesystems will need to use them.

Re:while we're here, what about linux zfs

[butlerm]

Including ZFS (or something like it) would not have any affect on the ability of other filesystems to use the Linux LVM / RAID layers. Nothing would stop a ZFS user from using those layers as well, instead of or in addition to the RAID functionality ZFS offers. The ZFS FUSE people do that, for example.

The downside of traditional software RAID is the RAID 5 "write hole", to the degree that I don't see why any sane individual would run software RAID 5 at all. The RAID 5 functionality in the Linux kernel is not particularly useful in real life. ZFS RAID-Z does not have that problem, and is.

Re:while we're here, what about linux zfs

[Big Boss]

Or Linux could implement ZFS the same way Solaris did and have 2 I/O stacks. Insisting on the old layers for a new setup like ZFS is backward when ZFS has clearly proven to be quite capable and has a clean design internally. The old way has good reasons for existing, but the ZFS way does as well. I have a very hard time seeing objections like this as anything other than NIH. I run OpenSolaris at home, and after using ZFS as it was intended, I have no reason to even consider anything else for a large fileserver. And my server is small compared to some I've read about. 12 1.5TB drives in 2 raidz2 arrays.

That said, I run Ubuntu in a VM on the same server for my MythTV needs. Linux still has a lot going for it, but for dealing with large datasets with redundancy and error-checking/correction, ZFS has any other software based solution beat by a long shot. And it's stable, working, with an excellent track record TODAY. Not some time in the distant future like BTRFS.

The licenses are a red-herring, IMO. If NVidia can have a closed source binary blob link to the kernel, I see no reason why an open-source project like ZFS can't do the same thing. License the bridge logic as BSD or CC and it shouldn't be a problem.

I won't comment on hardware RAID as I'm not willing to pay the asking price for the hardware. I'll be happy to compare should someone else wish to send me a RAID card and a set of disks to test.

Re:while we're here, what about linux zfs

[bhtooefr]

IIRC, Ubuntu provides free drivers with the OS, and it prompts to download the closed nVidia drivers.

Re:while we're here, what about linux zfs

[bill_mcgonigle]

the supposedly intractable licensing issues that prevent ZFS from being incorporated into the linux kernel? Is it still hopeless?

There are two problems:

1) CDDL code can't be integrated with GPL code due to license incompatibilities. That's the minor one.
2) Sun(Oracle) has a large patent portfolio on ZFS. You only get patent indemnification from Sun if you use their CDDL code. So, even though ZFS isn't a large code base, and a re-implementation is technically feasible, governments the world over may attack you on behalf of Sun/Oracle if you try it.

But even if you bold ZFS onto linux, which may be nice, the solaris kernel has additional efficiencies that make ZFS a win. I recently built a RAIDZ2 using commodity 7200RPM Seagate drives and it can write 750MB/s, without compression being on. I've never seen anywhere near that performance from a Linux install on the same kind of hardware. So, ZFS on Linux is also less interesting from the performance perspective - unless it's ZFS's "rampant layering violations" that make this performance possible. In that case then there's a real good debate to be had about the cost of those layers.

Re:while we're here, what about linux zfs

[drfireman]

Re: the licensing issues, I was really just asking if there's any chance Sun might consider re-licensing the ZFS code. Last I checked, this was a non-starter, but a lot has changed in the past year or so. That said, I forgot about the "rampant layering violations," and I can see how this would be a permanent problem even if the licensing issues could be worked out. I'm still struggling, though, because I hate maintaining an OpenSolaris box solely for ZFS. At the same time, I love ZFS, and it's going to be a while before I can feel equally comfortable with btrfs.

I'm surprised you've found Linux filesystems such comparatively poor performers, but that's good information to have. I live in a somewhat lower performance neighborhood, and would not have noticed this.

Re:while we're here, what about linux zfs

[bill_mcgonigle]

Yeah, if Oracle dual-licenses ZFS or OpenSolaris then there's all kind of interesting to be waged.

I'm surprised you've found Linux filesystems such comparatively poor performers, but that's good information to have. I live in a somewhat lower performance neighborhood, and would not have noticed this.

Yeah, I was astonished when I built my first server with this kind of setup. I've been building and selling ZFS storage servers since (mostly to back-end linux servers).

Re:while we're here, what about linux zfs

[Paul Jakma]

Yeo, and it is Ubuntu distributing the drivers.

Re:while we're here, what about linux zfs

[Anonymous Coward]

ZFS was designed as a huge blob of interdependent code

Not true, any more than the claim that Linux is. ZFS has internal layers and an architecture that is as sane as any system out there.

