There’s a rumour going around that Oracle is close to ending all development of Solaris, effectively killing the operating system.
Solaris being canned, at least 50% of teams to be RIF’d in short term. All hands meetings being cancelled on orders from legal to prevent news from spreading. Hardware teams being told to cease development. There will be no Solaris 12, final release will be 11.4. Orders coming straight from Larry.
It’s just rumours for now, but they’ve been gaining steam over the past few days.
Hopefully if it happens it’ll give a boost to OpenIndiana: https://www.openindiana.org/
I’ve had an urge to play around with an OpenSolaris VM but wasn’t sure which of the myriad of options were the best. Thanks for the link, I may try this one instead. I was previously leaning towards Dyson but had misgivings about its active support…
They are still hiring in the support department, so I doubt it.
I guess it’d be likely they would move the development to another party if they don’t want to do it themselves anymore. Or indeed release the code all freely.
There’s a big difference between continuing active development and providing support.
Yeah, they have to honour the contracts they have.
agree, the company that I used to work for just signed a 10 year support contract for Solaris so it will be interesting to see how that pans out if the rumours are true.
I would also add that the 10 years is only just buying us time + blow outs to port all our proprietary code from SPARC + Solaris to x86 + Linux (RHEL)
Edited 2016-12-05 08:07 UTC
This was obviously coming for some time. All of the new Solaris development since Oracle took control has been very meager. As an example, there have been no system ABI changes in the past six+ years (which is why OpenIndiana can run Solaris 11 binaries, and vice versa).
Especially given that Fujitsu is closing down their development of new UltraSparc/Sparc64 processors (they were by far the most active UltraSparc OEMs), and since one of Solaris’s best features was great UltraSparc support (better than any other OS, way better than Linux), it makes sense that Solaris development would slow or even stop now.
Edited 2016-12-01 23:13 UTC
Solaris development since Oracle has been meager? How is that? Theres been a consistent major release almost every year since Oracle acquired Solaris and the latest release coming out, expected to be called Solaris 12 has over 1400 enhancements with major updates to security & cloud
And FYI, Fujitsu has not stopped development of SPARC. More worthless rumors. Although they switched to ARM for their latest HPC system, its apparently a one off system for a specific customer request. Fujitsu still developing SPARC, have a 5+ year roadmap and are expected to release their next gen SPARC64 XII soon. http://www.fujitsu.com/global/products/computing/servers/unix/sparc…
I usually post sobre finely calibrated comments.
But Oracle.
http://giphy.com/gifs/cheezburger-olympics-U7P2vnWfPkIQ8
Some months ago I mentioned how bad and unstable Solaris was becoming… the amount of bugs We suffered in Solaris 11 prod environments was HUGE compared to Solaris tradition, very very nasty storage related bugs… and the Oracle attitude regarding them was so strange, almost “Ok, yeah, it’s a bug, I don’t give a shit, thanks”. Now, It’s very clear why that happened: Oracle was stopping Solaris development, reducing teams and so on. Result: crappy software.
But the thing is… Solaris is the only enterprise OS available for SPARC… if they kill Solaris… what OS Oracle will sell with SPARC iron? Unbreakable Linux for SPARC?! I don’t think so…
So… If they kill Solaris, they will have to kill SPARC too. And if Oracle kill Solaris and SPARC… and they don’t give a shit about Java anymore… why the hell they bought Sun in the first place?!? To keep Sun’s tech/IPs away from IBM?? To keep Sun’s tech/IPs away from Apple??
What was the fucking point? Oracle ruined the best IT company ever created for nothing.
Larry Ellison, I hate you bitch! xD
I’m always reminded of this bit:
https://www.youtube.com/watch?v=-zRN7XLCRhc#t=2000
“Don’t anthropomorphize Larry Ellison”
Except that Oracle has not been giving UltraSparc much love. Since they bought Sun, Fujitsu became the main supplier of Sparc systems. However, they recently announced there next HPC systems would use ARM instead of UltraSparc.
If the main producer of UltraSparc systems is throwing in the towel, it’s not too tough to imagine what will happen to the main OS for UltraSparc.
Since Oracle acquired Sun, all SPARC developments at Oracle are 100% Oracle. Fujitsu has their own SPARC development team. Yes, there is a lot of co-engineering to make certain a high level of compatibility, and Oracle resells Fujitsu systems outside of Japan but the two have very independent engineering teams. Since Oracle acquired Sun, they’ve introduced 7 new SPARC processors in 6 years. The latest SPARC M7 is currently leading all the world record benchmarks and even the next gen Intel and Power CPU’s still wont beat it. At over 10Billion transistors, 32-cores and integrated Software in Silicon and Silicon Secured Memory there is nothing in the industry that comes close.
