[rescue] Oracle kills Solaris

[Liam Proven]

I presume everyone's aware of this?

https://www.theregister.co.uk/2017/09/04/oracle_layoffs_solaris_sparc_teams/

Oracle has laid off ~2,500 Solaris and SPARC developers, leaving a
tiny remnant, presumably to fulfil some support contracts.

Apart from the forks -- Illumos, SmartOS etc. -- that's it. It appears
that Solaris is dead. Sad times.

--
Liam Proven b" Profile: https://about.me/liamproven
Email: ***@cix.co.uk b" Google Mail/Talk/Plus: ***@gmail.com
Twitter/Facebook/Flickr: lproven b" Skype/LinkedIn/AIM/Yahoo: liamproven
UK: [phone] b" DR/WhatsApp/Telegram/Signal: [phone]
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mauricio Tavares]

It is a shame. I used netbsd and then sunos 4.1.4 and
MacOS-something before I ever touched a Windows machine. First windows
machine I owned was a PC card in an Amiga2000. Next one was yet
another PC card but in a Sun Ultra30 running Solaris 9, which was then
replaced with a Ultra60 running Solaris 10. I still like solaris as an
OS and sparc as a CPU (much more elegant than Intel), but refused to
do business with Oracle.

I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

Post by Liam Proven
I presume everyone's aware of this?
https://www.theregister.co.uk/2017/09/04/oracle_layoffs_solaris_sparc_teams/
Oracle has laid off ~2,500 Solaris and SPARC developers, leaving a
tiny remnant, presumably to fulfil some support contracts.
Apart from the forks -- Illumos, SmartOS etc. -- that's it. It appears
that Solaris is dead. Sad times.
--
Liam Proven b" Profile: https://about.me/liamproven
Twitter/Facebook/Flickr: lproven b" Skype/LinkedIn/AIM/Yahoo: liamproven
UK: [phone] b" D R/WhatsApp/Telegram/Signal: [phone]
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan Katz]

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

OpenSXCE, but the maintainer went full Nazi advocating for ethnic
cleansing. And I'm not just invoking Godwin's Law for hyperbole. He
actually said that Muslims and Jews should be rounded up and put in
camps; I have the screenshot of his twitter. I think the project fell
apart before that, though. He didn't have anyone to help him maintain
it or host it.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mauricio Tavares]

Post by Jonathan Katz

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

OpenSXCE, but the maintainer went full Nazi advocating for ethnic
cleansing. And I'm not just invoking Godwin's Law for hyperbole. He
actually said that Muslims and Jews should be rounded up and put in
camps; I have the screenshot of his twitter. I think the project fell
apart before that, though. He didn't have anyone to help him maintain
it or host it.

Do send me the screen capture off-list if you can.

Did the other forks purge all of the sparc code or just ignored it?

Post by Jonathan Katz
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan Katz]

Post by Mauricio Tavares
Did the other forks purge all of the sparc code or just ignored it?

This I don't know.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Stephen Conley]

I've been a lurker on the OpenIndiana list for quite a long time.

My understanding is the lack of sparc support is due to lack of manpower
and equipment.

Not sure if they're the pony I'd bet on for the future of Solaris or not; I
tried to use OI about 4 or 5 years ago and it didn't work on my available
Intel hardware (I don't remember the details, but the curious thing was,
Solaris 11 official from Oracle *did* run on the hardware). I gave up on
Solaris after that, even though it remains to this day my favorite OS.

Anyway, long and short of it, I think OI at least could run on sparc if
they got some people behind it :)

Post by Jonathan Katz

Post by Mauricio Tavares
Did the other forks purge all of the sparc code or just ignored it?

This I don't know.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Dave McGuire]

Post by Stephen Conley
I've been a lurker on the OpenIndiana list for quite a long time.
My understanding is the lack of sparc support is due to lack of manpower
and equipment.

I lurked on that list until I got sick of it. I got the strong
impression that there are a bunch of x86 fanboys on there that think
anything that's not an x86 PC is "legacy", and they had zero interest in
supporting SPARC.

They seemed to fall under what I've been calling the "AS/400 problem",
people thinking that because there were AS/400s in 1988 that all AS/400s
are from 1988. Just because there were SPARC systems in 1993 doesn't
mean all SPARC systems are from 1993.