More sane actually. Just today, in Linux I had to do the following to get two LUNs joined together with an FS:
. fdisk /dev/sda: partition with Linux LVM
. fdisk /dev/sdb: partition with Linux LVM
. pvcreate /dev/sda1 /dev/sdb1
. vgcreate vgDepot100 /dev/sda1 /dev/sdab2
. lvcreate -n lv1001 vgDepot100
. mkfs.ext2 /dev/mapper/vgDepot100-lv101
. mount /dev/mapper/vgDepot100-lv101 /mymount (and then add it to /etc/fstab)

If the machine was Solaris 10, it would have been:
. zpool create mymount c0t3d0 c0t3d0

Built-in compression available as well, and in recent version of OpenSolaris, dedupe.

Mod parent up

[Anonymous Coward]

The MySQL fiasco(s) combined with getting absorbed in to the giant Oracle collective will hopefully put more spotlight on the more deserving open source SQL platform, PostgreSQL.

computer industry needs more standards...

[h00manist]

However, it didn't detect: the wifi adaptor the ethernet adaptor the sound

If there's one thing that would make the computer industry move ahead faster, it would be more standards. Why on earth can't simple mundane things like ethernet, sound, etc interfaces come with some sort of descriptors or standards which allow at least basic functionalities to be found more easily by an OS? Couldn't chipmakers, driver and OS writers try to save some work for themselves and talk? Every new OS version has to re-create, re-test, etc every driver for every device on the planet. The mere discussion of standards seems to have been killed by the whole 'de facto' notion, which is basically quitting. Even if we exclude MS, there enough active people now to have some debate over some driver and chip detection standards. VMware, linux, xbsd, the livecd scene, motherboard, device, and chipmakers, etc.

Re:computer industry needs more standards...

[Viol8]

Well, soundblaster used to be a de-factor standard for sound boards but that seems to have gone by the wayside and now there are a load of different varieties again. Same with VGA graphics - but then the 3D revolution brought along a slew of different boards that all required different drivers. Its also a mystery to me why at least a common base standard can't be thrashed out for common components but I guess it would be like herding cats with all the vested interests out there.

Re:computer industry needs more standards...

[somersault]

Probably would also get users plugging in their devices and never installing the proper drivers. But open interface standards for basic sound, graphics, networking etc would be truly fexcellent.

Re:computer industry needs more standards...

[Rysc]

This problem was solved--correctly--years ago, and by Sun no less! You want OpenFirmware, though sadly that Intel went all NIH and is pushing EFI for x86 instead, which is similar but not nearly as good.

Re:computer industry needs more standards...

[Anonymous Coward]

VGA is standard. SVGA isn't. The fragmentation started way before 3D revolution. There's a reason why win95 loading screen is in weird 320x400 mode.

Re:computer industry needs more standards...

[h00manist]

The problem is much greater than drivers. You can't even copy a phone contact list to anything else easily. It's just name and phone number, but you need to be pretty skilled to be able to copy it over anywhere. There's little discussion or respect for standards, testing, etc. Data standards, interface standards, software standards, cable standards, power standards, few things works together. USB and Bluetooth were pretty good standard inventions. The complexity of the mix of all things on the market is just completely overwhelming any consumer or tech. The big deal with Apple is basically that they control all their variables, and make things work more simply and easily. Yes, you could do it in Linux, or anything else, but it takes a lot more time and effort.

Re:computer industry needs more standards...

[Anonymous Coward]

Well, soundblaster used to be a de-factor standard for sound boards but that seems to have gone by the wayside and now there are a load of different varieties again.

AC '97? It's an actual standard instead of a de-facto one. Pretty much all on-board audio uses it, and on-board audio has become good enough that buying separate sound cards isn't something anyone should do (hell, most of them are 7.1 these days), unless they're doing something in a special niche. If you're doing something in a special niche, individual drivers for the extra functionality you need makes sense.

Re:computer industry needs more standards...

[Anonymous Coward]

http://www.phatcode.net/res/224/files/html/ch64/64-01.html
The story in that link illustrates the reason they want closed drivers, they think (perhaps correctly) that their competitors will gain advantage by reverse engineering source code, or if an open standard is followed, they're hobbled in certain ways.

Re:computer industry needs more standards...

[bill_mcgonigle]

simple mundane things like ethernet, sound, etc interfaces come with some sort of descriptors or standards which allow at least basic functionalities to be found more easily by an OS?

You're describing USB. Bluetooth adopted the USB driver model for this very reason.

USB3 can help on the low end, but something like USB profiles on PCIe might also be interesting given hardware encapsulation support.

Re:computer industry needs more standards...

