Patch availability change
ji...@specsol.spam.sux.com
now contract only?
Could this be part of a plan by Oracle to prod people into paying for
support contracts?
--
Jim Pennino
Remove .spam.sux to reply.
John D Groenveld
<ji...@specsol.spam.sux.com> wrote:
>Has anyone else noticed that a lot of patches that used to be public are
>now contract only?
I have not noticed, but which patch IDs are you having trouble
fetching?
John
groe...@acm.org
Chris Ridd
The recommended patch clusters for Solaris 10 x86 were not publicly
accessible when I looked a week or so back.
--
Chris
Tim Bradshaw
> Has anyone else noticed that a lot of patches that used to be public are
> now contract only?
There were changes to patch availability some time ago, which were
mostly enforcing rules which were not previously being enforced. In
some cases I think things are still available without a contract *but*
you now need a sunsolve login (or whatever it is called now) to get
them.
I don't think it's unreasonable for Sun/Oracle to try and make money
from support for the software they give away for free, actually. That
kind of is the OSS business model isn't it (when it's not "Collect
underpants", "?", "Profit")?
John D Groenveld
Chris Ridd <chri...@mac.com> wrote:
>The recommended patch clusters for Solaris 10 x86 were not publicly
>accessible when I looked a week or so back.
My Sun Online Account is bound to contracts so I can't test,
but the last time someone here complained they were able to
get those patches if not the tarball with Martin Paul's PCA.
<URL:http://www.par.univie.ac.at/solaris/pca/>
John
groe...@acm.org
ji...@specsol.spam.sux.com
> On 2010-02-10 17:41:40 +0000, ji...@specsol.spam.sux.com said:
>
>> Has anyone else noticed that a lot of patches that used to be public are
>> now contract only?
>
> There were changes to patch availability some time ago, which were
> mostly enforcing rules which were not previously being enforced. In
> some cases I think things are still available without a contract *but*
> you now need a sunsolve login (or whatever it is called now) to get
> them.
Yeah, I know and have a sunsolve login.
This just started happening this week.
> I don't think it's unreasonable for Sun/Oracle to try and make money
> from support for the software they give away for free, actually. That
> kind of is the OSS business model isn't it (when it's not "Collect
> underpants", "?", "Profit")?
Yeah, I understand that but there needs to be a reasonable balance for
patches.
For example, if patches for something that could cause the system to halt
at 3AM aren't readily available, they are shooting themselves in the foot.
Also, the current support plans offered seem like something out of the
big iron mainframe days.
Why, for example, does the number of CPU's matter for OS updates?
IMHO if there was a simple OS support plan that allowed for filing bug
reports electronically, patch download, no handholding, and was on the
order of $100/yr, I would think they would sell a LOT of them.
Chris Ridd
Using pca+patchdiag.xref I was only able to get earlier versions of
patches marked as recommended and/or security. Typically I could only
get version n-1, when version n was current according to
pca+patchdiag.xref.
This was consistent with navigating Sunsolve's patch finder in Firefox.
n-1 was public, but n was not.
I assume there's a sane relationship between the versions of
recommended patches and the recommended patch cluster? I couldn't get
the recommended patch cluster to find out!
--
Chris
Ian Collins
> Tim Bradshaw <t...@tfeb.org> wrote:
>
>> I don't think it's unreasonable for Sun/Oracle to try and make money
>> from support for the software they give away for free, actually. That
>> kind of is the OSS business model isn't it (when it's not "Collect
>> underpants", "?", "Profit")?
>
> Yeah, I understand that but there needs to be a reasonable balance for
> patches.
>
> For example, if patches for something that could cause the system to halt
> at 3AM aren't readily available, they are shooting themselves in the foot.
If I had a system that had to be running at 3AM, I'd make sure I had a
support contract!
> Also, the current support plans offered seem like something out of the
> big iron mainframe days.
>
> Why, for example, does the number of CPU's matter for OS updates?
>
> IMHO if there was a simple OS support plan that allowed for filing bug
> reports electronically, patch download, no handholding, and was on the
> order of $100/yr, I would think they would sell a LOT of them.
That's what they used to have, the basic plan was $120 a year. I agree
with your comment, I had a plan then, I don't now.
--
Ian Collins
ji...@specsol.spam.sux.com
> ji...@specsol.spam.sux.com wrote:
>> Tim Bradshaw <t...@tfeb.org> wrote:
>>
>>> I don't think it's unreasonable for Sun/Oracle to try and make money
>>> from support for the software they give away for free, actually. That
>>> kind of is the OSS business model isn't it (when it's not "Collect
>>> underpants", "?", "Profit")?
>>
>> Yeah, I understand that but there needs to be a reasonable balance for
>> patches.
>>
>> For example, if patches for something that could cause the system to halt
>> at 3AM aren't readily available, they are shooting themselves in the foot.
>
> If I had a system that had to be running at 3AM, I'd make sure I had a
> support contract!
I have only seen one Unix system in the past 30 years or so that didn't
run 24/7.
Whether or not they "had" to run is another issue, but at the minimum
most places do backups if nothing else in the wee hours.
>> Also, the current support plans offered seem like something out of the
>> big iron mainframe days.
>>
>> Why, for example, does the number of CPU's matter for OS updates?