-Dave

[Jonathan Patschke]

Post by Dave McGuire
I lurked on that list until I got sick of it. I got the strong
impression that there are a bunch of x86 fanboys on there that think
anything that's not an x86 PC is "legacy", and they had zero interest in
supporting SPARC.

You can thank Jonathan Schwartz for that.

I personally decommissioned the last SPARC workstation at my day job
because Sun quit making workstations. At the time, we'd rather have run
Cadence on Solaris. It worked, and the engineers had gotten used to it,
but we'd already been through a disruptive recall on our Blade 1500s, and
at the end of our hardware support contract, Sun had literally nothing
better[1] to sell us.

They got replaced with mediocre Dells running Red Hat. After the
engineers got over redoing a decade of accumulated customization, they
stopped caring. Dell won. The engineers were laying out ICs again, and I
had one fewer platform to support.

The least expensive SPARC machine I can find on Oracle's web site is, at
$11k, almost twice as expensive as the least expensive POWER8 system from
IBM and about five times as expensive as the least expensive Proliant from
HPE, all at the same memory size and the same or greater CPU core count.

The clear message from Sun and Oracle is that SPARC is not for you unless
you're doing very wide HPC or whatever silly "throughput computing" niche
the SPARC T was cut-out for.

Post by Dave McGuire
They seemed to fall under what I've been calling the "AS/400 problem",
people thinking that because there were AS/400s in 1988 that all AS/400s
are from 1988. Just because there were SPARC systems in 1993 doesn't
mean all SPARC systems are from 1993.

No, but all SPARC workstations are at least a decade old, and it's been
nearly that long since Sun, Oracle, or Fujitsu competed at
price/performance in small systems. If they cared about open-source
support for their platform, they'd at least release full documentation and
have some sort of hobbyist buy/lease program. No one is going to pay $11k
to port a fork of an abandoned proprietary OS to undocumented hardware.

When the only samples of a platform at reasonable affordability are
end-of-life, that's a workable definition of "legacy."


[1] The Ultra 25 did have a less infuriating chassis, I suppose.

[Patrick Giagnocavo]

Very sad situation. However, Illumos and related, lives on, and ZFS
has made it out into the wider world, so there's that.

I still believe that Solaris is better under memory pressure than
Linux, however, with RAM being so cheap these days, it doesn't matter
as much... I remember logging in to a heavily-laden Solaris machine
running many zones, with 2MB of free memory, yet SSH was still
responsive enough to use.

Should probably install Illumos or whatever and try to build a
software-service on top of it with zones. While I recognize Docker as
being useful, the sysadmin in me still finds it abhorrent.

Post by Jonathan Patschke

Post by Dave McGuire
I lurked on that list until I got sick of it. I got the strong
impression that there are a bunch of x86 fanboys on there that think
anything that's not an x86 PC is "legacy", and they had zero interest in
supporting SPARC.

You can thank Jonathan Schwartz for that.
I personally decommissioned the last SPARC workstation at my day job
because Sun quit making workstations. At the time, we'd rather have run
Cadence on Solaris. It worked, and the engineers had gotten used to it,
but we'd already been through a disruptive recall on our Blade 1500s, and
at the end of our hardware support contract, Sun had literally nothing
better[1] to sell us.
They got replaced with mediocre Dells running Red Hat. After the
engineers got over redoing a decade of accumulated customization, they
stopped caring. Dell won. The engineers were laying out ICs again, and I
had one fewer platform to support.
The least expensive SPARC machine I can find on Oracle's web site is, at
$11k, almost twice as expensive as the least expensive POWER8 system from
IBM and about five times as expensive as the least expensive Proliant from
HPE, all at the same memory size and the same or greater CPU core count.
The clear message from Sun and Oracle is that SPARC is not for you unless
you're doing very wide HPC or whatever silly "throughput computing" niche
the SPARC T was cut-out for.

Post by Dave McGuire
They seemed to fall under what I've been calling the "AS/400 problem",
people thinking that because there were AS/400s in 1988 that all AS/400s
are from 1988. Just because there were SPARC systems in 1993 doesn't
mean all SPARC systems are from 1993.

No, but all SPARC workstations are at least a decade old, and it's been
nearly that long since Sun, Oracle, or Fujitsu competed at
price/performance in small systems. If they cared about open-source
support for their platform, they'd at least release full documentation and
have some sort of hobbyist buy/lease program. No one is going to pay $11k
to port a fork of an abandoned proprietary OS to undocumented hardware.
When the only samples of a platform at reasonable affordability are
end-of-life, that's a workable definition of "legacy."
[1] The Ultra 25 did have a less infuriating chassis, I suppose.
--
Jonathan Patschke
Austin, TX
USA
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Meelis Roos]

Post by Patrick Giagnocavo
I still believe that Solaris is better under memory pressure than
Linux, however, with RAM being so cheap these days, it doesn't matter
as much... I remember logging in to a heavily-laden Solaris machine

It might matter with containers... I recently ran into trouble with
Linux 4.4 and and LXC containers - memory limit was 2G per container,
this used memory control group as backend for this limit, and cache
reclaim did not work well in memcg, leading to OOM killer visits with
about a third of the 2G allocated for cache (and reclaimable).

Same memcg is used under docker... I wonder if the current kernels are
any better - but for production we moved to full VM-s with same amount
of RAM and these have behaved well.

[Jonathan Patschke]

Very sad situation. However, Illumos and related, lives on, and ZFS has
made it out into the wider world, so there's that.

I appreciate ZFS on my FreeBSD systems almost literally every day. It's
probably saved me months of unavailability, cumulatively.

The code will drive you insane, though. There's a very thick layer of
glue to make FreeBSD's VFS system look like what ZFS expects, to make
FreeBSD's synchronization primitives look like what ZFS expects, and to
make ZFS's memory allocator not trainwreck into the rest of the kernel.
ZFS on Linux probably has similar glue (although I've not looked at that
code).

Thanks to the patent provisions of the CDDL, that glue will never go away
until the patents expire, Oracle changes the license, or Oracle disclaims
the patents. CDDL only grants patent protection for the unmodified
portions of code. So, while there are likely many opportunities for
performance enhancement in ZFS on Linux and FreeBSD, mining those
opportunities is a legal hazard in countries that recognize software
patents.

[Phil Stracchino]

Post by Jonathan Patschke
The code will drive you insane, though. There's a very thick layer of
glue to make FreeBSD's VFS system look like what ZFS expects, to make
FreeBSD's synchronization primitives look like what ZFS expects, and to
make ZFS's memory allocator not trainwreck into the rest of the kernel.
ZFS on Linux probably has similar glue (although I've not looked at that
code).

Yup, it does. It's called the Solaris Porting Layer.

Post by Jonathan Patschke
Thanks to the patent provisions of the CDDL, that glue will never go away
until the patents expire, Oracle changes the license, or Oracle disclaims
the patents. CDDL only grants patent protection for the unmodified
portions of code.

Aaaaah ... a succinct statement of the problem.

[Majdi S. Abbas]

Post by Phil Stracchino
Aaaaah ... a succinct statement of the problem.

When I was 17, and the patent umbrella covered literally
my entire lifespan; this was scary.

Post by Phil Stracchino
20 years down the road, a patent umbrella no longer

seems nearly as scary -- the patent will expire long before
Oracle manages to take advantage of it.

--msa
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Phil Stracchino]

Post by Majdi S. Abbas

Post by Phil Stracchino
Aaaaah ... a succinct statement of the problem.

When I was 17, and the patent umbrella covered literally
my entire lifespan; this was scary.

Post by Phil Stracchino
20 years down the road, a patent umbrella no longer

seems nearly as scary -- the patent will expire long before
Oracle manages to take advantage of it.

I'm not sure Oracle is imaginative enough *TO* take advantage of it. If
Oracle were imaginative, it would change the licensing just enough to
allow *Oracle* to integrate ZFS into Oracle Enterprise Linux and be the
only shipping Enterprise Linux distribution with full ZFS support.

[Michael-John Turner]

Post by Phil Stracchino
I'm not sure Oracle is imaginative enough *TO* take advantage of it. If
Oracle were imaginative, it would change the licensing just enough to
allow *Oracle* to integrate ZFS into Oracle Enterprise Linux and be the
only shipping Enterprise Linux distribution with full ZFS support.

Canonical seem to be happy shipping ZFS with Ubuntu these days, so I'm not
sure why Oracle don't do the same. I always assumed it was because they
wanted to keep their version of ZFS (ie pool versions 29 and higher) as
Solaris-only... (not to mention that the ZFS on Linux project isn't
something they can control).

Cheers, MJ

[Phil Stracchino]

Post by Michael-John Turner

Post by Phil Stracchino
I'm not sure Oracle is imaginative enough *TO* take advantage of it. If
Oracle were imaginative, it would change the licensing just enough to
allow *Oracle* to integrate ZFS into Oracle Enterprise Linux and be the
only shipping Enterprise Linux distribution with full ZFS support.

Canonical seem to be happy shipping ZFS with Ubuntu these days, so I'm not
sure why Oracle don't do the same. I always assumed it was because they
wanted to keep their version of ZFS (ie pool versions 29 and higher) as
Solaris-only... (not to mention that the ZFS on Linux project isn't
something they can control).

This is true, but I've seen it argued that Canonical is on shaky legal
ground in how they're doing it. That aside, it's my understanding that
they are getting away with it by shipping zfsonlinux as a separate
package, not actually built into the mainline kernel which I think is
what we'd all like to be able to do.

[Phil Stracchino]

https://meshedinsights.com/2017/09/03/oracle-finally-killed-sun/

Interesting read even just for the 'scorecard'.

[Phil Stracchino]

Post by Jonathan Patschke
Thanks to the patent provisions of the CDDL, that glue will never go away
until the patents expire, Oracle changes the license, or Oracle disclaims
the patents.

Actually ... when *DO* those patents expire?

[Jonathan Patschke]

Post by Phil Stracchino

Post by Jonathan Patschke
Thanks to the patent provisions of the CDDL, that glue will never go away
until the patents expire, Oracle changes the license, or Oracle disclaims
the patents.

Actually ... when *DO* those patents expire?

That's a great question. I haven't been able to find anywhere that Sun or
Oracle actually lists the patents implemented by ZFS. Whatever shipped
with the first public release of ZFS in Solaris 10u2 will expire around
2026. Stuff new in Solaris 11 will be closer to 2031.

The timeline is more than a little unfortunate.

[Phil Stracchino]

Post by Patrick Giagnocavo
Very sad situation. However, Illumos and related, lives on, and ZFS
has made it out into the wider world, so there's that.
I still believe that Solaris is better under memory pressure than
Linux, however, with RAM being so cheap these days, it doesn't matter
as much...

And it's STILL the best nfsd implementation, by a large margin, IMO.

(Shocking, shocking, I know.)

[Dave McGuire]

Post by Jonathan Katz

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

OpenSXCE, but the maintainer went full Nazi advocating for ethnic
cleansing. And I'm not just invoking Godwin's Law for hyperbole. He
actually said that Muslims and Jews should be rounded up and put in
camps; I have the screenshot of his twitter. I think the project fell
apart before that, though. He didn't have anyone to help him maintain
it or host it.

I worked with him briefly before he lost his shit. I was going to
host the mailing lists and distributions, as well as a build farm. But,
well, then he lost his shit.

-Dave

[Mauricio Tavares]

Post by Dave McGuire

Post by Jonathan Katz

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

OpenSXCE, but the maintainer went full Nazi advocating for ethnic
cleansing. And I'm not just invoking Godwin's Law for hyperbole. He
actually said that Muslims and Jews should be rounded up and put in
camps; I have the screenshot of his twitter. I think the project fell
apart before that, though. He didn't have anyone to help him maintain
it or host it.

I worked with him briefly before he lost his shit. I was going to
host the mailing lists and distributions, as well as a build farm. But,
well, then he lost his shit.

you make me think of Mel Gibson in South Park.

Post by Dave McGuire
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Dave McGuire]

Post by Mauricio Tavares

Post by Dave McGuire

Post by Jonathan Katz

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

OpenSXCE, but the maintainer went full Nazi advocating for ethnic
cleansing. And I'm not just invoking Godwin's Law for hyperbole. He
actually said that Muslims and Jews should be rounded up and put in
camps; I have the screenshot of his twitter. I think the project fell
apart before that, though. He didn't have anyone to help him maintain
it or host it.

I worked with him briefly before he lost his shit. I was going to
host the mailing lists and distributions, as well as a build farm. But,
well, then he lost his shit.

you make me think of Mel Gibson in South Park.

Yes. ;)

-Dave

[Rich Kulawiec]

Post by Dave McGuire
I worked with him briefly before he lost his shit. I was going to
host the mailing lists and distributions, as well as a build farm. But,
well, then he lost his shit.