[TheRaven64]

VGA is standard. SVGA isn't.

Yes it is.

Re:computer industry needs more standards...

[Anonymous Coward]

Oh, I meant stuff like port 0x3d4 etc. BIOS is essentially a driver.

Opensolaris != Solaris

[Anonymous Coward]

Note that Opensolaris and Solaris are two different things...Opensolaris is to Solaris what Fedora is to Red Hat. Oracle is going to support and invest in Solaris, it's opensolaris what may change.

Does Oracle own Sun yet?

[Viol8]

With all the legal wrangling , especially in europe , I've rather lost the thread of this ongoing buyout.

Re:Does Oracle own Sun yet?

[Paul Jakma]

Yes, since Feb 15th Sun pretty much ceased to exist (in the USA), imlu. Though Sun had already become a wholly-owned subsidiary of Oracle even before that point.

Not likely

[Anonymous Coward]

Even if ZFS was GPL'ed I very much doubt it could displace Btrfs in the Linux land. Not only because the COW-friendly B-trees of Btrfs look more clean, but because ZFS is not just a filesystem and would require a lot of work. ZFS is a complete reimplementation of everything between the VFS layer and the disk driver, including cache management. Solaris has two IO stacks living together, the old one (UFS, FAT, etc) and the ZFS one. I doubt the Linux hackers would accept something like that in Linux, they would probably require to drop everthing that it's not the filesystem (if Sun wasn't able to make UFS work with the ZFS block subsystem I doubt you can adapt it to work well with the myriad of filesystems that Linux supports). Btrfs in the other hand it's designed to fit in Linux perfectly, and it's already being used by early adopters anyway (I've been using it for 4 months on my desktop with no problems at all)

Re:Not likely

[0100010001010011]

ZFS is available now. Running numerous places including my own home network.

btrfs still has a ton of "EXPERIMENTAL." "DON'T USE FOR ANYTHING IMPORTANT" warnings everywhere.

That right there clinched it.

Backspace in the Opensolaris/x86 console pls

[crispi]

solaris console login: ^H

Grrrr.

Re:Backspace in the Opensolaris/x86 console pls

[bhtooefr]

IIRC, it's just as bad on SPARC - ^?/delete is backspace in the console, ^H is backspace in the GUI.

Of course, this wasn't as insane on the Type 4 keyboard, where delete was directly above backspace, and the same size, in the function key row. But Type 5 broke that.

Anyone actually read the article at Oracle yet?

[Anonymous Coward]

Those, who are crying here "OpenSolaris gone", read the fucking article CAREFULLY (never happens on Slashdot, though):

So use letter-by-letter approach if you're unable to see word-by-word or sentence-by-sentence:"Future releases of the Solaris OS will also be based on the OpenSolaris community codebase."

That means RedHat/Fedora model. Clear now?

http://www.sun.com/software/solaris/lifecycle.xml

What is the reason for abandon fashion lately?

[Ilgaz]

When will open source folks understand that older version support, especially for server oriented things is a big deal?

There is a company who makes living with OS upgrades/sales and they still release updates for Windows XP you know. An OS from 2003 or something.

Right, they don't release directx 11 for XP but at least their consumers (and IT guys) don't feel abandoned in sense of security updates.

Same mistake is being done almost monthly in open source scene and they wonder why companies choose a $2K price instead of their "free" product.

Re:What is the reason for abandon fashion lately?

[TheLink]

> Windows XP you know. An OS from 2003 or something.

From 2001 (oct 2001 retail release).

OpenSolaris is "dead man walking".

Re:What is the reason for abandon fashion lately?

[Anonymous Coward]

> When will open source folks understand that older version support, especially for server oriented things is a big deal?

LOL
Debian gets it perfectly. So "open source folks" is too broad a definition.

Re:What is the reason for abandon fashion lately?

[Dare nMc]

At least opensolaris is technically supportable. Sounds like Oracle is laying off the best developers. So you have the source, and you can hire the developer... When closed source does the same thing, you can hire the developer, but he could only start from scratch (legally.)
Granted this is pricey support, but you likely can't get better support than that IMHO.

Re:What is the reason for abandon fashion lately?

[Ltap]

No. There is plenty of support for older stuff - look at Apache 1.

You're saying "oh well, open-source devs should support users who update once every ten years." Why? If you have obsolete software, you suffer consequences like additional exploits and fewer features. This is a bad thing, and the companies' idea that they can just install something and then leave it written in stone is bad, and lazy, and should stop.

Re:What is the reason for abandon fashion lately?

[ToasterMonkey]

> When will open source folks understand that older version support, especially for server oriented things is a big deal?