>>
>> IMHO if there was a simple OS support plan that allowed for filing bug
>> reports electronically, patch download, no handholding, and was on the
>> order of $100/yr, I would think they would sell a LOT of them.
>
> That's what they used to have, the basic plan was $120 a year. I agree
> with your comment, I had a plan then, I don't now.
Yeah, me too as well as several other people/organizations I know of.
Richard B. Gilbert
> ji...@specsol.spam.sux.com wrote:
>> Tim Bradshaw <t...@tfeb.org> wrote:
>>
>>> I don't think it's unreasonable for Sun/Oracle to try and make money
>>> from support for the software they give away for free, actually.
>>> That kind of is the OSS business model isn't it (when it's not
>>> "Collect underpants", "?", "Profit")?
>>
>> Yeah, I understand that but there needs to be a reasonable balance for
>> patches.
>>
>> For example, if patches for something that could cause the system to halt
>> at 3AM aren't readily available, they are shooting themselves in the
>> foot.
>
> If I had a system that had to be running at 3AM, I'd make sure I had a
> support contract!
>
Any system that HAS to be available needs a support contract. It may
even need a clone to take over in case the original becomes seriously
ill! Probably needs one or more RAID arrays as well.
>> Also, the current support plans offered seem like something out of the
>> big iron mainframe days.
>>
>> Why, for example, does the number of CPU's matter for OS updates?
>>
>> IMHO if there was a simple OS support plan that allowed for filing bug
>> reports electronically, patch download, no handholding, and was on the
>> order of $100/yr, I would think they would sell a LOT of them.
>
> That's what they used to have, the basic plan was $120 a year. I agree
> with your comment, I had a plan then, I don't now.
>
I believe it's still available but as of the last time I heard it cost
more money. The last figure I'm sure of was $240 per year and I believe
that there has been at least one increase in that rate.
More money buys you more, or faster, service! If you want both more and
faster be ready to cough up BIG bucks!!!
Volker Borchert
> IMHO if there was a simple OS support plan that allowed for filing bug
> reports electronically, patch download, no handholding, and was on the
> order of $100/yr, I would think they would sell a LOT of them.
I think download-only for 30$ would be more attractive.
(Would that be tax deductible?)
--
"I'm a doctor, not a mechanic." Dr Leonard McCoy <mc...@ncc1701.starfleet.fed>
"I'm a mechanic, not a doctor." Volker Borchert <v_bor...@despammed.com>
Ian Collins
> Ian Collins wrote: a LOT of them.
>>
>> That's what they used to have, the basic plan was $120 a year. I
>> agree with your comment, I had a plan then, I don't now.
>>
>
> I believe it's still available but as of the last time I heard it cost
> more money. The last figure I'm sure of was $240 per year and I believe
> that there has been at least one increase in that rate.
>
> More money buys you more, or faster, service! If you want both more and
> faster be ready to cough up BIG bucks!!!
True, but I don't think the current $240 plan offers much, if anything,
more than the old $120 one.
--
Ian Collins
John D Groenveld
Ian Collins <ian-...@hotmail.com> wrote:
>> IMHO if there was a simple OS support plan that allowed for filing bug
>> reports electronically, patch download, no handholding, and was on the
>> order of $100/yr, I would think they would sell a LOT of them.
>
>That's what they used to have, the basic plan was $120 a year. I agree
>with your comment, I had a plan then, I don't now.
I suspect most of that market got onto Solaris Express and Indiana
trains.
John
groe...@acm.org
Richard B. Gilbert
AFAIK the only difference is the price. I could be wrong. As a
hobbyist I have NO support from Sun or anyone else. If I find something
that doesn't work, I'll either fix it myself or learn to live without it.
ji...@specsol.spam.sux.com
The basic plan seems to be $324/year, but isn't available for more
than 2 sockets.
That means to get all patches for my old V440 with 4 CPU's, I'd have to
go to $1440/year.
So what does the number of CPU's in a V440 have to do with OS patches?
At least they don't use a per core basis so my 4 core X86 machine could
get the $324/year plan.
John D Groenveld
Chris Ridd <chri...@mac.com> wrote:
>Using pca+patchdiag.xref I was only able to get earlier versions of
>patches marked as recommended and/or security. Typically I could only
>get version n-1, when version n was current according to
>pca+patchdiag.xref.
PatchIDs?
John
groe...@acm.org
Richard B. Gilbert
> ji...@specsol.spam.sux.com wrote:
>> IMHO if there was a simple OS support plan that allowed for filing bug
>> reports electronically, patch download, no handholding, and was on the
>> order of $100/yr, I would think they would sell a LOT of them.
>
> I think download-only for 30$ would be more attractive.
> (Would that be tax deductible?)
>
Tax deductable? If the hardware and software are used solely for
business purposes they are both deductible.
I have seen bug reports filed here and have seen them acknowledged by
Sun employees. Unless a paying customer complains, these bugs will be
fixed when someone gets around to it. Depending on the seriousness of
the problem and the difficulty of the fix such a bug report might be
acted on "same day" or when all other bugs have been fixed.
You want it now? PAY!!!!!
Ian Collins
>
> The basic plan seems to be $324/year, but isn't available for more
> than 2 sockets.
>
> That means to get all patches for my old V440 with 4 CPU's, I'd have to
> go to $1440/year.
>
> So what does the number of CPU's in a V440 have to do with OS patches?
