Re: PostgreSQL still for Linux only?

I don't want to split hairs, but wasn't Firebird originally Interbase? If
so, you'll find it was originally a *nix product before it was a Windows
database (back in the Ashton-Tate days for those with long memories).

Edwin New.

-----Original Message-----
From: Uwe C. Schroeder [mailto:uwe(at)oss4u(dot)com]
Sent: Wednesday, 9 March 2005 3:49 PM
To: pgsql-general(at)postgresql(dot)org
Subject: Re: [GENERAL] PostgreSQL still for Linux only?

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tuesday 08 March 2005 07:24 pm, Tope Akinniyi wrote:
> Hi,
>
> I am wondering at this display of extreme Linux mentality being displayed
> by the 'top bras' of the PostgreSQL community. And I ask, are we
> encouraging Windows use of PostgreSQL at all?
>
> Take a look at tools being rolled out at PgFoundry on daily basis; all for
> Linux except the Windows installer. I ask myself what is being done to
> encourage PostgreSQL Windows users. Nothing is available to them except
> the Database and PgAdmin. No replication tool, no this, no that.

To be honest - I wouldn't encourage the use of PostgreSQL on Win.
Neither would I for any database or data warehouse application (which
probably
is why SAP put onto their website that they prefer linux to windows
platforms).
I think it could even damage the quite good reputation of PostgreSQL - if
your
windows box crashes and takes the DB with it - most likely it's not the
fault
of a lousy OS, nor the fault of an incompetent sysadmin who forgot to make
backups - it will be this "shitty" free database system that's to blame.

I wrote quite some software that uses postgresql - never would I tell any
customer that he could now run it on windows. As a matter of fact I put code

like:

if os="win" {
errormessage("this software is not ported to windows yet");
exit(99);
}

into the startup routine - just to make it impossible for the customer to
run
it on windows.

> I was troubled when CommandPrompt, the leading Windows support provider
> responded to a post that their plPHP is for Linux only.
>
> Sorry for this: Firebird provides equal tools for Linux and Windows
users.
> We are not the one to tell the Windows users whether they need them.

Firebird was a DOS ISAM DB. It just made it's way to *nix a couple years
ago.

> Whether Windows is bad or good; Linux is the angel and Windows the devil
is
> not the issue here. PostgreSQL has gone the Windows way and must not be
> shown to be deficient.

The problem is, that it's a question of perception. Most windows fans don't
see that "their" OS is pretty instable. So it's not a question if the
community can do anything to make PostgreSQL look deficient - it's a
question
of what people do with it on Win. I had a similar case recently with a
customer: His MS Office suite crashed at least 3 times a day. So I switched
him to OpenOffice. Now OO crashed once after a month of perfect operation -
guess what, the customer is back to MS Office because OO crashed on him and
MS has this new version that's sooo much better. Call it dumb - but that's
how a lot of people are. Well, he paid a couple $k to get new licenses and
is
back where he was a month ago.

> I am not holding anybody responsible, but I think we need to do a massive
> re-orientation of the community not to carry the Linux-Windows game too
> far.

It's just a fact: any unix is a better platform for databases than windows.
Windows was designed (and mostly still is) as a Desktop operating system -
and it's fairly good on the desktop. Never trust a server that needs a mouse

attached to operate properly. Unix was designed with scalability, stability
and multiuser-operation in mind - and that's what it's good at. I wouldn't
want my payroll on a windows box - much less my company data.

UC

- --
Open Source Solutions 4U, LLC 2570 Fleetwood Drive
Phone: +1 650 872 2425 San Bruno, CA 94066
Cell: +1 650 302 2405 United States
Fax: +1 650 872 2417
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFCLoAijqGXBvRToM4RAu4ZAJ9Ed1kgGzNaFmVCgJSfZS1kAkm9HACfZ5bI
rSX4FvU1RxHR63sg6icE+gU=
=+NPW
-----END PGP SIGNATURE-----

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo(at)postgresql(dot)org

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thinking about it you may be right. I guess I'm misstaking it for something
else. Too many "foxes" out here nowadays :-)

To the topic: I don't argue the benefit of a native windows version from a
marketing point of view (although not so from a technical point of view). As
long as MS hasn't filed a chapter 11 the rest of the world will have to deal
with them. Therefor a native windows version is possibly the only way to make
postgresql more popular and sneak it into the one or other fortune 500
company.

On Tuesday 08 March 2005 09:02 pm, Edwin New wrote:
> I don't want to split hairs, but wasn't Firebird originally Interbase? If
> so, you'll find it was originally a *nix product before it was a Windows
> database (back in the Ashton-Tate days for those with long memories).
>
> Edwin New.
>
> -----Original Message-----
> From: Uwe C. Schroeder [mailto:uwe(at)oss4u(dot)com]
> Sent: Wednesday, 9 March 2005 3:49 PM
> To: pgsql-general(at)postgresql(dot)org
> Subject: Re: [GENERAL] PostgreSQL still for Linux only?
>
> On Tuesday 08 March 2005 07:24 pm, Tope Akinniyi wrote:
> > Hi,
> >
> > I am wondering at this display of extreme Linux mentality being displayed
> > by the 'top bras' of the PostgreSQL community. And I ask, are we
> > encouraging Windows use of PostgreSQL at all?
> >
> > Take a look at tools being rolled out at PgFoundry on daily basis; all
> > for Linux except the Windows installer. I ask myself what is being done
> > to encourage PostgreSQL Windows users. Nothing is available to them
> > except the Database and PgAdmin. No replication tool, no this, no that.
>
> To be honest - I wouldn't encourage the use of PostgreSQL on Win.
> Neither would I for any database or data warehouse application (which
> probably
> is why SAP put onto their website that they prefer linux to windows
> platforms).
> I think it could even damage the quite good reputation of PostgreSQL - if
> your
> windows box crashes and takes the DB with it - most likely it's not the
> fault
> of a lousy OS, nor the fault of an incompetent sysadmin who forgot to make
> backups - it will be this "shitty" free database system that's to blame.
>
> I wrote quite some software that uses postgresql - never would I tell any
> customer that he could now run it on windows. As a matter of fact I put
> code
>
> like:
>
> if os="win" {
> errormessage("this software is not ported to windows yet");
> exit(99);
> }
>
> into the startup routine - just to make it impossible for the customer to
> run
> it on windows.
>
> > I was troubled when CommandPrompt, the leading Windows support provider
> > responded to a post that their plPHP is for Linux only.
> >
> > Sorry for this: Firebird provides equal tools for Linux and Windows
>
> users.
>
> > We are not the one to tell the Windows users whether they need them.
>
> Firebird was a DOS ISAM DB. It just made it's way to *nix a couple years
> ago.
>
> > Whether Windows is bad or good; Linux is the angel and Windows the devil
>
> is
>
> > not the issue here. PostgreSQL has gone the Windows way and must not be
> > shown to be deficient.
>
> The problem is, that it's a question of perception. