OpenSolaris Governing Board Closing Shop?

echolinux writes "Frustrated by Oracle's refusal to interact with the OpenSolaris community or speak with the OpenSolaris Governing Board, the OGB has issued an ultimatum to Oracle: designate a liaison to the OGB by August 16th or the board will 'take action at the August 23 meeting to trigger the clause in the OGB charter that will return control of the community to Oracle.'"

Comments

Sad

[0racle]

Oracle seems determined to destroy everything they acquired from Sun. We had 2 OpenSolaris machines since Zones and ZFS are just hot shit and several SunFire servers. We're moving the OpenSolaris installs to FreeBSD and are probably going to be looking at HP or IBM machines in the future.

Re:Sad

[allcar]

It is sad. I find it really depressing to find the Oracle logos all over the Sun site and Java downloads. I guess that Sun were just too nice a company to prosper in the cut throat world of modern IT.

Re:Sad

[bleh-of-the-huns]

Oddly enough, I went the other way around about 6 months ago because I wanted a better implementation of ZFS and iSCSI. Freebsd's implementation is far behind OSOL, and iscsi is not even close to being functional in 8. iscsi works in 7, but zfs does not.

It's unfortunate, but I see OSOL dieing a slow horrible death at the hands of Oracle.

Re:Sad

[0racle]

Oh, we unfortunately are not using ZFS on FreeBSD because it's not as stable, but it was the Jails that pushed us to FreeBSD instead of Linux which is what runs most of our other systems.

Re:Sad

[Third Position]

This is not exactly a surprise. Oracle has a deep and abiding interest in Oracle's bottom line. How does Open Solaris contribute to that? It doesn't, hence Oracle losing interest fast.

As painful as it may be to acknowledge, this is actually a rational approach. Look no further than the fact that Oracle ended up eating Sun, not the other way around. I like Free Stuff as much as the next guy, but that doesn't change the fact that if you're in business to make money, you'd damn well better focus on things that make you money.

Re:Sad

[Darkness404]

Well developed open projects however allow for greater mindshare leading to more people using their commercial offerings. Look at Red Hat, because RHL was well used on people's personal desktops, it made sense for them to push a company towards Red Hat's commercial products. Same thing with Ubuntu, because many people who use Linux are comfortable with Ubuntu, when a small business looks to consider Linux, Ubuntu is their first choice. Solaris has a lot of features that could be very handy for businesses, but without experience, most tech people are going to recommend BSD or Linux because it is what they have worked with.

Support a community well and it will pay you back. Alienate a community and you are suddenly competing against better entrenched products.

Re:Sad

[Mark Round]

They’ve completely alienated and scared off the community around OpenSolaris, killed any lines of communication by clamping down on employee blogs and ignored open letters from highly influential and important community members begging for *any* kind of information. They’ve forbidden Sun/Oracle employees from heading up the Solaris user groups and booted the meetings out of their buildings; turned Solaris 10 into a 30-day trial, and pushed back the 2010.x release of OpenSolaris with no word as to it’s planned release date, or even if it is being continued as a product.

Oracle are doing a superb job of killing Solaris - at least, as we knew it to be.

Oracle just really doesn't care about Solaris as a general purpose OS (there's no money in it), and it makes sense although I personally find it tragic. It's probably why they're also killing all their OEM deals. I strongly suspect Oracle's overall aim is to have Solaris relegated to the role of running as the bottom layer in an Oracle "database machine" or Java appserver bundle.

It excels in these tasks, and it would obviously fit into Oracle's stated goal of being a one stop shop, where if you want to run Oracle, they'll sell you the bundle - hardware, storage, OS and software. If they no longer want it to be a dominant general purpose datacenter OS, then their approach makes sense. They don't need a "community" around the product, they don't need open source developers porting applications to it, and they certainly don't need the overhead of running and managing a community portal anymore.

I think the way they are going about it reprehensible, and it's a tragic end for such a historic and innovative OS but you can see why. Larry is all about the $$$, and Sun's approach just wasn't bringing in the big bucks.

Re:Sad

[0racle]

OpenSolaris as the basis to Solaris 11, or simply dropping Solaris 10 and going with OpenSolaris as the primary OS would have brought a more modern environment and significantly improved package management and patching while still maintaining the expected stability from Solaris. Oracle seems to prefer to keep Solaris archaic. They just killed their best beta platform.

Oracle got rid of the free to download and use Solaris 10 as well. Sun moved that way to entice developers to develop and test on their platform. Oracle, instead of continuing that to keep developers, moved Solaris back to a pay only. Why would anyone pay to develop or test on Solaris when the competitors are free and just as good? Who is going to buy Solaris when the only thing tested on it is Oracle Database? Oracle is shooting itself in the foot.

Re:Sad

[LWATCDR]

Oracle could leverage Open Solaris as the ideal Oracle platform.
They could push for high end web solutions to use Oracle+Solaris+Java.
So yes it could be a really good solution for them. Also if they can get enough people using OpenSolaris they will have a well trained pool of sysadmins. OpenSolaris could compete with RHEL and Sun/Oracle can sell support like Red Hat.

Re:Sad

[Mark Round]

Not if they don't want or need a community around their product. I posted about this above, but I suspect Oracle are of the opinion that there's no money to be made selling Solaris as a general purpose OS. They probably just want it as the bottom layer in a hardware & software bundle, tuned to running Oracle or Java workloads.

Re:Sad

[DesScorp]

Oracle seems determined to destroy everything they acquired from Sun. We had 2 OpenSolaris machines since Zones and ZFS are just hot shit and several SunFire servers. We're moving the OpenSolaris installs to FreeBSD and are probably going to be looking at HP or IBM machines in the future.

And just how much money did Oracle make from your two OpenSolaris machines? If you bought them pre-merger, they made nada. Zip.

Oracle doesn't care about you unless you're willing to spend a lot of money. That's not bad, that's just their business model. They're trying to become the Mercedes-Benz of IT. They're going mostly to the high end of the enterprise. They expect to be well paid for both hardware and software. And I suspect Sun will start making more money on the high end if they do it right. I hear people moaning at Slashdot all the time about how Oracle doesn't care about the little guy, the small shop. Well, you know what? You're right, they don't. You're not their target. You don't have enough money to interest them. So it's best for all involved that you DO move your installs to BSD. So Oracle can concentrate on the big paying customers, and you can concentrate on your shop that doesn't pay for operating systems.

Re:Sad

[hitmark]

when the market goes down, the companies goes vertical.

heck, it may well be that oracle wont sell a stack, but rather lease it; with some kind of yearly support contract.

Re:Sad

[codepunk]

They could but it would compete with their Linux offering thus adding no value vs development and maintenance costs.

Re:Sad

[Timex]

Support a community well and it will pay you back. Alienate a community and you are suddenly competing against better entrenched products.

If you'll pardon the dated reference, this isn't the first time something like this has happened. One case that comes to mind is Apple's ending the life of the Apple II line. Sure, it would have been a virtual nightmare to keep backward compatibility as they moved forward with the series, but because of the way they went about it, many of their big fans jumped ship to the PC-compatible camp, rather than shifting to the Mac. Apple could have had a larger following with their Mac line, had they tried to make the change a little more gently, but they didn't, and they are only recently beginning to recover from it.

If Oracle is careful, they won't make any waves in doing what they think is the best action to take, but somehow I get the feeling that they're past caring what anyone else thinks.

Re:Sad

[Doc Hopper]

Oracle could leverage Open Solaris as the ideal Oracle platform.
They could push for high end web solutions to use Oracle+Solaris+Java.

Actually, Oracle DOES leverage OpenSolaris as an Oracle platform. The 7410 storage platform exclusively runs OpenSolaris under the hood. Bog-standard Solaris wasn't up to the job. We've bought a number of these storage platforms and are testing them out right now; other than annoying production delays due to unavailability of really-honking-big SSDs, they are extremely cool and high-performance storage solutions.

Also the newer T5240 boxes run way better on OpenSolaris than on stock Solaris 10. No ifs, ands, or buts. Better hardware support and faster I/O. You have to be running the 10/09 release of Solaris 10 to even support these boxes at all, and OpenSolaris supported them before they were even released.

Re:Sad

[Hylandr]

When I was on the Oracle / Solaris website recently I noticed the environment was very old guard. Hailing to a day when Unix knowledge was a rare and expensive commodity.

You could see it in the language that was used, 60's business rhetoric regurgitated for today's business masses that might still buy it. Now, I am looking for contracts that are migrating away from Solaris. Going to try and catch this wave early.

- Dan.

Re:Sad

[Doc Hopper]

Oracle doesn't care about you unless you're willing to spend a lot of money.

Correction: Oracle wants you to spend a lot of money, but they care about you as a potential paying customer. For instance, you can pick up a two-user license of Oracle Database for free and run a large production web site on it if you want. Think about your typical MySQL deployment: One user for the web site, and maybe a second user for the administrative user (usually "root"). Oracle gives this away for free for unlimited use.

The goal is to eventually rope you into a larger deployment with more capabilities so you become a paying customer. Always has been.

But I can assure you the sales guys "care" to get your money even if you're only spending a little bit of it. And Oracle spends a ton of money on trial programs, free software (Oracle Enterprise Linux, for instance), and other promotions to eventually drive revenue.

Re:Sad

[qwijibo]

OpenSolaris provides insight into the directions Solaris is going. I've used my experience with OpenSolaris at home to discuss possible future directions at work once the features work into mainstream Solaris. OpenSolaris is to Solaris what Fedora is to RedHat Enterprise Linux. Giving your customers a chance to preview what's up and coming gives them an opportunity to suggest a direction before it's in the mainstream release. Sure, a lot of deadbeats benefit from making the OS freely available, but getting contributions from the unpaid community and giving your customers reasons to promote your products should be a good enough benefit to justify the business case to keep it going.

I work for a hosting group in a large financial services company. We have over a thousand Linux and Solaris systems we support, with more being added all the time. We have numerous internal groups that need large Oracle databases on Solaris and we're happy to provide that. The people who write standards for the company are telling us to move away from Solaris and either move to Oracle on Linux, or for large databases, go to DB2 on AIX. When we're being told to drop Solaris, we really need something compelling to argue in favor of keeping it around. That is going to be much harder with Oracle's tight lipped approach to letting customers know what's coming up.

At home, I run OpenSolaris using ZFS for all of my storage. I run Ubuntu under VirtualBox to get all of the benefits of ZFS on the hardware with the more user friendly features of Ubuntu as a desktop. This is working out great for my whole family. Unfortunately, it looks like I'm going to have to look for another OS for the bottom of the stack since Oracle appears to be dropping it entirely.

OpenSolaris may not be a direct money maker for Oracle, but it has a very real contribution to their bottom line. With an enterprise database, you can get away with being secretive about everything. With an operating system that is much more widely used, that approach is not wise. Unfortunately, Oracle probably won't realize that until they've permanently lost several large customers.

Re:Sad

[jackspenn]

I guess that Sun were just too nice a company to prosper

Well that and they were way over priced ...