LOL
Debian gets it perfectly. So "open source folks" is too broad a definition.

That's not fair. The F/OSS spectrum is composed of free software, open source software, and Linux software. Almost everything derogatory said of free/open source software can be blamed on last one, which Debian is not a part of. It's only confusing because Debian uses the Linux kernel.

I mean, if the idea of taking the Linux out of your "Linux system" and replacing it with something else bothers you emotionally, you are not really in the first two camps; you are a freetard.

This is coming from a former freetard.

Wonder if someone can help me here

[sleeponthemic]

Is there a way I can filter all KDawson articles out? Rather than endlessly whine about it, I'm looking for a way to return slashdot to the way it used to be. That is, with some integrity. I think I've read three piece of shit, antagonistically misleading articles posted by this bastion of all that is wrong with journalism today alone.

Re:Wonder if someone can help me here

[Ant P.]

Um, did you even try looking?

Re:Wonder if someone can help me here

[nomadic]

slashdot to the way it used to be. That is, with some integrity

Man I wish I had some mod points to mod you up funny.

IBM & AIX - the last man standing

[Colin Smith]

Nobody cares about open solaris. Nobody in their right mind would have chosen it as a platform.

I'm not surprised that IBM is the last company, AIX the last proprietary unix platform. Power the last proprietary hardware platform...

HP & Itanium? Laughable... And Linux on x86 has eaten the rest.

IBM 'get' services in the way the rest never have. They get that it's the bloody hardware which matters. This is why power is hitting 5GHz. The OS is just there to make it work. You want the fastest, lowest latency, highest throughput. You use IBM. You just want it to work and are on a budget? Linux.

The 'executives' of the rest of the companies clearly didn't know or care what their customers want, or what their business really is.

Re:IBM & AIX - the last man standing

[BitZtream]

Nobody cares about open solaris. Nobody in their right mind would have chosen it as a platform.

Ah, the beauty and ignorance of youth and inexperience. In short there are plenty of places where Solaris fits perfectly and Linux doesn't. I can show you several data centers with more Solaris running on Sun hardware with than Linux on any hardware, and they like Linux, its simply that it doesn't fit the bill.

Guess what, Solaris doesn't fit every situation either, which is why they use Linux too, and Windows, and some IBM gear.

Power the last proprietary hardware platform...

You do realize the x86 and x64 architectures are proprietary right? You do realize that Intel just happens to license the rights to use it to people like AMD and nVidia ... right? You realize they've been doing everything they can to back out of those licensing agreements so they become the sole producer of x86 chips again ... RIGHT?

Re:IBM & AIX - the last man standing

[afabbro]

Nobody cares about open solaris. Nobody in their right mind would have chosen it as a platform.

Ah, the beauty and ignorance of youth and inexperience. In short there are plenty of places where Solaris fits perfectly and Linux doesn't. I can show you several data centers with more Solaris running on Sun hardware with than Linux on any hardware, and they like Linux, its simply that it doesn't fit the bill.

Ah, the beauty and ignorance of youth and illiteracy. Read what you quoted again - he said Open Solaris. Still have those several data centers to show us?

Re:IBM & AIX - the last man standing

[Anonymous Coward]

Highest throughput? That would be UltraSPARC Niagara. It is built for throughput. On Siebel v8 benchmarks, you need 28 of the 5GHz POWER6 to match 4 of the 1.4GHz Niagara. The POWER7 is in general 3x times faster than POWER6 (sometimes slower, sometimes faster). So you need 9 of the high clocked POWER7 to match 4 of the 1.4GHz Niagara. POWER is not very efficient, it needs high Hz, similar to Pentium 4. You need four 5GHz POWER6 to match two 2.93GHz Intel Nehalem on official TPC-C benchmarks. Intel Nehalem-EX will be comparable to POWER7 in terms of performance, but for a much lower price. Frankly I dont see what expensive POWER7 offers compared to much cheaper x86 CPUs? 12-core AMD bulldozer, 8-core Intel Nehalem-EX, etc. I expect POWER to diminish in favour of new generation x86 CPUs. OTOH, Niagara has extreme throughput, it is many times faster than POWER7 and Nehalem-EX.

Re:IBM & AIX - the last man standing

[Colin Smith]

Ah, the beauty and ignorance of youth and inexperience. In short there are plenty of places where Solaris fits perfectly

I've run Solaris datacenters, on Sun systems. That's fine, it makes sense. You wanted the Sun hardware, they had some nice kit. But Solaris is just there to make it go. OpenSolaris. Solaris on x86 etc make absolutely NO sense at all. Nobody cares.

so they become the sole producer of x86 chips again

If not AMD, then VIA or Transmeta or Nvidia, or whomever. Anyone with the knowledge and fabs can make an x86 instruction set CPU. There is no such thing as an x86 license, it's a bundle of patents on specific implementations. You know that... right? RIGHT? Avoid those particular implementations and you don't need to license any patents from Intel.

 

Re:IBM & AIX - the last man standing

[Colin Smith]

You're conflating throughput with speed. Speed is latency (how fast). Thoughput is how many.

You need four 5GHz POWER6 to match two 2.93GHz Intel Nehalem on official TPC-C benchmarks

What are you talking about? TPC-C benchmarks entire systems not cpus. Mostly I/O bound... RAM, network and disk.

http://www.tpc.org/tpcc/results/tpcc_perf_results.asp

The IBM box is one machine with 32CPUs 64 cores. The Sun, a cluster of what? 12 systems? with 4 CPUs each, 384 cores. See what I mean by IBM getting it?

I think you're right though. x86 will eventually eat power too, simply because the market for "ultimate performance" is much smaller than "good enough".

Re:IBM & AIX - the last man standing

[ishobo]

Gee, I feel the same way about Linux

Re:IBM & AIX - the last man standing

[Anonymous Coward]

Glad to see someone else here understands the modern day enterprise market. Of course POWER hit 5Ghz because they ripped out out-of-order execution from power6 and in doing so killed a lot of the real-world performance improvements from it. Looking like POWER7 is fixing that though (of course clock speed is going down again).

The reality is that x86 in 2010 looks very different from x86 in 2000. NUMA, on-die memory controllers, 12-core processors, etc etc. x86 is pretty viable for "large" workloads these days even though most arm-chair server engineers still think you need a slow-as-shit SPARC box for the 'highend'.

Re:IBM & AIX - the last man standing

[jwhitener]

Nobody cares about open solaris. Nobody in their right mind would have chosen it as a platform.

I'm not surprised that IBM is the last company, AIX the last proprietary unix platform. Power the last proprietary hardware platform...

HP & Itanium? Laughable... And Linux on x86 has eaten the rest.

IBM 'get' services in the way the rest never have. They get that it's the bloody hardware which matters. This is why power is hitting 5GHz. The OS is just there to make it work. You want the fastest, lowest latency, highest throughput. You use IBM. You just want it to work and are on a budget? Linux.

The 'executives' of the rest of the companies clearly didn't know or care what their customers want, or what their business really is.

Why is this insightful? And it is especially not insightful after a merger with Oracle, who has declared they will increase support and development for Solaris.

Solaris is currently the most common platform for Oracle installations in higher ed. and many other business types. It runs on both very powerful proprietary cpus (Ultrasparcs) and on commodity x86 hardware. It contains support for multiple variations of virtualization, ZFS (learn a bit more about the power of ZFS if you think it is a minor advantage), and can run practically any software that linux or other unix systems can. See sunfreeware.com

In addition to the added power of Oracle's backing and software lineup, Solaris has a complete stack of solutions for the enterprise. Ldap, email, identity management, etc.. All of which will run on other platforms, but in my experience, run best on Solaris.

Google around: http://news.netcraft.com/archives/2004/11/16/solaris_remains_top_choice_among_fortune_100.html

Solaris is everywhere. Just because you may not be in areas of business that use it, declaring "that nobody in their right mind" would use it is just silly.

It is a culture problem

[Ilgaz]

Interestingly, all the issues you have exists on a typical Desktop/Laptop. Especially sound card and Wi-Fi. I guess the issue with OpenSolaris is the company and its culture. Sun is a company who makes gigantic servers having insane amounts of uptime and most of their products (except couple of workstations) doesn't even have the parts like sound card or wifi.

A good example is Java, for a decade, people using Java plugin had to deal with their hard disk going nuts right after running a basic applet. What did they (finally!) do? A simple, 1 MB application running in low priority that caches most used classes. Problem instantly got fixed. Same guys, while fixing that issue, had the marvelous idea of adding something to startup to check for java updates, running 24/7. That is PR suicide on Windows land, that is one thing users hate more than a virus. Of course, they are disconnected from average desktop user so they thought it won't bother. It didn't change the mind of thousands to flame them. They could, use Apple's method of using system's own scheduler on Windows (for software update) and get away with it.

I really think it is a culture problem for Sun, they should really get rid of "lets go big on desktop" mad idea and fix their already problematic products like Desktop Java, Open Office. They could start with taking over OS X Java development from Apple, Apple clearly doesn't care and doesn't bother at all. Open Office? They managed to copy MS Office with all its problems in open source. I remember what a great thing it was while it was Star Office. It is like, they code it in a way that everyone has a Solaris Workstation with 4GB of RAM.