Well, you have a high end system! Support pricing (and not just Sun's)
has always been based on CPU (now core) count. I have a client who pays
through the nose for database support because they run it on an old 8
processor v880. They could get lower support costs and better
performance on a newer, smaller, system but the pain of migrating from
Solaris 8 it too high.
> At least they don't use a per core basis so my 4 core X86 machine could
> get the $324/year plan.
Ah, a toy system....
--
Ian Collins
ufo
> >> Has anyone else noticed that a lot of patches that used to be public
> >> are now contract only?
> >
> > There were changes to patch availability some time ago, which were
> > mostly enforcing rules which were not previously being enforced. In
> > some cases I think things are still available without a contract
> > *but* you now need a sunsolve login (or whatever it is called now) to
> > get them.
>
> Yeah, I know and have a sunsolve login.
>
> This just started happening this week.
I have a contract, but was unable to download patches via pca about two
or three weeks ago. A note somewhere on SunSolve's web page told me to
login and somehow renew the account in order to be able to download
patches again using pca (can't recall the details). From there on pca
worked again. Some hassle from the takeover I guess.
Uwe
ji...@specsol.spam.sux.com
> ji...@specsol.spam.sux.com wrote:
>>
>> The basic plan seems to be $324/year, but isn't available for more
>> than 2 sockets.
>>
>> That means to get all patches for my old V440 with 4 CPU's, I'd have to
>> go to $1440/year.
>>
>> So what does the number of CPU's in a V440 have to do with OS patches?
>
> Well, you have a high end system! Support pricing (and not just Sun's)
> has always been based on CPU (now core) count. I have a client who pays
> through the nose for database support because they run it on an old 8
> processor v880. They could get lower support costs and better
> performance on a newer, smaller, system but the pain of migrating from
> Solaris 8 it too high.
A V440 wasn't a "high end system" even when new.
>> At least they don't use a per core basis so my 4 core X86 machine could
>> get the $324/year plan.
>
> Ah, a toy system....
Not according to the Sun glossy brochure's...
Chris Ridd
The only one I can find right now is 119964-19; patchdiag.xref lists
-20 as being current, and Sunsolve's patchfinder list thinks they're
all available to me. Guess I tried to get -20 slightly too early?
Interestingly the corresponding patch for SPARC is 119963, and *no*
versions of that are accessible without support contracts. Clearly
SPARC users are being treated differently from x86 users.
Neither of these are in the recommended patch clusters according to the
cluster READMEs.
--
Chris
Martin Paul
> Has anyone else noticed that a lot of patches that used to be public are
> now contract only?
>
> Could this be part of a plan by Oracle to prod people into paying for
> support contracts?
Get ready for some bad news: According to my sources there will be no
more free access to any Solaris patch, be it security or not. A support
contract will be required for every patch download, no matter whether
it's done interactively through the website or hands-off with wget/pca.
I've been told that this policy change won't by publically announced by
Oracle.
It already seems to be in effect. I cannot download any of these patches
with a Sun Online Account which is not connected to a support contract:
Patch IR CR RSB Age Synopsis
------ -- - -- --- --- -------------------------------------------------
119254 52 < 72 RS- 22 SunOS 5.10: Install and Patch Utilities Patch
119255 -- < 72 RS- 22 SunOS 5.10_x86: Install and Patch Utilities Patch
125332 08 = 08 RS- 13 JDS 3: Macromedia Flash Player Plugin Patch
125333 -- < 08 RS- 13 JDS 3_x86: Macromedia Flash Player Plugin Patch
141505 -- < 07 RS- 8 SunOS 5.10_x86: ipf patch
141506 -- < 07 RS- 8 SunOS 5.10: ipf patch
Sooner or later access to patchdiag.xref and all other content currently
on SunSolve will be limited to paying customers as well.
No opinion, but a few thoughts:
Using an operating system where the only way to fix security problems is
waiting for the next version is a no-go, so the minimum cost for running
Solaris on a real-world system is now that of a basic service plan (324$
and up per year).
Sun didn't make enough money, so it's obvious that Oracle handles things
differently. Whether this is a wise decision is left to the future. The
fact that there's no public announcement reveals a lot.
There should be a much simpler way to get basic patch support (think
about a PayPal button at the end of the OS installation to get the
idea). And it definitely should cost less than the service plans.
Personally, I'd prefer a one-time fee for the OS (which should be
included in the price of a server when bought from Sun).
Martin.
--
SysAdmin | Institute of Scientific Computing, University of Vienna
PCA | Analyze, download and install patches for Solaris
| http://www.par.univie.ac.at/solaris/pca/
John D Groenveld
Chris Ridd <chri...@mac.com> wrote:
>Interestingly the corresponding patch for SPARC is 119963, and *no*
>versions of that are accessible without support contracts. Clearly
>SPARC users are being treated differently from x86 users.
Not sure which BugID in 119964 qualified it as a free patch under
the old policy and whether the bug applies to SPARC.
John
groe...@acm.org
John D Groenveld
Martin Paul <m...@par.univie.ac.at> wrote:
>Get ready for some bad news: According to my sources there will be no
>more free access to any Solaris patch, be it security or not. A support
Thanks for the heads-up.
<URL:http://sunsolve.sun.com/search/document.do?assetkey=1-61-203648-1>
| A support contract covering Solaris is required to access,
| download, install and use all Solaris Software Updates. This
| applies to all Solaris releases.