Well that and the Open Source Community caught up with them ...

Well that and they didn't have a long-term solution/strategy to ensure new entries into the tech field could gain experience/skills on their products so they would be comfortable recommending them. Sun relied on the old guard to recommend Sun, while newer entries onto the computer field were more comfortable recommending solutions they had experience with.

As a result apache replaced Sun's web server as the standard.

Red Hat (and others) took away Solaris server market share.

New startups began by running Oracle and other databases on Linux (or even Windows) servers in the initial low funding development stages and then when it came time to go into production, some of them didn't bother with moving to Sun hardware and Solaris, and instead remained with what worked and building it out to be "good-enough" for less money and less headaches.

I was in college from 1995-1999. The guys who loved going to the lab became Solaris die-hards, because that was what the school at that time ran (it is now LINUX, LINUX and more LINUX). But I preferred working in my apartment, so when I had took C, LISP, and JAVA classes that were focused on the fundamentals of code, things like recursion or objects, my teachers didn't demand I used the Solaris workstation, just that I solved the problem and got a strong foundation. So I installed Red hat on a backup PC and worked by using the same languages, with the same libraries, with the same text editors only on LINUX as the labs used on Solaris. At the time, I was the minority, but with each new class the LINUX users increased and those willing to invest in learning Solaris decreased, not to mention a larger and larger percentage of Solaris guys knew both.

When I went to work for a startup in California, they couldn't afford the quotes for Sun, so I purchased three DELL servers and installed Linux on them to accomplish the same task. Now nobody asks for Solaris admins, they ask for Linux admins.

Re:Sad

[TheLink]

> They're trying to become the Mercedes-Benz of IT. They're going mostly to the high end of the enterprise.

Climbing up the ladder may work, but don't be surprised by how high and fast Intel, AMD and their partners can climb ;).

DEC and SGI also went for the high end of IT.

HP's HP/UX is good as dead, and they've done a good job killing off Tandem and VMS.

So far IBM is still holding out with their POWER stuff.

Re:Sad

[ultranova]

I find it really depressing to find the Oracle logos all over the Sun site and Java downloads.

Yeah. It really makes me worry about Java's future. Is there any other comparable cross-platform language that runs in a managed environment?

Re:Sad

[interval1066]

"They probably just want it as the bottom layer in a hardware & software bundle, tuned to running Oracle or Java workloads."

I think that is the point exactly. I think Oracle inherited all this OSS stuff from Sun and they're still scratching their heads trying to figure out how it will fit into their bottom line. I know Oracle is not new to OSS, I think InnoDB was open, but the Sun acquisition dramatically expanded their OSS offerings and I think their still playing catchup with understanding exactly how all this technology fits with their profit plans.

Re:Sad

[CAIMLAS]

We're moving the OpenSolaris installs to FreeBSD

That, is a mistake. I strongly recommend you do some reading about the base requirements for ZFS on FreeBSD as well as its many shortcomings (at least compared to the OpenSolaris implementation).

Just a couple of the shortcomings I've hit against in the past couple months:

* stability issues. Even with the supposed "stable" 8 RELEASE and the 'required' ZFS tuning and hardware, I've had ZFS lock the system. It would appear the only significant difference between the 7.3 and 8 ZFS implementations is that in 8, they've removed the "EXPERIMENTAL!" warning on the opensolaris driver.
* boot mechanisms. There is no 'official' way to boot off a ZFS zpool, and all the ways that exist to get around that shortcoming are poor compromises, won't work from one release to another, or require use of unstable code (USB boot device, grub2, etc.)
* ZFS requires a *minimum* of 4GB of RAM for supposed stable operation. It will use that memory, even on an infrequently accessed file server. You will have stability issues with less, even with the recommended FreeBSD ZFS tweaking.
* Compared to Linux or OpenSolaris, FreeBSD stability - largely related to device drivers - is pathetic and amateur.
* A general "unprofessional asshole" attitude on the mailing lists. "I've discovered a bug, here it is" seems to result in things like "we're not going to fix that, we'll replace the system in the next release" or similar - if any response is made at all (admittedly, the only list I'm currently following is freebsd-usb).
* ALong those lines, the inclusion of incomplete/dysfunctional systems (presumably) simply on the basis of superior design.

Zones, however, would probably be pretty well implemented via jails. Those are cool. But ZFS is, IMO, not a good choice for picking FreeBSD. FreeBSD does a subset of things very well (networking, documentation, infrastructure design and naming), but ZFS is, unfortunately, not one of them (yet).

I'm very concerned that Oracle is going to kill ZFS off. It is one of the coolest, most useful things to come to storage in a long, long time. Hopefully the Linux folks can pull their pants up quickly and come out with something feature comparable.

Re:Sad

[pak9rabid]

Who is going to buy Solaris when the only thing tested on it is Oracle Database?

People looking for a stable, tested OS to run Oracle on?

Re:Sad

[Anonymous Coward]

Support a community well and it will pay you back.

Yeah, that worked out great for Sun, didn't it?

Re:Sad

[Galactic Dominator]

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

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

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

Re:Sad

[TheRaven64]

For some value of 'tested'. When the only people filing bug reports are Oracle paying customers, you get something a lot less tested than when anyone can download and try it.

Re:Sad

[jgagnon]

Oracle probably only cares that Oracle software runs on it. Everything else is superfluous.

Re:Sad

[KonoWatakushi]

Oracle doesn't care about you unless you're willing to spend a lot of money. That's not bad, that's just their business model.

No, that is bad; in the long term it will ensure their own demise. Actually, never mind, that is good. It is just a shame that they are wasting all of Sun's innovation in the process.

This is one of the great problems with investment in America; senseless greed ensures that the bulk of the money is concentrated in short-term high-return investments, which produce little overall value outside of these investments. The companies are set on a course for self-destruction, and the investors jump ship when appropriate.

Re:Sad

[hairyfeet]

I don't think old Larry is trying to destroy Solaris as much as he has a plan for it that I doubt includes OpenSolaris. My guess is right now old Larry is working the hell out of the Solaris developers to integrate Oracle DB as tightly as possible with Solaris OS and Sun hardware to make sure that Solaris+SPARC+Oracle equals the fastest implementation of Oracle DB you can possibly get. Then of course old Larry is gonna make back his money plus a hefty profit by charging out the ass for the combo, which they WILL pay as those hooked on Oracle DB tend to be pretty loyal.

So "free as in beer" simply doesn't figure into that equation, which is why I doubt they'll give a shit whether the OpenSolaris team folds or not. Old Larry may be a bastard like Bill or Steve, but he is a bastard that knows how to get a ROI and make some serious cash. A tightly integrated top to bottom stack gives him total control of the platform, as well as ensuring he has what he needs to make sure Oracle DB runs the fastest on HIS equipment. I just don't see giving away an OS figuring into that equation.

Re:Sad

[poet]

Although I agree, kinda. We are wrong.

The market cap of Red Hat is 6.06B
The market cap of Oracle is 119.57B

Oracle doesn't need "a community" in any way. Communities are great if the bottom line isn't the priority. RHAT makes the bottom like "a" priority but they are certainly not making it "the" priority. They can't because they are an Open Source company and without the community they are hosed. Oracle needs a community like Bill Gates needs a loan.

Re:Sad

[rubycodez]

nice? pumping millions into sco during their lawsuit against Linux?

Re:Sad

[morgan_greywolf]

You mean like Red Hat Enterprise Linux or CentOS, which is built from the exact same source?

Running Oracle on Linux isn't different from running Oracle on Solaris.

Re:Sad

[gilesjuk]

Solaris could very easily fall behind in terms of support for modern hardware and principles.

Solaris is a small advantage over Linux and competitors, but they seem more interested in Linux.

Re:Sad

[Capt.DrumkenBum]

But since Solaris does not allow me an encrypted file system or even a reasonable way to encrypt part of the file system, Solaris is useless to me.

Re:Sad

[pak9rabid]

Bingo

Re:Sad

[Bert64]

The more recent sun machines were very competitively priced, i had a few of their fairly lowend amd based 1u boxes and they were roughly the same price as similar spec machines from dell and hp.

They could run windows, linux, bsd or whatever, but they could also be bought fully configured and supported with solaris.. There's a lot to be said for buying hardware and software packaged up and designed to work together... That's part of the attraction of Apple too. You know that if you buy a sun box everything will work out of the box with solaris and it will be extremely stable...
With OpenSolaris you got all the freedom of linux combined with the stability of a fully supported hardware/software bundle.

Re:Sad

[Bert64]

A good packaging system is pretty much the domain of open OS's, commercial vendors seemingly don't want to make it easy to manager software... No commercial OS really has ever really had a decent package management system.

Re:Sad

[M. Baranczak]

I wouldn't worry about Java. It's open-source and very widely used. If Oracle cuts it loose, there are plenty of others to pick up the slack. Worst case: we'd have to make a fork and call it something other than Java.

Re:Sad

[Bert64]

Also look at the Itanium processor line, it was too expensive so hobbyists couldn't afford them, and too new so they couldn't buy old ones...
Availability of a platform to the masses increases user experience of it, and users like to run what they're familiar with and have used before.

Re:Sad

[Anonymous Coward]

Well developed open projects however allow for greater mindshare leading to more people using their commercial offerings. Look at Red Hat, because RHL was well used on people's personal desktops, it made sense for them to push a company towards Red Hat's commercial products.

Red Hat is still a magnitude of order smaller than Sun's software business was. However, there also are Mandriva, SUSE, and so on, most of which at best break even, have been taken over, etc. pp.
No, Red Hat is not successful because they are open, have a community, or whatever. All those commercially unsuccessful Linux distributions have that, too.
Red Hat is successful because they are by far the biggest player in the Linux market. Self-concentration of the market on the biggest player, regardless of quality. See also under Microsoft.

Re:Sad

[Doc Hopper]

It's not that it wasn't up to the job, it's that the features weren't/aren't backported to Solaris (10) yet.

Right you are. I stand corrected. My main experience with the 7000-series storage devices comes from some training classes, followed by hands-on recently as we've received a few of the devices with many, many more back-ordered due to the global solid-state disk shortage.

Thanks for the clarification.

Root file systems on ZFS were originally OpenSolaris-only, but are now possible in recent updates of Solaris 10.

Yep. Happy day! I'm running Solaris that way right now (though typing this from my Ubuntu Linux box).

Re:Sad

[Anonymous Coward]

Wow. You must be kidding. See below:

We're moving the OpenSolaris installs to FreeBSD

That, is a mistake. I strongly recommend you do some reading about the base requirements for ZFS on FreeBSD as well as its many shortcomings (at least compared to the OpenSolaris implementation).

Just a couple of the shortcomings I've hit against in the past couple months:

* stability issues. Even with the supposed "stable" 8 RELEASE and the 'required' ZFS tuning and hardware, I've had ZFS lock the system. It would appear the only significant difference between the 7.3 and 8 ZFS implementations is that in 8, they've removed the "EXPERIMENTAL!" warning on the opensolaris driver.