Re:It is a culture problem

[Digital Pizza]

If Microsoft wasn't hostile to Java then maybe Sun wouldn't have needed to do those stupid shenanigans to get Java to run halfway decently on Windows; it could even have been included with Windows (not the bastardized version Microsoft made, but the true, licensed Sun version).

But I don't really blame Microsoft too much for this; I blame Sun for declaring Java a "Windows Killer" from the get-go. The idea was that you'd run all your apps in Java in a web browser, and this would "Kill Windows". Yeah, declaring war on Microsoft was REALLY smart. I can't really blame MS for not going along with them on that. Probably just another example of Scott McNealy running his mouth.

(Of course it didn't help that Java was slow as hell back in the beginning; it isn't now, but the damage to its reputation is done).

Sun/Oracle and Sun 7000 Storage w/OpenSolaris

[slieberg]

One thing that is interesting is that Oracle has stated that they are committed to the Sun Open Storage Line(Sun 7000 Line) --> http://searchstorage.techtarget.com/news/article/0,289142,sid5_gci1379858,00.html and currently that line is built on OpenSolaris. So it will be interesting to see what's going to happen. I had heard they will be refreshing some of their models in the next couple months. I just put in an order for a Sun 7110 actually.

Drivers are available by download

[Anonymous Coward]

Solaris and OpenSolaris have a well defined device driver interface, sample and real driver source code, and a stable ABI, so third-party ethernet and WiFi drivers are available for the common network devices that Sun themselves don't yet support:

http://homepage2.nifty.com/mrym3/taiyodo/eng/
http://www.lmgtfy.com/?q=solaris+ethernet+drivers

Put the downloaded drivers on a flash drive or your hard disk.

Pure FUD

[Anonymous Coward]

This is why I hate the Slashdot community.
Half retarded FUD posts like this spread like wildfire over the web.

Is there paid for support/software updates for Fedora? No you get your 6 months and then on to the next release. How much of stretch is to assume that the same will be true of OpenSolaris? OpenSolaris will have a similar relationship with Solaris 11 as Fedora has with RHEL.

So quit crying wolf. It makes you all look like terribly uninformed scaremongers.

Mission critical servers?

[Anonymous Coward]

Where exactly are those servers?

And what is the critical mission?

There are tons of Linux servers running mission critical stuff (like business applications worth millions per minute if the servers fail) but I have not seen a single one that does not have the support of either RedHat or in a smaller scale, Novell for SUSE.

Unsupported Linux servers may be running important stuff, but if you are betting your bacon on them without somebody that has a business commitment with you to keep the servers working, then you are a fool man indeed.

Time for everybody to gwow up.

[BrokenHalo]

On the contrary... if you want zfs, and stability under load, there's nothing wrong with it.

I wouldn't disagree, but there are usually alternatives. I would personally never make a choice of OS on the availability of zfs (though I have no doubt the filesystem is quite nice). And stability under load can be had with any Linux or BSD with an appropriate configuration. So yes, there's nothing wrong with OpenSolaris, but then there's nothing so right with it that we can't afford to do without it.

I know there's a resistance among the various *Solaris/SunOS communities about the other UNIX variants, but it's time for these people to get over themselves. The world has moved on, and the alternatives are plenty mature enough to cope, and whining about them just makes these people look like craniorectal adolescents.

Re:Time for everybody to gwow up.

[buchner.johannes]

Linux is for the desktop, Solaris is for servers. You can't use Linux on servers, because it has too many bugs.

At least that's what this NASA administrator said in 2006: http://searchenterpriselinux.techtarget.com/news/article/0,289142,sid39_gci1157924,00.html

Re:Time for everybody to gwow up.

[h4rr4r]

1. he was wrong then
2. 4 years later he is even more wrong.

Re:Time for everybody to gwow up.

[Anonymous Coward]

He is wrong because Linux is not for the desktop

Re:Time for everybody to gwow up.

[Wovel]

Tomorrow he will be more wrong than today and yesterday.

And this is why I no longer push Solaris

[WindBourne]

Years ago, I use to support Sun and their OSS efforts. First time that they opened Solaris, I was all over it. The second time, I said that I did not fully trust sun. Pending what really happens, it proves exactly how I feel about code that comes from companies like this.

Solaris is not gone, neither is SPARC.

[Anonymous Coward]

Oracle also gets services.

They are selling the full integratade package, the layers they were missing where Virtualisation/OS and hardware. With Sun now they have that.