As apt a description as any. This incoherent rant reminds me of some
of the stuff a few select people used to post to Usenet back in the day:

http://opensxce.org/how2use_Sun-11.x_xhci_on_OpenSolaris/README.txt

---rsk, who always rants coherently of course
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Bill Bradford]

Post by Rich Kulawiec
http://opensxce.org/how2use_Sun-11.x_xhci_on_OpenSolaris/README.txt

Wow. Someone needs their meds adjusted just a bit.

Bill

[Lionel Peterson]

Yikes! That is the best example of why you don't discuss politics at work...

Lionel

Post by Bill Bradford

Post by Rich Kulawiec
http://opensxce.org/how2use_Sun-11.x_xhci_on_OpenSolaris/README.txt

Wow. Someone needs their meds adjusted just a bit.
Bill
--
Bill Bradford
Houston, Texas USA
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mark Linimon]

Post by Bill Bradford
Wow. Someone needs their meds adjusted just a bit.

I feel sorry for that guy. He needs professional help.

mcl
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Mark Linimon]

Post by Jonathan Katz
OpenSXCE

From just a technical standpoint, I tried it on both
sun4u and sun4v machines a couple of years ago. It
booted on neither.

AFAIK if you are on sun4v and want Open Source you are
limited to OpenBSD or NetBSD. The FreeBSD support was
never completed. After investigating I think it would
take at least 2 months to complete.

I have been fighting to keep sun4u ssupport in FreeBSD
but I think with this news I am going to lose that battle.

mcl
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Meelis Roos]

Post by Mark Linimon
AFAIK if you are on sun4v and want Open Source you are
limited to OpenBSD or NetBSD. The FreeBSD support was
never completed. After investigating I think it would
take at least 2 months to complete.

And Linux, of course - I am running Linux on most of my sun4v's.

Oracles own distribution allows to run Linux as LDOM host, some of the
vdc patches have been submitted to mainline, some not yet, and the
userspace software not yet - althou they had plans for that.

The state of sparclinux support from Oracle remains to be seen - in
recent year, quite a number of Oracle developers have contributed, but I
do not know the current state - but hope fo the best.

[Mark Linimon]

Post by Meelis Roos
And Linux, of course - I am running Linux on most of my sun4v's.

Which distro? Most everything I researched at that
time looked more and more x86-orientrd.

mcl
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Meelis Roos]

Post by Mark Linimon

Post by Meelis Roos
And Linux, of course - I am running Linux on most of my sun4v's.

Which distro? Most everything I researched at that
time looked more and more x86-orientrd.

Most of my machines are still at old debian with 32-bit userspace that
is updated no longer. Some of my newer machines are debian sparc64 port
- 64-bit userland following debian unstable. This had some installation
glitches but after that has worked fine, and this is in active
development.

It looks Gentoo port is alive too but I have not had a need to try it on
sparc.

[Nemo]

Post by Mauricio Tavares
I wonder which fork can run on sparc hardware; everything I saw
focused on intel.

There is a partial list at Illumos. Tribblix looks interesting. I
have never tried any because I run Solaris on my boxes.

N.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan Patschke]

Oracle has laid off ~2,500 Solaris and SPARC developers, leaving a tiny
remnant, presumably to fulfil some support contracts.
Apart from the forks -- Illumos, SmartOS etc. -- that's it. It appears
that Solaris is dead. Sad times.

It's probably for the best. Oracle's been pretty clear about being
uninterested in developing Solaris or selling it ($1000/socket/year?).

The code quality of Solaris went to hell after Oracle bought Sun. Few of
the administration tools share a common "feel" or even similar
command-line styles. Lots of features act half-implemented and fragile.
I've run into a few race-conditions in the iSCSI stack, for instance.

It's a night-and-day difference to the solid (if spartan) distribution
that Solaris 9 was.

It would be Really Really Nice if Oracle would open-source (under an
actual open-source license like Apache, X11, or even GPLv2) Solaris and
disclaim their patents over ZFS. But, it's evident that the purchase was
really about controlling Sun's patents and choking MySQL to uselessness.
The last decade has been about milking support contracts, rent-seeking
litigation, and nothing more.

I've long hoped that Oracle would buy Adobe so that the company that
added backdoor-sufficient programmability to Flash, murdered FrameMaker,
and keeps trying to close-off PDF would see a fitting end.