The limitations of running ZFS on 32-bit systems is well-documented. Try an amd64 box instead with the same amount of RAM you would get in a Sun box, like, oh, 8 GB or so. ZFS is *ported* to FreeBSD. It isn't magically shrunk down to require less resources.

* boot mechanisms. There is no 'official' way to boot off a ZFS zpool, and all the ways that exist to get around that shortcoming are poor compromises,
won't work from one release to another, or require use of unstable code (USB boot device, grub2, etc.)

Untrue. Our production SAN boxes are all GPT/ZFS-boot. Nothing compromising about it, it's rock-solid and quick. You have to follow some easy directions (the install program won't do it for you, and you must restore the boot code with gpart if you overwrite it). Google "FreeBSD Root on ZFS" for the instructions.

* ZFS requires a *minimum* of 4GB of RAM for supposed stable operation. It will use that memory, even on an infrequently accessed file server. You will have stability issues with less, even with the recommended FreeBSD ZFS tweaking.

Yes, which is less than an equivalent Sun box would require, isn't it?

* Compared to Linux or OpenSolaris, FreeBSD stability - largely related to device drivers - is pathetic and amateur.

You ignore historical Netcraft surveys which show BSD boxes run longer than Linux on average. Our own FreeBSD servers have a 99.99% uptime. We stick with name-brand devices which may explain our lack of driver issues. I would put driver quality in FreeBSD up against Linux any day, since I run both in our data center. Linux requires more care and feeding than FreeBSD. This has been my experience for over 15 years.

* A general "unprofessional asshole" attitude on the mailing lists. "I've discovered a bug, here it is" seems to result in things like "we're not going to fix that, we'll replace the system in the next release" or similar - if any response is made at all (admittedly, the only list I'm currently following is freebsd-usb).

There aren't as many folks working on the BSDs - no Fortune 500 companies employing armies of kernel developers like Linux. Sorry you got that impression. We're just overworked, not assholes. It sounds like someone was focused on a new system rewrite and didn't have the bandwidth to address a minor bugfix. Major bugs are addressed as a matter of course.

* ALong those lines, the inclusion of incomplete/dysfunctional systems (presumably) simply on the basis of superior design.

There's an old generalization which is fairly true: "BSD is developed by those coming from a Unix software background. Linux is developed by those coming from a PC software background." Superior design is part of the Way of Unix. It pays off handsomely later on. I wouldn't expect youngsters to understand.

Zones, however, would probably be pretty well implemented via jails. Those are cool. But ZFS is, IMO, not a good choice for picking FreeBSD. FreeBSD does a subset of things very well (networking, documentation, infrastructure design and naming), but ZFS is, unfortunately, not one of them (yet).

I'm very concerned th

Re:Sad

[Cramer]

The more recent sun machines... are "PCs" -- AMD and Intel (Xeon) based. The one's actually made by Sun -- the earlier ones weren't, btw -- are nice hardware. But they aren't exactly cheap. And in the Oracle world, you need a support contract just to see a picture of one. 4 year old BIOS updates -- that had always been free -- now require a contract. Documentation of any kind now requires you login -- even for free crap.

Re:Sad

[Bill, Shooter of Bul]

Well BTRFS, is coming along nicely. Distros are looking at using it by default in their next desktop releases. I would advise waiting for at least another year after that before using it to store production critical data.

Re:Sad

[LWATCDR]

Solaris and OpenSolaris do have advantages over Linux. ZFS for example.

Re:Sad

[afabbro]

Oracle doesn't care about you unless you're willing to spend a lot of money.

Correction: Oracle wants you to spend a lot of money, but they care about you as a potential paying customer. For instance, you can pick up a two-user license of Oracle Database for free and run a large production web site on it if you want.

Uh, you can? I thought Oracle Express Edition was free (1 cpu max, 1GB RAM max, 4GB DB size max), but anything past that cost money. And if you hook it up to the Internet, you're paying per-processor, not-per-user.

Re:Sad

[kungfuj35u5]

We're moving the OpenSolaris installs to FreeBSD

That, is a mistake. I strongly recommend you do some reading about the base requirements for ZFS on FreeBSD as well as its many shortcomings (at least compared to the OpenSolaris implementation).

Just a couple of the shortcomings I've hit against in the past couple months:

* stability issues. Even with the supposed "stable" 8 RELEASE and the 'required' ZFS tuning and hardware, I've had ZFS lock the system. It would appear the only significant difference between the 7.3 and 8 ZFS implementations is that in 8, they've removed the "EXPERIMENTAL!" warning on the opensolaris driver. * boot mechanisms. There is no 'official' way to boot off a ZFS zpool, and all the ways that exist to get around that shortcoming are poor compromises, won't work from one release to another, or require use of unstable code (USB boot device, grub2, etc.) * ZFS requires a *minimum* of 4GB of RAM for supposed stable operation. It will use that memory, even on an infrequently accessed file server. You will have stability issues with less, even with the recommended FreeBSD ZFS tweaking. * Compared to Linux or OpenSolaris, FreeBSD stability - largely related to device drivers - is pathetic and amateur. * A general "unprofessional asshole" attitude on the mailing lists. "I've discovered a bug, here it is" seems to result in things like "we're not going to fix that, we'll replace the system in the next release" or similar - if any response is made at all (admittedly, the only list I'm currently following is freebsd-usb). * ALong those lines, the inclusion of incomplete/dysfunctional systems (presumably) simply on the basis of superior design.

Zones, however, would probably be pretty well implemented via jails. Those are cool. But ZFS is, IMO, not a good choice for picking FreeBSD. FreeBSD does a subset of things very well (networking, documentation, infrastructure design and naming), but ZFS is, unfortunately, not one of them (yet).

I'm very concerned that Oracle is going to kill ZFS off. It is one of the coolest, most useful things to come to storage in a long, long time. Hopefully the Linux folks can pull their pants up quickly and come out with something feature comparable.

What sort of hardware were you running this on? I did just fine with 4 disks and 64 bit hardware (very cheap 64 bit hardware). It was only when 4 disk became 12 disks did I need a faster processor, but now it's humming along just fine with a mini-ITX board. My dedicated intent log (single SSD) is actually being choked by the PCI bus right now, but apart from that I've seen flushes to disk of 465MB/sec and I run it stably for months on end. I eventually popped another 2GB of memory in the machine just because it was dirt cheap. I'm using an onboard crappy nforce controller, a Silicon image controller (can't remember the exact chipset, but it's a SATA3 based card), an adaptec SaS controller, and a really really garbage priced RocketRAID controller. And since when did the GPT ZFS bootloader ever require grub2? You've been able to install a ZFS on root config for a long time now. The only thing to watch out for is the break in ABI between userland and kernel during a run of "freebsd-update". Ways around this include: building from source to update (not really that difficult), using a seperate cheap UFS root or special emergency console based environment, or booting into single user mode after the reboot and mounting the zfs partitions forcibly without using the zfs userland utilities (now supported). I'm not trying to call you out on these things, I just want to know why your experience with FreeBSD is so much different than mine.

Re:Sad

[Hognoxious]

Oracle probably only cares that Oracle software runs on it. Everything else is superfluous.

Bit of a pisser, because their customers might want to run superfluous stuff like, I dunno, their applications.

Re:Sad

[afabbro]

I don't think old Larry is trying to destroy Solaris as much as he has a plan for it that I doubt includes OpenSolaris. My guess is right now old Larry is working the hell out of the Solaris developers to integrate Oracle DB as tightly as possible with Solaris OS and Sun hardware to make sure that Solaris+SPARC+Oracle equals the fastest implementation of Oracle DB you can possibly get.

Nope. Oracle is a huge Linux booster. Solaris they're keeping around because a lot of people still use it, but the Oracle internal mindshare is primarily Linux. New versions of Oracle are always released on Linux first, they run their internal stuff on Linux, their filesystem/volume manager (ASM) runs on Linux, they have their own Linux distro, etc.

Re:Sad

[hairyfeet]

That is true now, but will it be true in 16-24 months? look up "Larry Ellison full stack approach" and I'm sure you'll find plenty of talks by Larry talking up the advantage of the "Big Blue" style of top to bottom management. And besides it simply makes sense: If a new feature comes to the kernel in Linux that hurts Oracle DB by 8%, but speeds up the entire system otherwise, what are the odds Larry can pick up the phone and have that feature dropped? Probably zero.

But with Solaris Larry controls the entire OS. HE decides what direction it will take, HE decides what is a priority and what isn't. I have NO doubt that within 24 months, hell maybe even by early next year, Larry will be touting a "New" Solaris Oracle stack where Solaris is basically a stripped to the bone OS designed for max throughput for Oracle DB. Same will probably be done with SPARC. Sure Oracle will continue to run on Linux and Windows, but those that want max throughput will switch to SPARC+ SOlaris+ Oracle DB stack. After all there is only one vendor, which means only one source of support. In the enterprise a "one stop shop" is VERY appealing, and Larry knows that.

But if you don't think that is so what do YOU think Larry is gonna do to make max ROI on Sun? He didn't buy it out of the goodness of his heart you know, and I certainly can't see Java or OO.o being worth the money he paid for Sun.

Re:Sad

[ultranova]

This is not exactly a surprise. Oracle has a deep and abiding interest in Oracle's bottom line. How does Open Solaris contribute to that? It doesn't, hence Oracle losing interest fast.

Yes... And this rises some interesting questions about just how well capitalism works for an economy mostly based on intangibles. Based on this, and MPAA/RIAA's actions, I'd say: not very well.

Specifically, the questions are: what should replace capitalism, and how can it be replaced? When mercantilism came to its end, nation-states were still the greatest powers around; but now corporations are far stronger, to the point of pretty much dictating nation-state legislature for their benefit. How do you transition something like that to a new model?

Maybe we just have to go the route of the French Revolution again, with corporate CEOs acting the role of Louis. It's unfortunate, but then again: repetition is the mother of learning. And humans are still too stupid to learn that it's unwise to give any single person total dominion over them, be they a CEO or King, and be the dominion based on appeals to God of Abraham or the Almighty Dollar.

Re:Sad

[Hooya]

> Not if they don't want or need a community around their product.

The thing about that is that while you may not NEED a community around a product for making the next quarterly estimates, long term, I highly doubt that Solaris would have any edge over other OSs that do have a community around them.

> I suspect Oracle are of the opinion that there's no money to be made selling Solaris as a general purpose OS

Which is another reason why no money will be spent on Solaris to maintain any edge over other OSs.

So, Solaris will die a death brought on by both edges of the sword. No money AND no community.

Re:Sad

[CAIMLAS]

I'm using mostly SuperMicro boards (Intel everything, "recommended" FreeBSD hardware), but I have some other generic stuff in the shop.

The one FreeBSD ZFS machien I've got that has more than 4GB (it has 8), as well as the low-utility ones, are the only systems I've not had stability issues with. FreeBSD 7.2, 7.3, and 8.

Where is it that you see that you can use ZFS root only "for a long time now"? Are you one of those people who thinks "stable" is stable enough, even though it's strongly recommended against using it in production?