Yesterday Oracle presented in London part of their strategy (in the Royal Oper House, nice venue, great food, did I win the iPad?) and both Sun hardware/virtualization and Solaris are there.

Right now Oracle is making available virtual machines in a similar fashion to what is possible with VMware. Even LDOMs have been renamed to OracleVM . And you can download them for free without paying any licenses.

If Oracle is planning to ditch Solaris and SPARC then they have to explain why they had Sun hadware in central stage as the solution for "private clouds" they were parading all day yesterday during their first main presentation in Europe since they got Sun.....

WHat are you talking about?

[Anonymous Coward]

Pretty much all the GNU chain is availilable.

Sun itslef put GNOME in place as the dsktop envirnoment.

Apache, perl, MySql, PHP.

What exactly are you missing?

Business focus.

[Anonymous Coward]

Yesterday in Oracle's Cloud Computing presentation in London you saw a clear example of this.

Most people were Oracle, there was a chap from Sun.

While the Sun chap provided tons of technical data (boring us all to death), the Oracle guys provided a vision.

You may think this is glib, but everybody understood what Oracle wants to achieve and you may or may not buy into it, but at least are not left scratching your head regarding their strategy.

OpenSolaris and "cutting-edgeness"

[SIGBUS]

I recently set up an OpenSolaris ZFS file storage server on my home network, and am using the latest development snapshots as they come out. Development has not stopped on OpenSolaris, by the way. At any rate, I think the Fedora vs. RHEL comparison is a pretty good summary - though we don't have a Solaris equivalent of CentOS.

One very cool thing in OpenSolaris is the "boot environment" capability; it uses the ZFS cloning feature to create what is in effect a separate installation when you upgrade from one version to the next. If the latest update doesn't work right, you can boot back into your previous setup without having to try to downgrade your installed packages.

IMO, that makes keeping up with the bleeding edge a lot less risky. I gave up on Fedora years ago when I had a "yum update" blow up in my face one too many times.

The downside: Hardware support is a bit limited. OpenSolaris doesn't have support for HighPoint controller cards, which work fine in Linux and FreeBSD.

Re:OpenSolaris and "cutting-edgeness"

[perlchild]

On the CentOS equivalent front...

Wouldn't Nexenta be a possible close match?

Re:Anonymous Coward

[uassholes]

Must be someone who thinks MSWin is an OS. But I think his post should have been modded funny rather than troll.

I wish Oracle would GPL ZFS

[melted]

And some other Solaris tech. Heck, they should GPL the whole thing and get a group of engineers to port the juiciest morsels of it to Linux. That way, Solaris going away would be much less of a loss, and Linus would be a happy man (he said, half-jokingly, he wants Solaris to die :-).

Re:I wish Oracle would GPL ZFS

[Hurricane78]

Well, in case the whole thing would be GPLed, nothing would stop anyone from freely mixing things.

A Solaris kernel with a GNU user space. A Linux kernel on an OpenSolaris user space. Why not?
After all, that’s the biggest beauty of Linux: The freedom to get, whatever fits you personally like a glove!

(And that’s why Gnome and KDE (and XFCE) should work more on their integration [via FreeDesktop].)

Re:I wish Oracle would GPL ZFS

[SolitaryMan]

I'm afraid that "GPL something" is the last thing you can expect from Oracle.

Re:I wish Oracle would GPL ZFS

[RAMMS EIN]

I wouldn't be too surprised, actually. Oracle pushes Linux a lot. Now that they own ZFS, they could actually contribute that to Linux. Having said that, I am not sure how much of a benefit that would be to them - how much does their database depend on the characteristics of the filesystem?

Re:I wish Oracle would GPL ZFS

[SolitaryMan]

They don't need to GPL ZFS in order to be able to use it with Linux. They can just distribute the driver in binary form or, which is more likely scenario IMO, license it for use only with *their own* Linux distro and/or DB.

Will that help AMD survive?

[newdsfornerds]

I sure hope it will. It really would suck if AMD died leaving Intel as supreme dictator. These days it would also mean that the government would be forced to bail out Intel if ever Intel got into financial trouble.

Whither Nexenta?

[newdsfornerds]

What happens to projects that depend on Open Solaris?

Solaris and SPARC: standing tall

[Anonymous Coward]

Nobody cares about open solaris. Nobody in their right mind would have chosen it as a platform.

I'm not surprised that IBM is the last company, AIX the last proprietary unix platform. Power the last proprietary hardware platform...

SPARC, the last open hardware platform. Many chips are under the GPL, and you can license yourself to manufacture chips. Ellison has publicly stated they're investing more R&D into the platofrm.