[Phil Stracchino]

Post by Jonathan Patschke
It would be Really Really Nice if Oracle would open-source (under an
actual open-source license like Apache, X11, or even GPLv2) Solaris and
disclaim their patents over ZFS. But, it's evident that the purchase was
really about controlling Sun's patents and choking MySQL to uselessness.

I actually disagree about that last part. Oracle has worked *hard* on
MySQL, and has fixed performance and scalability problems and
storage-engine feature disparities that it has had for many years that
were just languishing under Monty Widenius. (Though mysqldump still
needs to die and rot. The mydumper project is what mysqldump SHOULD
have evolved into ten years ago, but didn't.) Also, Oracle is finally
within reach of killing the doddering MyISAM storage engine.

Oracle buying Sun (and thereby MySQL) was also the boot up the arse that
it took to get Monty moving on MariaDB. (I still think he's smoking
crack with his Aria engine though. It's transactional, wait, no, it's
not, I mean it CAN be ... *make up your damn mind*. The age of
non-transactional storage engines should be over by now.)

[Jonathan Patschke]

Post by Phil Stracchino

But, it's evident that the purchase was really about controlling Sun's
patents and choking MySQL to uselessness.

I actually disagree about that last part. Oracle has worked *hard* on
MySQL, and has fixed performance and scalability problems and
storage-engine feature disparities that it has had for many years that
were just languishing under Monty Widenius.

I'd argue that those fixes are merely the natural fallout of developing
the closed-source "enterprise edition," which allowed them to capitalize
on the age of "LAMP" within the market segment of companies that like
paying for useless support.

Otherwise, when compared to the strides that PostgreSQL made in the same
period of time, one can only puzzle at how the _much_ more popular
codebase advanced by so little.

[Phil Stracchino]

Post by Jonathan Patschke

Post by Phil Stracchino
I actually disagree about that last part. Oracle has worked *hard* on
MySQL, and has fixed performance and scalability problems and
storage-engine feature disparities that it has had for many years that
were just languishing under Monty Widenius.

I'd argue that those fixes are merely the natural fallout of developing
the closed-source "enterprise edition," which allowed them to capitalize
on the age of "LAMP" within the market segment of companies that like
paying for useless support.

While it's true that a lot of Oracle's work has been into proprietary
extensions such as the Enterprise Query Analyzer, pretty much
*everything* they have fixed or enhanced in the core of mysqld itself
has gone straight into Community Edition. They have actually made huge
improvements that just aren't immediately visible if you don't use MySQL
on a daily basis. They've added features in MySQL that have made my
employer's senior Oracle DBAs ask plaintively, "Why don't we have that
in Oracle?" (Preheating the InnoDB buffer pool on startup, for example.)

Buying an Enterprise MySQL license doesn't actually get you a different
MySQL engine, it just gets you support and the proprietary tools. (And
in Oracle's [gasp] defense there, I have to say I've used the Enterprise
Monitor with Query Analyzer, and it's pretty damned shiny.)

Post by Jonathan Patschke
Otherwise, when compared to the strides that PostgreSQL made in the same
period of time, one can only puzzle at how the _much_ more popular
codebase advanced by so little.

At the risk of being somewhat catty, I /could/ suggest that PostgreSQL
had a lot more room for improvement to start with. And they can't do
anything about some of the fundamental design decisions that make me
shake my head and wonder what the &#$(&$@%!$@! they were thinking.
PostgreSQL's origin as a conglomeration of separate CS students' senior
projects still shows badly.

[Toby Thain]

Post by Phil Stracchino
...
PostgreSQL's origin as a conglomeration of separate CS students' senior
projects still shows badly.

If I wanted to be catty I might point out that MySQL's origins and early
design decisions aren't very glamorous either.

(Disclaimer: I've been a heavy user of MySQL. But the lack of
transactional DDL *alone* might soon make me stop using it.)

--Toby
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan Patschke]

Post by Phil Stracchino

Post by Jonathan Patschke
Otherwise, when compared to the strides that PostgreSQL made in the same
period of time, one can only puzzle at how the _much_ more popular
codebase advanced by so little.

At the risk of being somewhat catty, I /could/ suggest that PostgreSQL
had a lot more room for improvement to start with.

To be sure! The rowsize limits in the early releases were crippling.

Post by Phil Stracchino
And they can't do anything about some of the fundamental design
they were thinking.

They were thinking, "Barring a hardware failure, We do not want the user
to be able to accidentally lose data; in the service of this, nearly any
inconvenience is permissible."