That’s true, I think SPARC development was pretty good on par with PowerPC.
But right now, Solaris is the only commercial OS available for SPARC… if you stop or slow down Solaris development… what’s the point of SPARC existence?
Even Fujitsu SPARC systems are used as Solaris servers.
IMHO if you kill Solaris, you kill SPARC too.
Well, you do have Linux, but there are no distros that provide a decent level of support.
OpenBSD provides the best support outside Solaris. When I discovered how much better they supported UltraSparc than Linux, I began to understand why they insist on all development and builds for an architecture be self-hosting. If a platform is unfit to host its own development (barring embedded platforms) or compiling its self, your support is pretty poor.
Larry must be buying a country. Yachts are not where it is at now.
Seriously,
I drive past the old SUN HQ in the UK almost every day.
They had a lot of really good engineers who cared about the stuff they sold. That all changed when Oracle took over.
This seems to mirror the HP takeover of Compaq/DEC. Soon all those of us who cared where shown the door.
A huge case of NIH perhaps?
The Sun kit could never return the margins that Oracle gets on their software.
AFAIK, Oracle stopped making he patches for Solaris available unless you had a paid support contract. That for my boss at the time was the writing on the wall. I still have the last of our servers from that time in my Garage. It is going to re-cycling next week.
The end of an era. Ellison will be remembered by many of us for all the wrong things.
Lastly, I hope your boat gets thrashed in the America’s Cup.
How is that different from the way Red Hat or HP do business? You can download Solaris for free, you get updates for free (pkg.oracle.com), but for more, you need a support contract. Same with AIX, RHEL, etc., except for the free download and updates.
The patch thing is true only for legacy Solaris releases (Solaris 8-10). For current release (Solaris 11), you get the updated packages at any time.
Sun had the best engineers in the world because that was the company culture.
Oracle changed that incredible, innovative and unique culture for the standard and mediocre “don’t give a shit” culture that every other big IT company in the world have (not only Oracle, all of them).
I’ve been working as Unix administrator for the last 16 years and I had the opportunity to solve an infinite amount problems with the help of Sun support engineers… they took their jobs with a passion I’ve never seen in other IT companies. Let me give you an example: one time, 10 years ago, a german Sun engineer sent emails to my personal address weeks after a big problem We had with a Solaris bug to know if I needed more help and how everything went…
They didn’t want to close a “ticket” in a queue… they actually wanted to find the root cause and solve the problem!!
I know that might sound stupid… and maybe that level of involvement is usual in other businesses or in the scientific world I don’t know… but in the IT business that’s just incredible.
😉 Also scaring… Such Is the Power of the Modern Framing of that euphemistic -‘World Order’.
Answer: To destroy MySQL (because Oracle is a vendor of a proprietary and rather expensive DB).
Most people forget that Sun Microsystems was sponsoring the development of FOSS projects like OpenOffice, VirtualBox and MySQL (among others) quite well. MySQL, in particular, had evolved to the point it was starting to encroach into the lower-tiers of what used to be the domain of Oracle’s DB products.
Sure, the community will fork, but they won’t have the support in terms of full-time professional devs working on the projects on a regular basis that Sun used to provide, maybe except for the case of LibreOffice. Thus, getting rid of the problem known as “the constantly improving MySQL is encroaching into our turf” was enough justification for Larry to buy Sun. Going with the vertical integration thing was always a secondary “maybe it can work” objective.
PS: I always laughed at tech writers trying to see any other real goal in the Nokia takeover and the Sun takeover beyond the obvious one: Hostile takeover. Oracle bought Sun to stop them from sponsoring MySQL development, MS bought Nokia to stop them from joining the greenbot camp and to stop MeeGo smartphones and tablets from appearing on shelves.
Edited 2016-12-02 14:31 UTC
Was MySQL really a serious contender? Last time I checked MySQL was an entry-level DB, for small projects, not really in the same ballpark as Oracle (aimed at larger corporate users). It was also sub-par in features when compared to PostgreSQL – another open-source DB.
MySQL was and is no serious contender to Oracle.
PostgreSQL, on the other hand, is almost close to being a real contender to Oracle with several years of solid multiple master support.