John
groe...@acm.org
Chris Ridd
Hm. Does the word "Update" there refer to major Solaris updates (eg
Solaris 10 update 9) or patches?
--
Chris
John D Groenveld
Chris Ridd <chri...@mac.com> wrote:
>Hm. Does the word "Update" there refer to major Solaris updates (eg
>Solaris 10 update 9) or patches?
Patches.
Solaris 10 October 2009 is an "update release":
<URL:http://www.sun.com/software/solaris/releases.jsp>
John
groe...@acm.org
David Kirkby
> Get ready for some bad news: According to my sources there will be no
> more free access to any Solaris patch, be it security or not.
I think a failure to offer security fixes would be a major obstacle to
making Solaris more popular. For companies on a tight budget, Linux
would seem more attractive then. Any arguments by IT professionals
that Solaris is a more stable system would be met with arguments like
"yes, and when a major security flaw is found, you have to pay to have
it fixed".
Buy a cheap PC from Tesco (in the UK) and it comes with Microsoft
Windows. All updates for that are free for life. I could buy a Windows
PC with free updates for life for less than I could buy a Sun
maintenance contract.
Solaris has a small enough market share now. Making security fixes
chargeable would decrease that.
FWIW, I tried to download a kernel update, which is marked as
recommended and security using pca last night. I was unable to do so,
even though my Sunsolve account is tied to a contract. I then logged
into Sunsolve directly with a browser, and was able to download it.
For non-security fixes, then I can see why updates are chargeable.
Dave
John D Groenveld
David Kirkby <drki...@gmail.com> wrote:
>I think a failure to offer security fixes would be a major obstacle to
>making Solaris more popular. For companies on a tight budget, Linux
I think the Solaris marketing wonks see Indiana as the volume
operating system for developers, content creators, early adopters
and those who don't want to buy support contracts.
>would seem more attractive then. Any arguments by IT professionals
>that Solaris is a more stable system would be met with arguments like
>"yes, and when a major security flaw is found, you have to pay to have
>it fixed".
I only have access to one Redhat Enterprise Linux installation and
up2date is bound to an annual support subscription.
Are RHEL security fixes available without a subscription or does
one need to run Fedora?
Or are you referring to a different Linux distributor?
>FWIW, I tried to download a kernel update, which is marked as
>recommended and security using pca last night. I was unable to do so,
>even though my Sunsolve account is tied to a contract. I then logged
>into Sunsolve directly with a browser, and was able to download it.
You had to periodically run pca with --update=check as Sun
would break the "wget" interfaces.
$ ./pca --update=check
No new version available
$ ./pca --version
pca 20091216-02
John
groe...@acm.org
Sami Ketola
> FWIW, I tried to download a kernel update, which is marked as
> recommended and security using pca last night. I was unable to do so,
> even though my Sunsolve account is tied to a contract. I then logged
> into Sunsolve directly with a browser, and was able to download it.
Did you forget to enable wget patch access on your sunsolve account?
Sami
Martin Paul
> FWIW, I tried to download a kernel update, which is marked as
> recommended and security using pca last night. I was unable to do so,
> even though my Sunsolve account is tied to a contract. I then logged
> into Sunsolve directly with a browser, and was able to download it.
As John says - often updating to the latest release of pca fixes
download problems.
As somebody else already mentioned on the thread, it might be necessary
to accept the new Software License from Oracle to make hands-off
downloads work again:
http://www.mail-archive.com/p...@lists.univie.ac.at/msg01583.html
On the other hand, a different problem showed up for me this morning:
SunSolve has forgotten about my support contract, it's not listed on the
"Update Account" subpage on sunsolve.sun.com anymore. Re-adding it
doesn't work either, even though there's no error. I'm not alone with that:
http://www.mail-archive.com/p...@lists.univie.ac.at/msg01612.html
David Kirkby
> David Kirkby wrote:
> > FWIW, I tried to download a kernel update, which is marked as
> > recommended and security using pca last night. I was unable to do so,
> > even though my Sunsolve account is tied to a contract. I then logged
> > into Sunsolve directly with a browser, and was able to download it.
>
> As John says - often updating to the latest release of pca fixes
> download problems.
I have the latest PCA
> As somebody else already mentioned on the thread, it might be necessary
> to accept the new Software License from Oracle to make hands-off
> downloads work again:
I had *not* done that, but I can easily do it.
> Martin.
But I'd also like to know exactly what patches are publicly available.
I've tried to build an open-source software package (Sage) on Solaris
10 03/2005. It fails, despite it works on a patched version of
Solaris. It would be nice to know exactly what patches are available
for Solaris. I just created a new account on sunsolve, purposely to
determine what someone without an account can get. I've not managed to
find anything so far, but perhaps there are things.
I want to get an old version of Solaris, fully patched, then see if
this software will build. If not, the minimum requirements for the
installation of Sage will either to be a later version of Solaris, or
a maintenance contract. That will be a shame, but might be
inevitable.
I think Oracle could be shooting themselves in the foot by applying
too many restrictions in the availability of patches. Let's be honest,
a lot of the software in Solaris is open-source. Oracle gain from the
development of open-source software.
It's anyones guess if Oracle will kill off Solaris to those without a
maintenance contract and it going back to being chargeable, like it
was on Solaris 7 and earlier.