I'm fairly new (6 months) to FreeBSD, having been a linux administrator since '98, using many different filesystems. This includes XFS and Reiser 3 and 4 when independent kernel patching was still required. It isn't since those days that I've seen the stability issues I've seen with ZFS on FreeBSD.

Your idea of "supported" is apparently drastically different than mine, because I've yet to see information presenting these things as facts; are they only on the mailing lists? The ZFS documentation I've seen has been sparse, poorly written, or (essentially) provided by Sun (Oracle), with little FreeBSD specifics aside from "this should work, try it".

The way I'm currently doing ZFS is to use a USB key for initial boot. This was working well until 8.0 came out and broke the USB stack, making booting from USB... "unreliable", at best. (Having the keys randomly fall off the chain until they were physically unplugged was also fun).

Re:Sad

[CAIMLAS]

Apparently you haven't read this:

http://en.wikipedia.org/wiki/zfs

Hell, even XFS is closer despite its lacking features, if only because it's stable and available. btrfs fits neither of those.

Re:Sad

[gmack]

The primary reason the Itanium died was because it preformed poorly for all but niche tasks. Intel's dream of super compilers that can take over instruction scheduling was wishful thinking.

The final nail in the Itanium coffin was AMD forcing Intel to come out with 64 bit Xeons that ended up outperforming the Itanium.

Re:Sad

[ToasterMonkey]

Support a community well and it will pay you back. Alienate a community and you are suddenly competing against better entrenched products.

Do you mind explaining how competing open source communities exist?

but without experience, most tech people are going to recommend BSD or Linux because it is what they have worked with.

I think both you and I know the real problem is inexperienced people making recommendations. Is supporting or competing with ignorance demonstratively better than the other? I'm still confused which side desktop friendly Solaris fits on. Don't get me wrong - user friendliness goes a LONG way, but "make it like Linux" is not impressing me. With as much resources as Sun/Oracle has, I'm hoping they'll set the bar much higher.

Re:Sad

[Bing Tsher E]

They didn't support a community very well. The years and years of thumbing their nose at Linux with Java was just one example.

Re:Sad

[drsmithy]

A good packaging system is pretty much the domain of open OS's, commercial vendors seemingly don't want to make it easy to manager software...

That's because the primary reason packaging systems exist at all is to address dependency hell, which is basically a non-problem on commercial platforms.

Re:Sad

[jlarocco]

The way Sun was being run they would have gone bankrupt and all the projects would have been lost anyway.

Your comment describes the situation perfectly. You want all the goodness, but you don't want to pay for it.

Re:Sad

[dna_(c)(tm)(r)]

Support a community well and it will pay you back.

Yeah, that worked out great for Sun, didn't it?

Their business model worked well until their main sources of revenue - Solaris and hardware - came under pressure from Linux and commodity hardware, i.e. x86 servers.

Until that point everything they did with Java, OpenOffice, Netbeans, MySQL brought them a lot of Solaris business... They simply hadn't answers for changing markets.

Re:Sad

[masdog]

GP is correct. That is just Oracle's business model, and it won't kill them off. It may hurt them in the future, but they can limp along like Microsoft for a long time.

The businesses that can afford, and need, Oracle's products are 1)very large, 2)prefer to pay for support contracts, and 3)will not find an open-source equivalent. How many open-source databases can scale to Oracle database? How many open source ERP and/or HRMS packages can operate on the same scale as JDE and Peoplesoft? Would you trust some open-source ERP package running with MySQL backend to handle a million+ daily DB queries and transactions in a multi-site manufacturing environment?

Re:Sad

[Anonymous Coward]

Actually, the software on the Storage 7000 platform is a horrible mess in my experience. We were persuaded by Sun reps about 1,5 years ago to buy 7410's and since we're historically a heavy Sun shop and the price difference between similar platforms was downright huge, we bought it, about 8 of them. Since then we've had nothing but trouble with them in all areas of the platform, from memory leaks in firmware that kills the service processor and takes the entire OS with it (still no fix) to critical NFS, CIFS, ACL, replication and backup bugs. We had to abandon the platform for virtualization use and it's barely making production as a simple file storage. Sun's support on the platform is non-existant, as we are constantly bug-testing their new releases. For each new bug we find, we're promised a fix in the next release, which might be months away. With no exception, all new major releases has introduced new critical bugs to us. Needless to say, had we gone with the much more expensive competitor to begin with, we would have spent less money than we have now. We should have backed out early, but we really wanted to believe that they could stand up to the competition, and now it's too late.

I truly hope your experience will be better.

Re:Sad

[kungfuj35u5]

I'm currently using a CF based root, but I've got a crappy celeron (Pentium III era) laptop running zfs root currently. It's balls slow, but that's really the laptops fault (and it only has 1GB of memory). I managed to do a portupgrade when the last portupgrade was done around FreeBSD 7. It finished just fine after a day or two (putting the CPU at full load for two days by compiling on a compressed zfs volume no less should be enough of a testament of its stability). I wouldn't have any experience with the USB problem as I've only made quick USB boot devices for recovery consoles when dealing with UFS dumping.

Re:Sad

[kungfuj35u5]

Here's the crappy machine I have running on ZFSRoot:

[adam@bsdtop ~]$ uptime
9:11AM  up 48 days, 10:26, 2 users, load averages: 0.42, 0.31, 0.26

[adam@bsdtop ~]$ uname -a
FreeBSD bsdtop 8.0-RELEASE-p3 FreeBSD 8.0-RELEASE-p3 #0: Wed May 26 05:45:12 UTC 2010     root@i386-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC  i386

[adam@bsdtop ~]$ zpool status
  pool: tank
state: ONLINE
scrub: none requested
config:

        NAME        STATE     READ WRITE CKSUM
        tank        ONLINE       0     0     0
          ad0s1d    ONLINE       0     0     0

errors: No known data errors

[adam@bsdtop ~]$ zfs list
NAME                       USED  AVAIL  REFER  MOUNTPOINT
tank                      4.32G  31.1G    18K  none
tank/root                 27.9M  31.1G  27.9M  /
tank/tmp                  18.0M  31.1G  18.0M  /tmp
tank/usr                  3.66G  31.1G  2.15G  /usr
tank/usr/ports            1.22G  31.1G   346M  /usr/ports
tank/usr/ports/distfiles   905M  31.1G   905M  /usr/ports/distfiles
tank/usr/src               299M  31.1G   299M  /usr/src
tank/var                   626M  31.1G   626M
/var

[adam@bsdtop ~]$ dmesg | grep -i celeron
CPU: Intel Celeron (896.69-MHz 686-class CPU)

Re:Sad

[TheLinuxSRC]

Running Oracle on Linux isn't different from running Oracle on Solaris.

Until Oracle drops support for running its databases on Linux. Already, Oracle will not support CentOS, even though it is binary identical to RHEL.

Re:Sad

[Doc Hopper]

I truly hope your experience [[with the 7410]] will be better.

As do I. We've experienced reasonable success in certification for our needs -- principally NFS, with a sprinkling of CIFS -- and are in a fairly unique position to leverage the platform and influence the direction of development. We're not just buying eight of these units. We're buying a few hundred of them...

Re:Sad

[NormalVisual]

FreeBSD does a subset of things very well (networking, documentation, infrastructure design and naming), but ZFS is, unfortunately, not one of them (yet).

Add Xen to that list of "doesn't do well" too. If you're wanting to use FreeBSD as a dom0, you're just SOL. OpenSolaris was great in that you could use Zones for lightweight VMs, but if you needed a couple of Linux partitions on the same physical box you could do that and still be able to keep ZFS because the kernel itself supported all of those options. Being able to just effortlessly pop off another filesystem is great, as is being able to dynamically resize them at will. No such luck now, and it looks like a switch back to Linux will be in my future soon. LVM management is truly painful after you've gotten used to ZFS.

Re:Sad

[KonoWatakushi]

Microsoft is the perfect example here. The bad will that they have generated with their business practices will ensure that they have no future, or at best, one in which they "limp along."

Open source alternatives to Oracle may not exist today, but they will arrive in time. As for not trusting OSS solutions, that is plain nonsense; many critical systems (including stock exchanges) already run on OSS platforms.

Re:Sad

[masdog]

First off...I'm not distrusting OSS solutions. As you can see, my scenario was very specific to the products that Oracle sells - high-end databases and Line-of-Business applications. Many times, these will be installed on Linux servers, so they are partially OSS.

That said, many Open Source ERP systems run MySQL as a backend. Would you trust that for a mission-critical 24x7 5 9's availability system that gets hammered throughout the day?

As for Stock Exchanges, which ones run on an entirely open source platform? If you say London, you'd be wrong. The database that runs their new system from MilleniumIT (which has yet to be implemented) is Oracle.

Re:Sad

[masdog]

Also, the "bad will" will not ensure that they have no future. Microsoft continues to do well in their core segments - Windows, Office, MS Exchange, Sharepoint, and SQL Server. Yeah, they've taken their lumps with Xbox, Kin, Windows Mobile, etc...but they are still sitting on a pile of cash and they are still making a profit.

What will eventually kill Microsoft is 1)a better Office Suite (sorry, OpenOffice isn't there yet), 2)Mass Adaption of Linux on the desktop, and 3)Open-Source replacements for Active Directory and Exchange.

If Microsoft were to drop their Home Entertainment, Mobile, and Internet divisions, they'd be cutting out a lot of cruft that is dragging the company down.

Re:Sad

[Bert64]

Commercial platforms tend to address dependency hell by having a small set of standard libs on the system and then bundling everything else with the app - a horribly inefficient system.. You can do the same on open platforms too but its generally frowned upon.

Packaging systems just provide so much convenience, i cant imagine having to install and update applications manually and i would never install anything that isn't package managed.

Re:Sad

[Bert64]

Performance had very little to do with it, unless you were trying to emulate x86...

High prices of the hardware...
The vicious circle - a lack of commercial software caused by a lack of users, and a lack of users caused by the lack of software...

Itanium systems would have been fine for open source hobbyist use, except for their high price.

Re:Sad

[drsmithy]

Commercial platforms tend to address dependency hell by having a small set of standard libs on the system and then bundling everything else with the app - a horribly inefficient system.

Actually it's a fairly large set of standard libraries, and whether it's "inefficient" depends rather dramatically on your definition of "efficient". Personally, I find not having to worry about dependency hell vastly more "efficient".

and alone one dark night

[nimbius]

on the internet, i felt my coffee mug rumble and the tubes begin to quake...i felt a fork looming in the distance.

Re:and alone one dark night

[tenco]

Yeah, except CDDL doesn't allow forking.

I guess that's why the CDDL is better than the GPL?

Re:and alone one dark night

[Anonymous Coward]

Actually, the CDDL does allow forking.

Re:and alone one dark night

[Chris Mattern]

Yes, it does. The problem is that there's no supporting user community to maintain a fork; almost all of OpenSolaris's updates come from Oracle/Sun.

Re:and alone one dark night

[O Blimey]

3 am IS early.

lolwut?