[Phil Stracchino]

Post by Jonathan Patschke
They were thinking, "Barring a hardware failure, We do not want the user
to be able to accidentally lose data; in the service of this, nearly any
inconvenience is permissible."

Which is not an area in which MyISAM can claim any laurels...


To be fair to MyISAM, one of its principal design constraints was that
it needed to work *acceptably well* on a small server shared with other
applications, at a time when a _large_ server was one that might have an
entire 32MB of RAM.

[Andrew M Hoerter]

Post by Phil Stracchino

Post by Jonathan Patschke
They were thinking, "Barring a hardware failure, We do not want the user
to be able to accidentally lose data; in the service of this, nearly any
inconvenience is permissible."

Which is not an area in which MyISAM can claim any laurels...
To be fair to MyISAM, one of its principal design constraints was that
it needed to work *acceptably well* on a small server shared with other
applications, at a time when a _large_ server was one that might have an
entire 32MB of RAM.

To me at least, reliability and "ACIDity" are preconditions for being a
traditional RDBMS, not a design consideration that can be traded off in
favor of something else. Admittedly MySQL has improved over time, but
certain qualities in software (like security) can be very difficult to
retrofit later if you didn't adopt the right philosophy from the very
beginning. See also: Linux containers.

Of course, people with different priorities can make whatever tradeoffs
they want -- but hopefully they're educated about the risks and know
what they're doing. I suspect that's rarely true of people who pick a
database engine based on microbenchmarks.

Also, I'd mention SQLite as an obvious counterexample against needing to
sacrifice reliability for performance or small footprint. Ok, that's a
somewhat unfair comparison when taking into account the relative feature
sets, but I don't think it's *too* far off the mark.
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jerry Kemp]

I'm always disappointed when one of these rumors comes out, and people from this
list jump on the band wagon, buy it all, hook, line and sinker. Start stating
how sad it is that Solaris is now gone.

This literally just happened less than a year ago from some place called
"thelayoff.com". Turned out that Oracle decided that they were jumping on the
same OS naming convention that m$ windows and Apple Mac OS X was doing. But
Solaris was "dead" then also.

And ultimately, didn't anyone see Alan Coopersmith, head X11 guy, post on the
OpenIndiana list just a couple of days ago?

......................................................
Discussion list for OpenIndiana <openindiana-***@openindiana.org>
From: Alan Coopersmith <***@oracle.com>
Message-ID: <d5af3d37-eb20-cab7-4df2-***@oracle.com>
Date: Sun, 3 Sep 2017 21:51:28 -0700
User-Agent: Mozilla/5.0 (X11; SunOS i86pc; rv:52.0) Gecko/20100101
Thunderbird/52.2.0

Subject: Re: [OpenIndiana-discuss] So is Solaris officially dead?


No.
--
-Alan Coopersmith- ***@oracle.com
Oracle Solaris Engineering - https://blogs.oracle.com/alanc

......................................................


The IT world is a crazy place, and sure, anything can happen. I will believe
its dead when I hear it from Oracle. I'm sad for anyone who gets laid off and
looses their job. The Register is just a "trade rag" with mixed accuracy at best.


Jerry

Post by Liam Proven
I presume everyone's aware of this?
https://www.theregister.co.uk/2017/09/04/oracle_layoffs_solaris_sparc_teams/
Oracle has laid off ~2,500 Solaris and SPARC developers, leaving a
tiny remnant, presumably to fulfil some support contracts.
Apart from the forks -- Illumos, SmartOS etc. -- that's it. It appears
that Solaris is dead. Sad times.
--
Liam Proven b" Profile: https://about.me/liamproven
Twitter/Facebook/Flickr: lproven b" Skype/LinkedIn/AIM/Yahoo: liamproven
UK: [phone] b" DR/WhatsApp/Telegram/Signal: [phone]
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue

[Jonathan Katz]

Post by Jerry Kemp
I'm always disappointed when one of these rumors comes out, and people from
this list jump on the band wagon, buy it all, hook, line and sinker. Start
stating how sad it is that Solaris is now gone.

I think this time it's different:

http://dtrace.org/blogs/bmc/2017/09/04/the-sudden-death-and-eternal-life-of-solaris/
_______________________________________________
rescue list - http://www.sunhelp.org/mailman/listinfo/rescue