Those without a contract would be well advised to make sure they have
the latest Solaris 10, and keep about 10 copies of it in different
places. I would not be surprised if the next release of Solaris 10
needs a contract.
Dave
David Kirkby
wrote:
I had not done that, but will do.
Martin Paul
> But I'd also like to know exactly what patches are publicly available.
As it seems, the question is quickly answered with the new policy: None.
> I want to get an old version of Solaris, fully patched, then see if
> this software will build. If not, the minimum requirements for the
> installation of Sage will either to be a later version of Solaris, or
> a maintenance contract. That will be a shame, but might be
> inevitable.
I think stating a certain release of Solaris (e.g. "Solaris 10 8/07")
will be the easiest solution, both for the maintainer and the
prospective user.
The alternative - a possibly long list of patches - can be really
painful to verify. Patches have dependencies and might pull in multiple
kernel patches, patches get obsoleted by other patches which either the
maintainer or the user has to resolve, etc. If it's just one certain
patch which is required, listing that might be better though, as
upgrading to a newer release sometimes isn't possible (despite Live
Upgrade).
David Kirkby
> In article <c871baa8-653a-43c0-ab44-463cd9be7...@q16g2000yqq.googlegroups.com>,
> David Kirkby <drkir...@gmail.com> wrote:
>
> >I think a failure to offer security fixes would be a major obstacle to
> >making Solaris more popular. For companies on a tight budget, Linux
>
> I think the Solaris marketing wonks see Indiana as the volume
> operating system for developers, content creators, early adopters
> and those who don't want to buy support contracts.
That's quite possibly true, but on SPARC hardware, that is not really
very viable.
In any case, I rekon at least some people that use Open Solaris now
might think twice about it in future. I suspect Oracle could be
killing that off
A lot of people buy old SPARCs and run Solaris. Killing that off,
would IMHO, decrease the already small update of Solaris.
> >would seem more attractive then. Any arguments by IT professionals
> >that Solaris is a more stable system would be met with arguments like
> >"yes, and when a major security flaw is found, you have to pay to have
> >it fixed".
>
> I only have access to one Redhat Enterprise Linux installation and
> up2date is bound to an annual support subscription.
> Are RHEL security fixes available without a subscription or does
> one need to run Fedora?
> Or are you referring to a different Linux distributor?
There are plenty of free linux distros - Ubunta, Debian etc. For
startup companies on a tight budget, they are looking more attractive
than Solaris. Hence those companies are likely to go with Linux rather
than Solaris.
I also note that patches (apart from hardware patches) will not be
available to those with a hardware only warranty. So despite my Ultra
27 is only a few months old, I would be unable to get the Solaris 10
patches for the operating system which came with it - Solaris 10. It
so happens that I've removed Solaris 10 and put Open Solaris on it.
I'm sure when I bought that Ultra 27, I probably signed something
which gave me very limited access to software updates. It would be
interesting to see if these warranty terms are legally enforceable in
the UK. I'm not a laywer, but from what I understand of UK law, you
can't take someone legal rights away from them. So for example, when
you go to a car part, and see big notices that "cars left at owners
risk", that is not legally enforceable The car park owners have a duty
of care, and sticking notices up like that does not remove that duty
of care.
I happen to know someone who parked his car on a hospital car park -
he worked for the hospital. How he managed to get a car parking place
in Central London I do not know, as he was a fairly junior member of
admin staff. Anyway, his car got broke into, and he tried to claim
from the hospital (i.e. his employer). They declined to pay up, so he
took them to court, arguing they had a duty of care. The hospital paid
up before it went to court.
Selling someone hardware or software, then failing to fix defects
might not be legally enforcable in the UK - irrespective of whatever
someone might have signed.
Dave
JKB
Re: Patch availability change,
David Kirkby ?crivait dans comp.unix.solaris :
> On Feb 11, 7:09 pm, groen...@cse.psu.edu (John D Groenveld) wrote:
>> In article <c871baa8-653a-43c0-ab44-463cd9be7...@q16g2000yqq.googlegroups.com>,
>> David Kirkby <drkir...@gmail.com> wrote:
>>
>> >I think a failure to offer security fixes would be a major obstacle to
>> >making Solaris more popular. For companies on a tight budget, Linux
>>
>> I think the Solaris marketing wonks see Indiana as the volume
>> operating system for developers, content creators, early adopters
>> and those who don't want to buy support contracts.
>
> That's quite possibly true, but on SPARC hardware, that is not really
> very viable.
I use sparc and Solaris for a long time (even on an SS20 with four
Hypersparc that has never worked under Solaris...). I'm not sure that
sparc has future with Oracle. I have some servers that run with Solaris,
but I think that next sparc servers I shall buy will run with Linux or
NetBSD, not with Solaris. I have installed Linux on a diskless T1000 and
Linux runs better on T1000 than Solaris 10 !
> In any case, I rekon at least some people that use Open Solaris now
> might think twice about it in future. I suspect Oracle could be
> killing that off
>
> A lot of people buy old SPARCs and run Solaris. Killing that off,
> would IMHO, decrease the already small update of Solaris.
There is no patch available from Solaris 9/sparc for last december.
>> >would seem more attractive then. Any arguments by IT professionals
>> >that Solaris is a more stable system would be met with arguments like
>> >"yes, and when a major security flaw is found, you have to pay to have
>> >it fixed".