[wmbetts]

So they're trying to force Oracle to give them a liaison by threatening to cut their own throats? Great move I'm sure Oracle will get exactly what they want.

Re:lolwut?

[valeo.de]

It looks like they don't have any real power anyway, so they're basically telling Oracle they will no longer work to Oracle's benefit for free.

Re:lolwut?

[iammani]

Its like threatening to quit if they dont give you a pay rise. Often not a great move, but people do it all the time.

Re:lolwut?

[Brett Buck]

It's the Judean People's Front crack Suicide Team to the rescue!

Re:lolwut?

[CAIMLAS]

Isn't a project fork still possible, though? "Here's your turd, we're starting our own project!" and so forth. Granted, they'd be stuck with the license, but at least things like Nexenta could continue - yes?

Re:lolwut?

[TrailerTrash]

It's the People's Front of Judea. Bloody splinter group.

I blame Jörg Schilling

[Anonymous Coward]

Is he involved in the negotiations?

Uhhh...

[Jaysyn]

Uhhh.. that will show them?

"If you don't give in to our demands...we'll give up & stop existing?"

Re:Uhhh...

[bsDaemon]

No, it seems more like "If you don't appoint 1 person to sit at a table, we'll dump responsibility for the whole thing on your lap... where you still won't have anyone pay attention to it, so we may as well all just cut to the chase and declare OpenSolaris dead."

Re:Uhhh...

[ak3ldama]

...so we may as well all just cut to the chase and declare OpenSolaris dead.

This is absolutely what is happening. From that post I linked to:

I once advocated this kind of self-implosion tactic back in the Sun days. The reason was to re-organize the OpenSolaris leadership to be more engaged and industry focused. That was a good idea back in the days when I had faith that Sun would "do the right thing". However, those times have past. Oracle has made it clear that it either controls things or it doesn't... there is no give and take. I don't think we can demolish the structure and believe that Oracle will re-organize in such a way as to give the community more power. It was a long shot with Sun anyway.

However, the most important tidbit he reveals lower in his post:

We're in no worse a position right now than we were during the Sun days. They didn't communicate, we had no visibility or impact on the OpenSolaris distribution, etc. Don't fall into the lie that things are now "worse" than they were... they aren't. Its status quo. The difference is that the OGB is no longer composed of Sun insiders who can get a sense of control from hallway conversations and are now as blind and weak as those of us in the community always have been.

My apologies to Ben Rockwood for raping his blog post of content, but this is /. and no one reads anything linked to apparently.

Re:Uhhh...

[Mysticalfruit]

Oracle had one plan when they bought Sun. Kill it and pump a few extra rounds into it, just to make sure.

If these guys are smart, they should just bang the gavel, stop trying to talk to Oracle, officially fork OpenSolaris. I'm sure Oracle will probably go bat shit lawyer crazy and start yelling it's their property, but since they've given it so little interest... who knows.

Re:Uhhh...

[rattaroaz]

If you follow the discussions, the community around Opensolaris is not enough to maintain a fork. 99.9% of the OS is developed and maintained by Oracle now. It's not like the Linux kernel where numerous people/companies contribute. Legally, you can fork Opensolaris given the CDDL. But maintaining a fork is just not realistic. If it was as popular as Linux, then okay, but that is the problem.

Re:Uhhh...

[Doc Hopper]

Oracle had one plan when they bought Sun. Kill it and pump a few extra rounds into it, just to make sure.

Nope. Oracle had a major goal when they bought Sun: create a vertically-integrated platform where they control everything from the hardware through the OS, applications, and support contracts. IBM has that sort of leverage with DB2 + Web Services on AIX, along with a killer international sales and support force with its fingers everywhere in the Fortune 500. IBM is really Oracle's main competition and has been for several years because they could offer whole-life-cycle, end-to-end support at a fraction of the cost of Oracle's offerings. The acquisition of Sun allowed Oracle to compete where it was getting hammered.

Re:Uhhh...

[pak9rabid]

"If you don't give in to our demands...we'll give up & stop existing?"

Now where have I heard this before? Oh, right.

Re:Uhhh...

[MBGMorden]

Legally, you can fork Opensolaris given the CDDL

Also important to note is that given that the CDDL is not compatible with the GPL, you can't merge in the good things about OpenSolaris into Linux (not sure about it's compatibility with the BSD license).

With that it mind, you literally would have to maintain it always as a completely separate project, compounding the lack of manpower.

All in all, I think it's best to just give up on it. It doesn't have the steam the Linux has. I say we just use this as a good example of just WHY you should be careful about what license you release your code under, and the licensing of projects that you contribute to.

Re:Uhhh...

[anilg]

Though a fork (in the sense of OpenBSD/FreeBSD) is not possible, a fork in the sense of Linus's tree, and Alan Cox's tree is possible. The Nexenta project itself already maintains such a tree (nexenta-gate) for the Nexenta and derivative distributions.

In short, though Oracle develops a major part of the kernel, it's open source nature still allows for multiple paths the community can take. The healthy Nexenta community is a testament to that.

We do have some plans for OpenSolaris in the near future. If you're attending DebConf in the first week of August, look me up (and my talk).

Re:Uhhh...

[Mysticalfruit]

That actually makes a lot of sense, though their ambivalence towards OpenSolaris is a bit troubling.

Very sad to witness this

[DrunkenPenguin]

I'm very sad to witness this mess, but I suppose it's just best to accept the fact that Solaris ceased to exist the day Oracle bought Sun. Solaris is and never will be the Solaris we old hackers got to know and love. Maybe it's better to accept it and move on. Still - very sad thing.

Re:Very sad to witness this

[C_Kode]

Solaris isn't dead. Oracle said as much. It's OpenSolaris that has become roadkill via the Oracle purchase of Sun.

Re:Very sad to witness this

[DrunkenPenguin]

Sun has set. It happened the day Oracle bought Sun. I want to personally thank each and every Sun employee who ever got anything to do with the amazing Solaris OS. Thank you. http://www.youtube.com/watch?v=wEUGF3NGbPg

Re:Very sad to witness this

[DrunkenPenguin]

I don't believe in Oracle Solaris. Sorry.

Re:Very sad to witness this

[hondo77]

Solaris is and never will be the Solaris we old hackers got to know and love.

Oh, I don't know. I had to use Solaris a couple of years ago and I felt like I was back in 1992 all over again...

Re:Very sad to witness this

[Sponge Bath]

I felt like I was back in 1992 all over again...

Ah, 1992. George H. W. Bush vomits into the lap of Prime Minister Kiichi Miyazawa on TV. Riots in Los Angeles. Hurricane Andrew. Windows 3.1 is released.

Good times.

Re:Very sad to witness this

[countSudoku()]

You must not get near Unix or Linux much then, because they ALL look like that at the command line, even the latest versions not coded in the '90s. Still, good for you for putting down the mouse for two minutes and getting into some REAL computing. As a very long time Solaris Admin (22 years +, kids) I am not surprised at how Oracle completely ignores the open source community in favor of more $$. Look how they are fumbling SPARC, Java, MySQL, and now Open Solaris. In case you're not aware, which is what it seems here, Open Solaris is/was where all the latest crap is/was built. It was a very stable "testbed" for Solaris 11, now it's about to get forked, but anything interesting will never make it back into the main line code for Oracle's Solaris 10 updates. It is sad, but I saw this coming, which is why I'm all about Linux and VMWare now. I've done everything with Solaris I need to do; if you've not used the Zones then you've missed out on a very well designed, free, virtual environment that will become a fond memory now. So, good for Oracle! Go make lots of money with your expensive, proprietary products. I've seen OAS and Oracle DB. I support open source in the enterprise, and I push open source into the enterprise. Solaris bought my house, but Linux/VMWare is going to buy the next one. So long, and thanks for all the fish, Solaris!

Re:Very sad to witness this

[arth1]

The road to the perfect OS sure has a lot of big roadkills:

- DEC OSF / Digital Unix / Tru64 - gave us 64-bit and support for more than one CPU type.
- Unicos - gave us NUMA and threading that actually worked.
- IRIX - gave us xfs, OpenGL, mixed 32/64-bit environments and tonnes of new GUI features now found in all OSes.
- NeXT - gave us heartburn, upset stomach, indigestion and envy of those who could actually BUY a system.

and now OpenSolaris goes too, and will undoubtedly drag Solaris with it in the long run.
With HP-UX on life support, HURD being just a wet dream, and BSD evolving as quickly as granite, there's not a lot to choose from anymore, at least not for servers.

Re:Very sad to witness this

[Rob Riggs]

Solaris isn't dead.

It's just pining for the fjords.

Re:Very sad to witness this

[h4rr4r]

With no OpenSolaris, Linux will continue to kill Solaris in every market.

Why the silence?

[saleenS281]

The thing I don't get is Oracle's utter silence on the issue If they're discontinuing Opensolaris (even though they've claimed they aren't twice now), why not just come out and say it? They stopped releasing builds, they aren't interacting with the governing board... you'd think SOMEONE there would realize that saying nothing is WORSE than just pulling the plug. If they are indeed pulling the plug, and came right out and said it from the start, I'm sure they'd find plenty of customer still willing to use Solaris proper. At this point though, going by the mailing lists and customers of mine who run solaris and Opensolaris, they've simply turned many paying and non-paying (but mindshare) customers into bitter people. The type of people who go out of their way as admins to make sure that a company's wares don't exist in the shop in any form if they can be replaced. The stuff they've been doing is the type that causes irreparable damage to their image. At some point they're going to have to figure out that people can very easily move away from an OS... it's not like their database customers that are generally stuck with them because it's so much work to move to something else.

Re:Why the silence?

[Anonymous Coward]

I was a Sun employee and hoped that Oracle management would have more ideas with what to do with the technologies than Sun. But we were told 'margin is king and nothing else matters'. So where is the margin in Open Solaris, MySQL, or most other Sun products that Oracle can cash in with ASAP?? If there is not a high margin then Oracle will kill it.

Of my group of about 25 when Sun was taken over, less than a quarter are left. And the others are looking.

Re:Why the silence?

[0racle]

I'm sure they'd find plenty of customer still willing to use Solaris proper

Are they willing to pay for it though? Oracle removed the ability to download and use Solaris 10 for free. This isn't 10 or 15 years ago, Linux and *BSD are more then capable of doing most of the loads you would throw at Solaris an RHEL has the cooperate support and a sane company backing it.

Oracle seems to be looking at Solaris the same as they look at their Database product. Oracle Enterprise Database, even for all its irritations and faults does get the job done very well and does shine against it's competitors, Solaris doesn't have that same position any more. There are many just as good products out there.

Re:Why the silence?

[Doc Hopper]

Oracle removed the ability to download and use Solaris 10 for free.

Thanks for playing. Please try again.

1. Register at sunsolve.sun.com.
2. Click "Downloads & Trials", and select "Top Downloads".
3. Under "Servers & Storage Systems" select "Solaris".
4. Download the option most suited to your needs. For certain releases, you may be asked some survey questions first. If you're not certain you want Solaris full-time on your workstation, I'd suggest going with the VirtualBox image.