OpenSolaris is like Fedora: a test bed for new ideas that will eventually become the next release of the commercial OS. It just happens that official support is available (unlike Fedora or OpenSuSE), and that is what's up-in-the-air.

Re:Solaris and SPARC: standing tall

[Colin Smith]

Ellison has publicly stated

Meh. Ellison is redundant. Sun is dead, with it almost certainly goes SPARC. Bit of a shame really.

IBM have an integrated business services division which understands their customers. Sun never did. Maybe Oracle/Sun can work together and make a real business support company competitive with IBM. Frankly, I doubt it. Both organisations are barely functional. It's far far more likely that Oracle will shut down Sun and sell anything valuable to someone who understands hardware better.

 

A chance to get ZFS on Linux?

[Hurricane78]

Do we then at least get ZFS (The fast original implementation. Not the slow and huge FUSE one) for Linux?

Or will it end like with Lotus SmartSuite (best. office. suite. ever. full stop.), and die, never to be opened? :/

Re:A chance to get ZFS on Linux?

[bhtooefr]

Technically, it's BSD-compatible open source.

Or, you could include the ZFS source code and the Linux kernel source, and build a Linux+ZFS on installation of the OS, I guess. Problem is, you'll have to have a scratch disk to do this on, and then boot off of that scratch disk to partition and install. Or, use FUSE. Or, use a BSD for the installer.

FUD about Re:FUD

[stilldead]

FUD about Re:FUD. While you are both right in theory ( http://support.microsoft.com/gp/lifean31 ) If you saw the weekly official patch cycle for WinXP from Microsoft you would feel mostly like... someone who had a crappy life testing and installing all of those patches.

mdb

[mzs]

Everyone mentions ZFS and Dtrace, I will really miss mdb if solaris ever goes away. mdb is the single best debugger ever created. I have shell scripts that use truss to stop on an bug in closed source code, then use mdb to change some registers, and then use prun to carry-on. That completely ignores all the useful dcmds too.

Dan Dobberpuhl, the Alpha lead works for PASemi

[tlambert]

Dan Dobberpuhl, the Alpha lead designer works for PASemi, which is o say Apple, now.

-- Terry

Lost opportunity

[recharged95]

I wonder what will happen to Nexenta OS? I know the effort is currently low-key, but OpenSolaris, ZFS, Debian pkgs.... sounded like a best of both worlds.

live-lock under Linux; not under Solaris

[Anonymous Coward]

On the contrary... if you want zfs, and stability under load, there's nothing wrong with it.

I wouldn't disagree, but there are usually alternatives. I would personally never make a choice of OS on the availability of zfs (though I have no doubt the filesystem is quite nice). And stability under load can be had with any Linux or BSD with an appropriate configuration. So yes, there's nothing wrong with OpenSolaris, but then there's nothing so right with it that we can't afford to do without it.
 

Availability of ZFS is a major win, as it makes many things so much easier as an admin.

As for load, I've never had Solaris live-lock on me, but in just the last year I've had two systems do that under Linux. And let's not get started on the OOM on my Perforce servers at least once a year (thankfully we're upgrading to new hardware next quarter).

Linux is good, but I've run into a lot of edge cases on it. Throw in the lack of zones, and I'm now preferring Solaris 10 when I can get it.

Now if it only had the Ports tree (and yes, I know about pkgsrc).

Move to FBSD?

[ducomputergeek]

FreeBSD fully supports ZFS and DTRACE in version 8. Version 7 ZFS was still listed as experimental. I know we looked at OpenSolaris, but once those features were in FreeBSD, we decided to stick with what we had. (That being said, we're still running FBSD 6.x on the production machines)

Oracle does not canning OpenSolaris!!

[hotfireball]

Anyone has been read the thing at Oracle? They say they will release their Solaris on top of the codebase of OpenSolaris. In other words, commercial Solaris from Oracle is a same as an OpenSolaris++. Oracle will add some proprietary features to Solaris that will be enabled only on their exclusive hardware.

In other words, this is Fedora/RHEL or OpenSuSE/SLES model. You like generic OS and you like to piss with it yourself, wasting a load of time, then go ahed, get OpenSolaris and GA support, if you like. But if you want advanced stuff and you have no time to waste a time for the cheap mess, then get Oracle Solaris for pay, get Oracle hardware and that's it.

So that's basically a message. Which is very good: it will actually push generic OpenSolaris to be up to date and financed by Oracle.

Oracle is the primary driving force behind btrfs

[melted]

Oracle is the primary driving force behind btrfs, which is a ZFS-like next generation FS for Linux. Now that they own ZFS, continuing development of btrfs makes far less sense.