>>
>> I only have access to one Redhat Enterprise Linux installation and
>> up2date is bound to an annual support subscription.
>> Are RHEL security fixes available without a subscription or does
>> one need to run Fedora?
>> Or are you referring to a different Linux distributor?
>
> There are plenty of free linux distros - Ubunta, Debian etc. For
> startup companies on a tight budget, they are looking more attractive
> than Solaris. Hence those companies are likely to go with Linux rather
> than Solaris.
I'm not sure that the main problem is budget. If you use for exemple
Linux/Debian, you can apply patches with apt-get update and apt-get
upgrade without rebooting. I use some T1000/T5xxx in disless
configuration and I cannot apply any patches because patchadd is
totaly broken over NFS root file ! I cannot apply patches because a
lot of patches require a reboot... Thus, I'm not sure that Solaris
is a good choice for all new installation even patch policy was not
changed by Oracle.
JKB
--
Le cerveau, c'est un véritable scandale écologique. Il représente 2% de notre
masse corporelle, mais disperse à lui seul 25% de l'énergie que nous
consommons tous les jours.
David Kirkby
> David Kirkby wrote:
> > But I'd also like to know exactly what patches are publicly available.
>
> As it seems, the question is quickly answered with the new policy: None.
>
> > I want to get an old version of Solaris, fully patched, then see if
> > this software will build. If not, the minimum requirements for the
> > installation of Sage will either to be a later version of Solaris, or
> > a maintenance contract. That will be a shame, but might be
> > inevitable.
>
> I think stating a certain release of Solaris (e.g. "Solaris 10 8/07")
> will be the easiest solution, both for the maintainer and the
> prospective user.
Why for the prospective user?
> The alternative - a possibly long list of patches - can be really
> painful to verify.
Yes agreed, but if it was "download the latest patch cluster" it would
have been quite easy. It seems that is unlikely.
> Patches have dependencies and might pull in multiple
> kernel patches, patches get obsoleted by other patches which either the
> maintainer or the user has to resolve, etc. If it's just one certain
> patch which is required, listing that might be better though, as
> upgrading to a newer release sometimes isn't possible (despite Live
> Upgrade).
I was hoping it was just the patch to fix the release of gcc shipped
with 03/05 (patch 123647-04). I'm not sure if that is however true. It
could have been another problem. I want to investigate that. I thnk
123647-02 is actually sufficient, but would need to check that.
If not, it will have to be something that at least makes gcc not
broken. I suppose one could install a binary gcc, but it would be good
if gcc 3.4.3 could be made sufficiently able to build any later gcc
from source.
Dave
Andreas F. Borchert
> On the other hand, a different problem showed up for me this morning:
> SunSolve has forgotten about my support contract, it's not listed on the
> "Update Account" subpage on sunsolve.sun.com anymore. Re-adding it
> doesn't work either, even though there's no error. I'm not alone with that:
>
> http://www.mail-archive.com/p...@lists.univie.ac.at/msg01612.html
I've observed the same problem today. And we hat similar issues in
the past. This gets particularly annoying when due to this calls
cannot be opened and urgent security patches not downloaded.
We are paying customers and we find this painful.
Andreas.
Martin Paul
> On Feb 12, 11:06 am, Martin Paul <m...@par.univie.ac.at> wrote:
>> I think stating a certain release of Solaris (e.g. "Solaris 10 8/07")
>> will be the easiest solution, both for the maintainer and the
>> prospective user.
>
> Why for the prospective user?
Maybe I chose the wrong word (English is not my native language)? What I
meant was "future" or "possible" user. Or just leave off the adjective :)
> I was hoping it was just the patch to fix the release of gcc shipped
> with 03/05 (patch 123647-04).
If it's just that patch that would be great, as it doesn't have any
dependencies.
Richard B. Gilbert
> David Kirkby wrote:
>> On Feb 12, 11:06 am, Martin Paul <m...@par.univie.ac.at> wrote:
>>> I think stating a certain release of Solaris (e.g. "Solaris 10 8/07")
>>> will be the easiest solution, both for the maintainer and the
>>> prospective user.
>>
>> Why for the prospective user?
>
> Maybe I chose the wrong word (English is not my native language)? What I
> meant was "future" or "possible" user. Or just leave off the adjective :)
There's nothing wrong with your English. "prospective" is correct!
pro-spec-tive adj. 1. Likely to happen; expected
2. Likely to become or be: a prospective client
(From "The American Heritage Dictionary" Second College Edition,
Houghton Mifflin Company, Boston 1985)
I wish some of our "native speakers" wrote as well as you do!! A
significant number seem to have slept through spelling and grammar
classes! (I did too but at least I use a spelling checker!)
Tim Bradshaw
> So what does the number of CPU's in a V440 have to do with OS patches?
Per socket licensing is a lot friendlier than per core, which tends to
be the norm.
Tim Bradshaw
> The only one I can find right now is 119964-19; patchdiag.xref lists
> -20 as being current, and Sunsolve's patchfinder list thinks they're
> all available to me. Guess I tried to get -20 slightly too early?
There definitely are occasional lags between patchdiag.xref and what
the patch server actually has.
ji...@specsol.spam.sux.com
Again, what does that have to do with OS patches?
The OS is identical no matter how many CPU's or cores you have.