The assertion that Oracle no longer allows you to download and use Solaris 10 for free is completely FALSE. I hate seeing this canard repeatedly trotted out as if it were true. There were a couple of days during the support transition and shutdown of legacy Sun data centers when Solaris downloads were affected, but that's been fixed for quite a while now.

Re:Why the silence?

[saleenS281]

That's FUD. You can still download and run Solaris for free. You only need to pay for it for commercial use.

http://www.ypass.net/blog/2010/04/solaris-licensing-changes-the-real-story

Linux and *BSD still have nothing to compete with zfs and dtrace. Their competition to crossbow (new networking stack) sucks. And even with SELinux, can't touch the fine grained access controls built into Solaris. But I suppose if you've never used it, it seems like it doesn't do much different.

Re:Why the silence?

[starfishsystems]

I've been involved with Sun since early days, and intensively with Oracle for the past couple of years.

The two cultures couldn't be more different, especially where support is concerned. It's a good litmus test. Sun would stand behind its products. If something wasn't right, all you had to do was say so and someone who knew what he was talking about would go and fix it. I don't mean that the process was perfect, but there was some agility to it. Oracle seems to stand in front of its products. If something isn't right, the default response is that it must be your fault. I don't know if the support staff are trained to play dumb or gradually get worn down by the Oracle culture, but the effect is the same either way. It's a huge distraction to have to educate the people who are supposed to be providing support, especially when their efforts seem to be bent on looking everywhere but at the issue.

So here's the bottom line. Often we're stuck for one reason or another with a particular product line. The difference is that I was happy to buy millions of dollars of Sun product. I loathe the very thought of having an Oracle product on site.

Re:Why the silence?

[CatsupBoy]

What buttons do I have to click to get my free patches? Oh that's right, they don't supply patches for free anymore.

If you think downloading a base image constitutes as using for free, then I'm afraid you are mistaken. It takes security patches and bug fixes to keep an OS in production quality working order.

Maybe you just re-install every 6 months when the new media set is released? right!

Re:Why the silence?

[mandelbr0t]

What buttons do I have to click to get my free patches? Oh that's right, they don't supply patches for free anymore.

When did they ever? Yes, Oracle is selling Solaris support. So what? If you need Solaris, then you should be willing to pay for it. If not, Oracle supports Linux as well. The Oracle bashing has gotten to be a bit much. Sun would have been dead one way or another; they're lucky someone with cash decided to trawl through all of it and decide some of it was worth keeping. It's unfortunate to see an open community die like this, but if it can't survive without Oracle, then there probably weren't many people there to begin with.

Re:Why the silence?

[CatsupBoy]

Yes, you could download patches for free from sunsolve for many years before Oracle.

This is not bashing, its the truth. Competitors such as IBM and HP currently provide patch information, downloads, and knowledge base articles for free, Oracle does not. And now they bled that mentality into Solaris.

It's unfortunate to see an open community die like this, but if it can't survive without Oracle, then there probably weren't many people there to begin with.

There were plenty of ppl involved and interested, but its based on a proprietary platform, of course it can't survive without the parent company's support. The larger point is that they are, seemingly, purposefully killing it.

Re:Why the silence?

[GarrettZilla]

I think this is the sentence in the license agreement that convinces people otherwise:
"Please remember, your right to use Solaris acquired as a download is limited to a trial of 90 days, unless you acquire a service contract for the downloaded Software."
http://www.sun.com/software/solaris/get.jsp

Re:Why the silence?

[Doc Hopper]

What buttons do I have to click to get my free patches? Oh that's right, they don't supply patches for free anymore.

Wrong again. You get LOTS of free patches with a free install of Solaris. RedHat set the pace for this: if you install RHEL, you have to use up2date which requires a registered system with the RedHat Network (RHN). If you don't want to register and pay for RHN, you wait for the next release and do your upgrade from that. Sun implemented a similar system -- in planning, testing, and preliminary deployment LONG before the acquisition -- requiring registration and a support contract number before allowing entitlement to certain patches in a more timely fashion than the traditional six-month release cycle.

Maybe you just re-install every 6 months when the new media set is released? right!

Actually, you can upgrade off the install CD from the media sets, too, without reinstalling. Always have been able to, and it's a simple, easy way to keep up-to-date, though it requires some downtime to install. Downside: no zero-day exploit fixes. Upside: free patch sets every six months. As long as I've been working with Solaris -- since 1999, and up through right now while I'm downloading the latest kernel exploit & StarOffice 8 security patch on my Solaris box -- the zero-day security exploits are listed in the patch entitlement for ALL Solaris systems, not just those with a support contract.

Upgrade old release (7, 8, 9): http://docs.sun.com/app/docs/doc/817-3799/6mjcan1v6?l=en&a=view
Upgrade newer release (10): http://docs.sun.com/app/docs/doc/817-0544/6mgbagb1c?l=en&a=view (x86 on this page; the SPARC install instructions are also in the documentation)ma

To get your security patches, go to Launch-> Applications -> Utilities -> Update Manager. Go through the registration wizard. Choose "Continue without providing a Service Plan Number". Accept the software license agreement. Finish the registration; if you want to use this as a base image for mass-deployment, click the "enable auto registration" option.

Next select all updates, and install them.

I understand you're concerned with not having the latest-and-greatest usability and functionality updates to your OS on a faster-than-6-month schedule. If it's of sufficient concern to you, register for a cheap Solaris support contract through SDN and be done with it. But for the rest of the world that wants to continue using Solaris for free, CRITICAL SECURITY PATCHES ARE AVAILABLE TO ANYONE WITH A SUN ONLINE ACCOUNT.

Free security updates online as soon as you get around to installing them. Free every-six-month usability and functionality updates. What exactly is the problem with this patch schedule? That those who choose to pay nothing for a great operating system don't get usability and functionality updates on the same schedule as paying customers?

OpenSolaris exists to fill that niche: customers who need bleeding-edge features on a very timely schedule and don't want to spend a lot of money. You can even patch production Solaris boxes from OpenSolaris patches if you wish, though I understand some assembly is required. Never done that myself. Never felt the need.

Overall, I think the Oracle acquisition of Sun has been a good thing for both companies. Sun gets to keep the lights on and payroll flowing, Oracle gets a bunch of hardware & software products in its portfolio.

Re:Why the silence?

[Anonymous Coward]

Unfortunately, you are both correct and incorrect. I've dropped Solaris (my favorite OS) as it's no longer free to use. The OS still is a free download, HOWEVER, read that license agreement... It says that the license is only valid if a support contract is purchased!

Re:Why the silence?

[Doc Hopper]

The language of the license is for production and production evaluation use only. Check out the real story here:
http://www.ypass.net/blog/2010/04/solaris-licensing-changes-the-real-story/

Excerpt:

The license and accompanying entitlement from the web, without a contract and without hardware, only entitle the downloader to non-commercial, non-production, or personal use in perpetuity. Production use and evaluation for production are good for 90 days.

The whole patch debacle is a tempest in a teacup. If you're a hobbyist or personal user, it's just business as usual: download, install, get your patches on a 6-month schedule, get your zero-day patches as soon as they're available on Sunsolve.

Re:Why the silence?

[Doc Hopper]

That's FUD. You can still download and run Solaris for free. You only need to pay for it for commercial use.

If I didn't have several posts in this thread, I'd mod this "informative". The anti-Solaris FUD spread far and wide since the acquisition has been ridiculous.

Most of the hysteria, I think, springs from the transition adjustments. Suddenly Oracle went from several years of an unwritten policy of "we do not buy any Sun hardware" to being the #1 customer of Sun hardware. Every trade-in, every re-purpose, every non-end-of-lifed system that Sun/Oracle gets its hands on is being recertified and used internally. Oracle had a HUGE pent-up demand for hardware to get a large number of projects off the ground, and Sun is scrambling to provide it. There's a lot of uncertainty right now just because of the growing pains, but with a new $300Mn data center going up in Utah due to open in 2011 (with three more expansions planned already), and maxxed-out data centers all over the world with hardware being refreshed, there's enough demand in the channel to backlog providers of everything from solid-state disks to capacitors.

Sun's scrambling to meet this huge demand -- in addition to a large number of companies that also held off orders due to acquisition-uncertainty -- with a reduced staff from those who left voluntarily or were let go during the acquisition. Oracle's spending out of the war chest, finally able to appease the hardware-hungry project base that had been stymied for years while it was hunting for a hardware vendor to buy.

There are, of course, growing pains. But the two companies, IMHO, are going to come out of this much stronger than before, and able to compete head-to-head in the IBM/ACN/HPQ/MSFT space.

Re:Why the silence?

[yuhong]

On the other hand, why do you think Sun was losing money?

Re:Why the silence?

[Doc Hopper]

And this whole process only authorizes you to use the download for 90 days on an evaluation basis. Beyond that, you have to cut a cheque. And, as another poster mentioned, you don't get any patches either.

Already debunked your assertions in this thread.

1. Hobbyists and personal use were unaffected by the license changes; what is affected is COMMERCIAL production or development use beyond 90 days: http://developers.slashdot.org/comments.pl?sid=1719254&cid=32904884 . Summary: Oracle's never going to come bash in your front door to extort license charges for that virtual Solaris you're running on your laptop or on that Sparc III you fished out of the dumpster.

2. Critical security patches are available generally and immediately; usability and functionality upgrades are on a six-month release basis for non-paying customers: http://developers.slashdot.org/comments.pl?sid=1719254&cid=32904728

Just 'cuz a load of Anonymous Cowards on Slashdot keep saying you can't get patches for unpaid Solaris installs doesn't make it so!

Re:Why the silence?

[CatsupBoy]

Sun changed the patch access policy early 2006, years before Oracle bought Sun. http://www.mofeel.net/1208-comp-unix-solaris/8197.aspx Ever since then you needed to have a support contract to legally download Solaris patches.

No, you just had to have a sunsolve account. It did not have to have an account tied to it (see the second comment on the link you provided).

It also is not true that IBM provides patches for free - you cannot download AIX itself for free, duh. AIX also doesn't run on non-IBM hardware. So you have to pay IBM at least twice before you can even use AIX.

IBM indeed provides patches to their AIX operating system without even logging in. Check out this fix central link for AIX patches:

http://www-947.ibm.com/systems/support/supportsite.wss/supportresources?brandind=5000025&osind=5329283&taskind=2

You are correct about not being able to download AIX for free, however when you purchase hardware you get a free license to run AIX on it. And with free patches you are not required to maintain vendor support in order to maintain it.

Re:Why the silence?

[Cramer]

All the way up to the day after Oracle took over. Remember "patchPro"? "smpatch"? Yes, there were contract only patches, but they were rare and generally fixed issues few people had (or needed fixed.)

For the record, I gave up on Solaris with the bastardized crap that is "Solaris 10". SMF! "We don't use shell scripts anymore." -- bullshit, they just aren't in etc anymore. And it smells too much like the Windows Registry.