The MySQL “multiple master” support is a paper tiger.
On a related note, MongoDB doesn’t even *do* multiple master. It’s a toilet paper tiger. But don’t tell the fan boys that their beloved NoSQL has any flaws.
MySQL/MariaDB was (still is I think) one of the most widely used databases in the world. It runs some of the largest sites, including Facebook.
Oracle installation don’t tend to have the scaling needs that MySQL and PostgreSQL or NoSQL DBs sites might have (though that’s a generalization). The very large database-driven sites don’t usually use commercial databases (except for possibly smaller back-end services) because of license costs. Plus, Oracle licensing (which is horrendously expensive) pretty much guarantees that any company looking at massive scale would never even consider Oracle.
Oracle is still the choice for certain enterprise DB needs, which tend to be older enterprise apps, banking, etc. To power middlewear, it’s a tossup I think between Oracle and other database providers, both open sourced and closed. But again, Oracle tends to be very, very expensive.
Edited 2016-12-04 09:46 UTC
you sure?
why taobao able to handle 100billion trade per day by using mysql at backend ? haha
I heard a rumor they were killing SPARC support but not Solaris.
Both.
Solaris 12 already exist & on last stages of QA. It has latest OpenStack & Docker technologies integrated and that 2 technologies are core prerequisite for Oracle “cloud shift” roadmap.
Pretty sure that Solaris 12.2 will be released somewhere because it is prerequisite to use DAX3 – software-in-silicon part of SPARC M9 processor, which already exist in silicon.
It is logical to have Solaris 12.1 between them. Looks like it will be released at the end of next year together with systems on SPARC M8/S8 processors (I already held M8 on my hands!) and will be prerequisite to use DAX2 – M8/S8 software-in-silicon part.
By the way, Oracle SPARC Linux exist too. It will be used in some SPARC based engineered systems in near future. But it will not replace Solaris on generic servers & “universal” engineered systems. May be it will be presented as an “option” to install it on servers, but, as far as I know, no decision yet.
Qetzlcoatl,
Oh how ironic it would be if it were fake.
All fake news should be accompanied with Star Trek’s classic “it’s a fake” scene from DS9:
https://www.youtube.com/watch?v=9HT_s0ITKjk
Haha, maybe then people will get the message
It’s not I’m afraid. They obviously aren’t going to publicise it.
Oracle is all about Java IPs.
Why do why people are still crying about Sun, Solaris and Sparc architecture ?!??!?
Let me break -again and again- their last hope : It is the end since 2009.
Oracle and Fujitsu are talking about support because Sun have long term contracts with some customers.
HW, CPU and OS architectures are far too expensive to maintain. And Oracle would like to stop paying the (high) price for its acquisition.
Stop nostalgia and wake-up, its 2016 : Does a reasonable person buy this kind of iron/os today for an IT need ?
Yes, (s)he does. A Partner company can make a difference with its ties to the customer. At least, that’s my current employer’s perspective and it works.
I acknowledge the decrease in a number of Solaris/SPARC deployments we sell, but this combination still wins us large projects.
“I acknowledge the decrease in a number of Solaris/SPARC deployments we sell, but this combination still wins us large projects.”
Still looking for Solaris people for projects??
Edited 2016-12-02 20:54 UTC
That’s the points :
– Niche applications
– High level (unaffordable maintenance contracts) engineers to maintain it.
For the same price you get an army of persons working on Linux. And they are still able to develop drivers, maintain the system and do the logistic.
Some says it is sad. But that’s the reality of the IT battle field for the present days.
Because they were one of the coolest technology I’ve ever used?
Big companies suck and enterprise products created by them suck even more. Sun was an exception to that rule.
Sun created some enterprise technology that was cool, smart, creative, beautiful and reliable… that’s why I cry when I see Oracle destroying it (or not giving a shit about it).
Almost any IT engineer in the world will feel the same because you admire and love when a company craft good products… because you know very well how hard is to “do the right thing” in big companies. Empathy they say.
It’s been dead long before Oracle got their hands on it. Sun went bust, remember?
If that happens, that would be an end of an era for me.
Another prediction of mine from a few years ago comes true. Solaris has really been dead since the late 90s ever since Linux ran on x86 and Sun didn’t know what to do. Sparc should have been canned a long time ago.
I can’t see it happening but if ZFS got into the Linux kernel I’d be pretty happy. In reality they’d probably just had it on to Oracle Linux as an ‘enterprise’ feature.