Tim Bradshaw
> Again, what does that have to do with OS patches?
>
> The OS is identical no matter how many CPU's or cores you have.
So is the RDBMS.
ji...@specsol.spam.sux.com
Yeah, so what?
Tim Bradshaw
> Yeah, so what?
You will almost certainly be paying a license per core (possibly socket
if you are lucky) for the RDBMS. That's just how life is: people with
big machines pay more for their software support as well as their
hardware support. It may not be fair, and it's particularly painful
for hobbyists (generally) who discover they can buy old, many-socket,
systems but can't get a cheap patch-download contract for them.
Unfortunately basically no vendors care about those people. It's also
hard to see how they would go about making exceptions for them - how do
they distinguish between some big organisation with a couple of E4500s
sitting somewhere running some legacy thing, who probably should not be
getting a cheapo license, and me with the same machines sitting in my
garage, which I bought for peanuts and like to play with?
There's actually no real reason they *should* care - I'm not really
helping them any with my vintage hardware, so why should they bother?
The case where they should care is if I *am* helping them - perhaps I'm
a software developer who it would be in their interests to support.
I'm guessing (I don't know, though I think this used to be the case)
that you can get special developer contracts which allow for just this
sort of thing.
ji...@specsol.spam.sux.com
Or in other words, set the pricing by the apparent deepness of the
customer's pocket.
FYI, most businesses don't run out and upgrade their computers when
a new model comes out.
There are still a lot of places running things like the V440 which were
purchased new.
Tim Bradshaw
> Or in other words, set the pricing by the apparent deepness of the
> customer's pocket.
Well, I think more accurately set the pricing to correspond to some
approximation to the "amount of computing power" the system had when it
was new.
(The reason for the "when it was new" is that otherwise old machines
end up being very cheap to support which both discourges people from
upgrading, which is bad for business, and also does not correspond to
the cost of supporting these old systems, which generally goes up over
time.)
>
> FYI, most businesses don't run out and upgrade their computers when
> a new model comes out.
>
> There are still a lot of places running things like the V440 which were
> purchased new.
I'm aware of this, of course: I don't know what the age-curve looks
like but I suspect the median would be 3-5 years old, with a very long
tail (somewhere there will be Sun3s in production use I am sure: fairly
recently someone I know was involved in decomissioning a system which
turned out to be a PDP11). I don;t think it affects the pricing model
particularly: in particular you *don't* want to make old machines very
cheap for commercial use, for the reasons given above.
ji...@specsol.spam.sux.com
> On 2010-02-12 22:13:11 +0000, ji...@specsol.spam.sux.com said:
>
>> Or in other words, set the pricing by the apparent deepness of the
>> customer's pocket.
>
> Well, I think more accurately set the pricing to correspond to some
> approximation to the "amount of computing power" the system had when it
> was new.
Or in other words, set the pricing by the apparent deepness of the
customer's pocket.
> (The reason for the "when it was new" is that otherwise old machines
> end up being very cheap to support which both discourges people from
> upgrading, which is bad for business, and also does not correspond to
> the cost of supporting these old systems, which generally goes up over
> time.)
Nonsense.
The fact that Solaris 10 patches work on SunFire V100's (likely the
cheapest SPARC machine Sun ever made), V440's, as well as the latest M9000's
doesn't add any cost.
And, since the software support contract for an X CPU V440 is the same as
an X CPU T5120, OS support cost is irrelevant to upgrading to the end user.
ji...@specsol.spam.sux.com
> On 2010-02-13 17:15:59 +0000, ji...@specsol.spam.sux.com said:
>
>> The fact that Solaris 10 patches work on SunFire V100's (likely the
>> cheapest SPARC machine Sun ever made), V440's, as well as the latest M9000's
>> doesn't add any cost.
>
> be tested on all the platforms it is intended for. New bugs may also
> be found in old hardware which require patches. Obviously most patches
> do not depend on the hardware platform, but some do, which requires you
> to have instance of that platform around.
No, it doesn't. There is no reason to test patches on all the machines
that have reached EOL, nor much motivation to develop hardware patches
for machines that have reached EOL.
>> And, since the software support contract for an X CPU V440 is the same as
>> an X CPU T5120, OS support cost is irrelevant to upgrading to the end user.
>
> This would be true if a v440 with 4 sockets had the same performance as
> (say) a T5440 with 4 sockets. My guess is it doesn't, and the question
> would really be: how many V440s could you consolidate onto the T5440,
> which will then cost you the same in software support as a single one
> of the systems you've replaced?
A V440 and a T5440 are not the same class machine.
The entry level T5440 is 2 X 8 cores at the same speed (basically) as
a V440, which could replace 4 4 CPU V440's.
The support cost for the T5540 is $324 while the 4 V440's would cost $5,760,
for a savings of $5,436.
However the T5540 start at $40,295.
So for $40,295 you can get the same performance and save $5,436/year on
support and the T5540 will be long past EOL when you break even neglecting
all the associated costs with moving to a different machine.
That is NOT the way most people decide to do upgrades.
-----------
Richard L. Hamilton
You're wrong, flat out. Different SPARC systems that still run Solaris 10
may include different hardware, from bus structure (maybe there's still some old
Sbus stuff still out there) to CPUs (anything from UltraSPARC II on up), to support
chips and peripherals. Everything that needs any extra code, including drivers,
to support it adds to the total amount of code being supported. Further, if specific
hardware is involved, it means keeping some of it around for testing.