Re:Why the silence?

[harmonise]

Maybe you just re-install every 6 months when the new media set is released?

Just like with Ubuntu and Fedora.

Re:Why the silence?

[IanBal]

> The assertion that Oracle no longer allows you to download and use Solaris 10 for free is completely FALSE

Read the licence agreement! "Please remember, your right to use Solaris acquired as a download is limited to a trial of 90 days, unless you acquire a service contract for the downloaded Software."

The changes that Oracle made to the licence agreement are best highlighted at http://www.cuddletech.com/blog/pivot/entry.php?id=1120

Solaris under Sun had a chance. Solaris under Oracle will fade away to some unknown unimportant Oracle byproduct.

Re:It is obvious

[stakovahflow]

Feels like kind of a shame, doesn't it? But I'm still glad I never made the "switch" from FreeBSD/OpenBSD to Solaris/OpenSolaris...

--Stak

Old Aggie joke

[BenEnglishAtHome]

An Aggie* comes home to find his wife in bed with another man. He pulls out a pistol and points it at his own head. His wife screams "No, don't do it!" The Aggie replies "Just wait; you're next."

* - Footnote for people not from Texas - Students at Texas Agricultural and Mechanical (A&M) University are called Aggies and are the subject of endless jokes insulting their intelligence.

Re:Old Aggie joke

[johnncyber]

And as a Red Raider I thank you.

Re:Old Aggie joke

[johnncyber]

That is I thank you for making my morning.

Re:Old Aggie joke

[Anonymous Coward]

Jokes? I thought those were news reports!

Re:Old Aggie joke

[sconeu]

Why did the three Aggies go to the Xmas party dressed as firemen?

Because the three wise men came from afar.

(read it out loud, it makes more sense).

Re:Old Aggie joke

[herring0]

That's a new one on me. I'll have to tell it to the rest of my Aggie family.

Re:Old Aggie joke

[pak9rabid]

And as a UT fan, I hate you.*

*-Footnote: I don't really hate you.

Accepting reality

[ThrowAwaySociety]

Uhhh.. that will show them?

"If you don't give in to our demands...we'll give up & stop existing?"

It's not like they can really threaten Oracle into submission. Sometimes, you just have to roll over and ask, "Honey, are you really in this for the long run, or are you just screwing me?" If you don't like the answer, you just pack up and leave. No need to go all psycho.

What were we talking about again? Oh yeah. If the organization disbands, Solaris loses some of its credibility as an open platform with a healthy, involved community. Not a death blow, but better than prolonging a charade.

What about VirtualBox?

[yuna49]

I don't really care about OpenSolaris, but I have been a happy user of VirtualBox since before it was acquired by Sun. Sun developed some nice, but proprietary, tweaks to VirtualBox in areas like graphics drivers. I do see development continuing as I get prompted to upgrade fairly regularly, but I've been nervous that VirtualBox will also eventually be treated as roadkill by Oracle. Obviously there will always be a free implementation since the "open-source edition" is GPL-licensed.

I can understand Oracle's lack of interest in OpenSolaris since they've supported Linux for a long time now. (Hell, they even compete directly against RedHat with their Oracle Enterprise Linux distribution.) I do wonder, though, whether they'll stay committed to VirtualBox down the road.

Re:What about VirtualBox?

[Doc Hopper]

I do wonder, though, whether they'll stay committed to VirtualBox down the road

I hate to engage in speculation, but Oracle now has two virtualization solutions:

1. Server-side "OVS" or "OVMS": Oracle Virtual Server. This is a Xen-based implementation used widely within Oracle under the framework of their Grid and Elastic Grid products. It's portable, scalable, and is a huge revenue-generator in areas like Oracle Education.

2. VirtualBox, which is more of a client-side, "run it on the desktop" app.

They both have their niches, so I don't see either going away any time soon. OVS is a beast to manage on more than a handful of servers, and paravirtualization (required for good virtualization of Windows) is just now getting rolling onto the "good" side of the usability & performance hump. While Vbox has worked great in that environment.

Speculation: I think we may see some sort of interoperability merge in the future between Vbox & OVS. I am fairly certain there is no development along those lines right now -- Oracle's really busy working on integrating all the web-services & database stuff acquired from Sun, PeopleSoft, and other acquisitions -- but I bet it's on a roadmap somewhere.

It's not just Solaris

[jimicus]

Most proprietary IT companies have started to - if not openly embrace - at least accept and try to work with F/OSS (both in communities and with sponsored products).

Some have been quicker than others; some have taken it more seriously than others. Some have quite obviously put up with F/OSS under protest rather than actively encouraging it.

Oracle, I would say, at least before they acquired Sun, has definitely been in the "not taking entirely seriously" camp. A traditional proprietary vendor, who consider anything free (either speech or beer) to be a toy, something the kiddies can play with and then when they grow up they can start using the Super Enterprise Product. So not only is this not a surprise, I'd say it's almost expected. Frankly, there are projects which I'm far more concerned about - MySQL and OpenOffice immediately spring to mind.

Re:It is obvious

[codepunk]

MS is completely 100% irrelevant to me I could care less what they do.

Ah Oracle

[sir lox elroy]

You have already showed what kind of a BOFH you can be. I was attempting this week to find drivers for some of our Ultra 20s, I can't even download drivers without a stinkin Maint Agreement. This is why I went with MySQL years ago, and not Oracle, hmmm time to change to PostGRES and dump all the Sun equipment.

Re:I'd just like to interject for a moment.

[S.O.B.]

I didn't realize RMS posted on Slashdot.

Sad for OpenSolaris

[guruevi]

Especially now that we have a need for Sun's Solaris, they close up shop and sell to Oracle. Sun was far ahead of their game implementing virtualization and extreme storage limits in a time when there was no need for it. Now the industry is finally ready to get started using these technologies and they decide to sell it. Sun's systems were finally starting to get used, systems like the Thumper have been in great demand. Too bad they wasted money on overpriced 1U servers ($10k for an Opteron system?) and projects that were very good but never went anywhere (trying to build on Java's success). Instead of selling the company, they should've cut off 80% of it (including their idiot C-level executives) and concentrated on selling ZFS, Xen and Zones

Re:It is obvious

[DesScorp]

Look it is obvious, Oracle is putting a nail in anything having to do with Solaris. Get over it, move on and start migrating.

No, Oracle is putting a nail in OpenSolaris. They're quite interested in developing commercial Solaris. They just want to be paid for it. You don't make money by not making money. You'd have thought everyone would get that now after the Internet bubble 10 years ago.

Today Solaris, tomorrow MySQL?

[Animats]

The future of MySQL is iffy, too. All Oracle has to do is put the second team on maintaining it, and it will die. A database program has to be very reliable to be usable at all.

We've already seen this with "MySQL Workbench". Since Oracle took over, all the MySQL GUI tools were wrapped into a central "MySQL Workbench" program. Which crashes frequently. (If you can install it at all.) If Oracle can bring MySQL down to the level of MySQL Workbench, nobody will be able to use it.

MySQL needs to be fully archived, including the revision history, outside of Oracle, just in case.

Re:Today Solaris, tomorrow MySQL?

[Blakey Rat]

We've already seen this with "MySQL Workbench". Since Oracle took over, all the MySQL GUI tools were wrapped into a central "MySQL Workbench" program. Which crashes frequently. (If you can install it at all.) If Oracle can bring MySQL down to the level of MySQL Workbench, nobody will be able to use it.

The MySQL GUI tools have *always* been this bad.

Although I will say this: although the UI was designed by aliens from planet Weebow whose only experience with GUI system was a vague explanation over the phone, at least it didn't crash. It was completely useless, but stable.

Re:Today Solaris, tomorrow MySQL?

[Lonewolf666]

MySQL needs to be fully archived, including the revision history, outside of Oracle, just in case.

There is already a somewhat credible attempt at maintaining a fork (http://askmonty.org/wiki/MariaDB#What_is_MariaDB.3F).
Monty may have been a bit of a hypocrite in first selling MySQL and then whining about what happened to it, but now he does approximately what you asked for :-)

It's not sad, it's normal

[VincenzoRomano]

When you have enough money, you rule. That's it.
Want to kill an open ource project? Buy the "supporting" company (if any, like MySQL AB) or buy^H^H^Hhire the top 5 or 10 developers.
Is the control of the project already in your hands? Simply forget about it.
Then sue all forks.
It's (sadly) normal.

Re:It is obvious

[tychovi]

Actually Microsoft bought Sysinternals hired the developers and made the whole thing better and is still giving it away for free. Those utilities are good enough that I would pay for them and I believe I did donate back before MS acquired them. Oracle on the other hand was saying:

"We've been in the open source business a very long time. We've been a distributor of Apache and we have our own version of Linux," Ellison said. "We have no problems having both Linux and Solaris and we want to make them both better." -- Larry Elison Interview on http://www.serverwatch.com/news/article.php/3861376/Whats-the-Future-of-Linux-and-Solaris-at-Oracle.htm

While the reality is that they're letting OpenSolaris die on the vine...

A view from the inside...

[Anonymous Coward]

I was a Sun employee, I'm now an Oracle employee. I've posted in the past about internal, but non-secret Sun stuff using my registered nickname because I didn't think it mattered all that much. These days, however, the corporate secrecy is verging on paranoia, and so I don't dare use my regular nickname.

Anyhow, I'll keep this short. First of all, Oracle does not say anything to anyone outside of Oracle about future plans. Period. It's repeated over and over in the brainwashing (er, onboarding) presentations. The rationale for this is that if customers think they know what new products are in the pipeline and when they'll come out, they'll plan their purchases accordingly. There's also the potential loss of competitive advantage.

Second, Oracle doesn't give a rat's ass about building communities and generating interest with Open Source. They'll re-brand Red Hat because they know people want Linux servers, but they don't care about trying to make Open Solaris a gateway to "real" Solaris. They'll make Solaris the premier platform for high-end Oracle DBs, and they'll use it for storage solutions which take on NetApp. Beyond that, they don't care about whether or not Solaris "wins" against Linux. They don't need it to. The goal is to leverage Solaris (on Sparc for Oracle DB, x86 for storage) into closed solutions which have huge profit margins. If it's not going to create large margins, it won't live long at Oracle.

Profit is king here. Anything else is overhead, and overhead eats into Larry's yacht fund.

Yes, I'm looking elsewhere. The best and brightest have been leaving in droves. I am neither, but I'm still pretty good; just somewhat less mobile. Working on that.

Re:A view from the inside...

[anilg]

/me wonders how long before someone on opensolaris-discuss posts this in a thread, leading to another flamewar and a new round of speculations on the nature of OpenSolaris ;)

Re:A view from the inside...

[Anonymous Coward]

The best and brightest have been leaving in droves.

Does that include Solaris Kernel Engineering folks? If so that is a very interesting situation. My reading was that the kernel folks never wanted to Open Source Solaris - at least not GPL it and so may be they are happy with the axe falling on OpenSolaris and they might just be staying. But it is hard to believe hard core Sun talent adjusting with the realities at Oracle. (Sun was Engineering dominated - as you pointed out, Oracle is all about $$$.)