The OS's job includes hiding most of those differences from user-land programs,
but that only means it has more to do, not less.
Further, every piece of hardware no longer being sold competes for limited
system programming resources with hardware currently being sold.
From a standpoint of paying for what it costs to produce, the price should go _up_
for hardware no longer in production, and continue to go up until it reaches a level
at which support is dropped.
Mind you, as an individual, I can't afford new or current hardware, period. (well,
I can't justify it, anyway) So I'd certainly be glad to be able to pay a fairly
nominal fee for patches, SunSolve access, and continuing software support for my
older hardware, _not_ including any specific incident support other than a way
of submitting bug reports. That is, something where the cost of providing it is
about the same regardless of how many people there are paying for that lowest level
of support.
And keeping individuals engaged is part of holding onto the commercial market.
So there's probably some balance that falls short of punishing owners of older
equipment to the degree that supporting them contributes to costs.
Richard B. Gilbert
Is it really reasonable to expect Sun to support antiques? In the
computer business three years is "old age". Six years entitles you to
play "unwrap the mummy"!
Yes, a lot of that stuff still works. I have an Ultra 5 and three Ultra
10s. Some of them still work but I'm going to have to find a source of
new "timer chips" or kludge some sort of battery power supply. Support
for these beasts would cost a great deal. The number of people who
would pay for support for machines that old is vanishingly small!
That's why support tends to vanish!
Volker Borchert
> Yes, a lot of that stuff still works. I have an Ultra 5 and three Ultra
> 10s.
Fancy new-fangled stuff, that is.
> The number of people who
> would pay for support for machines that old is vanishingly small!
Depends on the amount requested and availability of CDs/DVDs.
--
"I'm a doctor, not a mechanic." Dr Leonard McCoy <mc...@ncc1701.starfleet.fed>
"I'm a mechanic, not a doctor." Volker Borchert <v_bor...@despammed.com>
ji...@specsol.spam.sux.com
What part of the term "EOL" is it that you don't understand?
ji...@specsol.spam.sux.com
> Is it really reasonable to expect Sun to support antiques? In the
> computer business three years is "old age". Six years entitles you to
> play "unwrap the mummy"!
>
> Yes, a lot of that stuff still works. I have an Ultra 5 and three Ultra
> 10s. Some of them still work but I'm going to have to find a source of
> new "timer chips" or kludge some sort of battery power supply. Support
> for these beasts would cost a great deal. The number of people who
> would pay for support for machines that old is vanishingly small!
> That's why support tends to vanish!
Hardware support, or the lack of it, is a much bigger driving factor
for upgrading old systems than software support.
John D Groenveld
David Kirkby <drki...@gmail.com> wrote:
>That's quite possibly true, but on SPARC hardware, that is not really
>very viable.
Why?
>In any case, I rekon at least some people that use Open Solaris now
>might think twice about it in future. I suspect Oracle could be
>killing that off
Redhat could try to kill Fedora, but I'm not sure how that benefits
them.
Of course, Larry Ellison might see a profit in trying to make
Solaris a niche operating system by ending support for OpenSolaris
that I just don't see.
Is Ed Zander his confindant? Ed doesn't believe in volume.
>A lot of people buy old SPARCs and run Solaris. Killing that off,
>would IMHO, decrease the already small update of Solaris.
Those people will either need to buy a support subscription for
Solaris (or not and just live without patches until the next release
update) or install OpenSolaris (or *BSD or *Linux) and self-support.
>There are plenty of free linux distros - Ubunta, Debian etc. For
>startup companies on a tight budget, they are looking more attractive
>than Solaris. Hence those companies are likely to go with Linux rather
>than Solaris.
What's more attractive about those Linux distros than the
OpenSolaris distros?
I recently installed Ubuntu and Fedora in VirtualBox.
The Ubuntu was the KDE variant which is pretty slick.
I'm looking forward to KDE 4.4 packages.
John
groe...@acm.org
Richard B. Gilbert
> In article <55a52396-3856-4643...@l19g2000yqb.googlegroups.com>,
> David Kirkby <drki...@gmail.com> wrote:
>> That's quite possibly true, but on SPARC hardware, that is not really
>> very viable.
>
> Why?
>
>> In any case, I rekon at least some people that use Open Solaris now
>> might think twice about it in future. I suspect Oracle could be
>> killing that off
>
> Redhat could try to kill Fedora, but I'm not sure how that benefits
> them.
>
> Of course, Larry Ellison might see a profit in trying to make
> Solaris a niche operating system by ending support for OpenSolaris
> that I just don't see.
> Is Ed Zander his confindant? Ed doesn't believe in volume.
>
>> A lot of people buy old SPARCs and run Solaris. Killing that off,
>> would IMHO, decrease the already small update of Solaris.
>
> Those people will either need to buy a support subscription for
> Solaris (or not and just live without patches until the next release
> update) or install OpenSolaris (or *BSD or *Linux) and self-support.
>
That's what I do. Live without patches, I mean. I used to be a
professional system administrator. Now that I'm retired, I keep a
couple of Solaris boxes around along with a Linux box and an OpenVMS
box. I occasionally find a bit of work for one or another. Otherwise
they help keep the house warm. ;-)