If the kernel folks are leaving it's huge loss for Oracle if they intend to keep Solaris alive and kicking - it could be Linux's gain if those people made the next logical career choice. Any rate worth keeping an eye on.

Re:A view from the inside...

[yuhong]

These days, however, the corporate secrecy is verging on paranoia, and so I don't dare use my regular nickname.

Yes, I know. How do you think it would compare to say Apple?

Re:A view from the inside...

[Anonymous Coward]

Dear Colleague,

It all boils down to two words.

Regards,

ex-MySQL/ex-Sun/now-Oracle.

Re:A view from the inside...

[belthize]

I was at the LUG 2010 (Lustre Users group) and there was a very similar sense of disquiet where Lustre is concerned. The corporate line seemed to be: You'll be able to download Lustre for free but if you want any kind of support you'll have to install Lustre on a box from one of our preferred vendors running our Linux variant.

Granted it's not clear exactly what 'any kind of support' means. As it stands now the mailing list is very active and it doesn't really matter what your support status is. If that stays the same then wonderful. If the Lustre devs at Oracle are instructed to stop interacting with all but paying customers it's a real problem.

Re:It is obvious

[MBGMorden]

Depends. Big shops like commercial products with commercial support. They'll pay for it too. HOWEVER, the people to run that stuff? If you expect them all to pick up the skills on the job then they're going to be a rare breed, and rare talent = EXPENSIVE talent. If you have some version of your software available for free though, without support, then you let the tinkerers learn it in their spare time.

I'm convinced that's why things like IBM's i Series OS (OS/400) isn't picking up many more users. It still has a ton of people running legacy stuff from mainframe days when EVERYTHING required specialized help to run it, but these days? It's a pain to learn it. If they had some free virtualized version or something that would run on x86 I'd pick it up and learn it in a heartbeat, but the reality is that anyone who wants to learn that OS is either picking it up in an apprenticeship fashion from someone else, or they're laying down many thousands of dollars for a server to play on. It's just not realistic. On the other hand if company wants to use Linux? They have a huge pool of people who can pick up skills on that OS at home for next to free. Even Windows is only $100 or so if bought separately, and if included with a new computer typically is only $20-30 of the total cost, making it peanuts in comparison.

That's how you you make money by "not making money". Give your software to the enthusiasts and you'll make it more attractive to companies to PAY to implement with the bigger talent pool available.

How funny

[WindBourne]

I remember the first time that Sun opened SunOS. I supported it. But, when they closed it, it became obvious to me that it is trivial for commercial companies to pull this kind of BS. SO, a number of years pass and suddenly SUn re-opens Solaris. As I pointed out, that if and when Sun (or any owner) want to kill it again, then it would happen. How many of the sun marketers stormed in here declaring that it absolutely could never happen again. Had the Sun fan bois screaming that I was just FUDING. And yet, here we are AGAIN.
WHile I will likely do Solaris contracts, I will never again push their junk. It will only be a matter of time before they pull a MS/Sun style hit.

Re:FP

[couchslug]

"Where are my GNAA homies?"

RTFA, they are evidently running Oracle.

Re:It is obvious

[h4rr4r]

Solaris brought out opensolaris to try to deal with the fact that linux was killing them. This will move ensure Solaris ends up as dead as HP/UX.

Oracle thanks for giving us all an enemy again

[maliqua]

This is just sick. Oracle buying sun imo was the darkest day in recent IT history. I seriously hope oracle suffers some major financial losses because of the animosity there tactics with our beloved sun have created.

I hate to say this

[hellraizer]

but how can this really be a surprise ? this was bound to happen since Sun got sold, damn you oracle!!!!

Great!

[Hell0W0rld]

More Devs for Linux, especially for Ubuntu. :P

Re:Great!

[Anonymous Coward]

Another Sun/Oracle internal person here....

Yea, but will they listen and accept the seasoned incoming ex-Sun folks? I have a great libthreads mutex bug, I can get two or more threads in a protected block of code pretty consistently, and have sprinkled assert()'s all over to assure it's the library and not our code. Nope... Linux maintainer doesn't care. I'm not in the "inner circle" and can't make much trouble for him.

If the Linux people want the ex-Sun devs, they need to embrace them as peers, not vanquished competitors.

Re:It is obvious

[Bert64]

Sysinternals tools may be free on their own, but they require you to purchase windows in order to use them.

You don't need to buy anything from oracle in order to use opensolaris.

Re:It is obvious

[TheHedonismBot]

I could care less what they do.

I think someone needs to study up on The Caring Continuum.

Re:It is obvious

[afabbro]

Look it is obvious, Oracle is putting a nail in anything having to do with Solaris. Get over it, move on and start migrating.

No, Oracle is putting a nail in OpenSolaris. They're quite interested in developing commercial Solaris.

They're interested in doing the minimum necessary to keep Solaris customers. Oracle loves Linux and that's where its future energies will be predominantly spent. Put another way...I suspect you'll see a robust Linux on SPARC before you'll see many more versions of Solaris.

ffffork forkforkforkfork fork fork ....

[unity100]

do i need to repeat it more ? fork ....

In defense of Oracle...

[swordgeek]

Sun didn't turn from a bright shiny wonderful company to a sacrificial lamb at the helm of an evil empire in a single day with a single set of signatures on a sales contract. Not all of "Sun's" problems stemmed from Larry, or the sale.

Let's not forget a few years ago, when mentioning Sun on /. would get you spit on. Let's not forget that Jonathan Schwartz spent his entire tenure at the helm cutting, cutting, and cutting again, to make Sun an unstable empty shell of its former self - but very appealing as a buyout target.

Oracle wants to make money. If they're not going to be helping with OpenSolaris, then it'll go off on its own way. This might be the best thing for it. In the meantime, think about who else could have bought Sun and done differently (better) with it. All contenders - IBM, Microsoft, Cisco, NetApp, Dell, would have either fumbled as badly, or shuttered it more aggressively. And thanks to Schwartz's manipulations, they couldn't have survived this long without being bought.

Don't get me wrong--I don't like Oracle, I REALLY don't like what support from them is like (on our $~0.6bn worth of Sun gear), but it bugs me to see the former Sun executive--ESPECIALLY Schwartz--avoid their share of the blame.

Re:It is obvious

[grcumb]

No, Oracle is putting a nail in OpenSolaris. They're quite interested in developing commercial Solaris. They just want to be paid for it. You don't make money by not making money. You'd have thought everyone would get that now after the Internet bubble 10 years ago.

Don't ever ever become an economist. If all you can see in this is straight-up software sales, then concepts like opportunity cost (how much more you stand to spend because you've ceased to take advantage of free developer/tester time), secondary effects (like building goodwill in the software community in order to create better sales opportunities), diversification (brought on by allowing externalities to influence your tactics and strategy) will be very difficult for you to grasp.

I don't have a horse in this particular race, but killing off a FOSS project through neglect is a decision that should not be taken lightly. The knock-on effects of pissing off a bunch of talented and committed developers and leaving them with nothing to show for their efforts is a dangerous one. It's not too hard to imagine the Oracle salesman getting a frosty reception from the sysadmin who invested time and effort in getting the company's crown jewels running on OpenSolaris, only to be told there was no clear way forward (see the comments above for examples of why commercial Solaris is not a straightforward upgrade path).

Even RedHat were wise enough to provide Fedora with a springboard to take off from when they divested themselves of any commercial interest in desktop Linux. Yes, they made some false steps and they pissed off more than a few individuals, but they were at least open, honest and communicative about their decision and the reasons for it. They continue to be supportive of the Fedora community to this day. Viewed in retrospect, they did it well, albeit not perfectly.

Arguably, OpenSolaris was moribund and destined to fail because it was too similar to Linux. Be that as it may, simply cutting off the project's air supply, especially when it cannot survive without Oracle's support, seems negligent at best, punitive at worst.

Again, I don't have any particular insight into the process and I don't have a stake in the outcome. It could well be that there's more to this than meets the eye. That's for others to say. But as the old saying goes, 'Beware of geeks during rifts.'

Why doesn't OGB fork OpenSolaris?

[Luke has no name]

I know forking isn't a preferred method. However, if the mainline forks, isn't it just 'declaring independence' so to speak?

Imagine if a single company somehow magically controlled the current Linux mainline and the name "Linux". If Linus said "OK everyone, we're branching and MY version will be on linuxfuckingrocks.org" then everyone would follow him.

Re:Why doesn't OGB fork OpenSolaris?

[natehoy]

I guess it's a matter of how much effort is available to work on the fork, and whether Oracle would allow such a beast to exist without a patent shitstorm of some type.

Does OpenSolaris fit a market niche that justifies a lot of effort on the part of unpaid and corporate-supported developers? And within that niche do its merits justify the potential risk of having to survive an onslaught of Larry's Lawyers?

There's a critical mass of people necessary to maintain such a fork and sustain its use, and I'm not convinced there's enough developer population to sustain a fork in the absence of support from Big Brother Oracle, and even less so in the face of its active opposition.

Linux? Sure. There are a lot of people who want it, including a lot of corporations interested in using it in embedded apps and distributions (RedHat, Ubuntu, Cisco, etc), and enough of them are willing to add to the kernel and other projects that there's more than enough critical mass of interested contributors to keep it running for, well, a very very long time. Even us non-corporate, non-developer user types can help write the occasional document or contribute test results, bug reports, a few bucks here and there, and user-to-user help to various portions of Linux and its ecosystem of applications. And many of us do as a form of payment for the tremendous gifts bestowed upon us by the OSS community as a whole.

But what niche does OpenSolaris fill, exactly, as an open source OS? How many corporations (other than Oracle) receive benefit from expanding it? How many people who can contribute to the codebase would be interested in using a non-corporate-backed fork? How many would run the risk of a future lawsuit over some aspect that may be dependent on something Oracle bought the rights to and slaps a patent on?

Would you bet your business on a fork of code whose trunk is owned by Oracle? Are there enough businesses that would do so that they'd spend their developers' wages contributing to the codebase?

I think if Oracle allows the board to close, the only "fork" going into OpenSolaris is the one referred to in the phrase "put a fork in it, it's done." OpenSolaris will quietly wither as it sits, unchanged, and all enhancements will get integrated into Pay-Solaris only.

Re:Why doesn't OGB fork OpenSolaris?

[Luke has no name]

I'd just as soon Opensolaris die. I don't know of any reason why it sets itself apart from other OSes besides DTrace (maybe?). I haven't watched Opensolaris in a while, but with ZFS in FreeBSD, and its package manager sucking compared to , I can't see a justification for its existence.

Last thought: Crossbow was cool. Don't know if that exists in BSDland.

Re:I'd just like to interject for a moment.

[ci4]

Wait, your Username is "Oracle"? Speaking as someone with a significantly lower UID than you; Go to hell.

Go to hell, and die.

Wait, if I say the same, should you drop dead then; I've got one digit less in my UID...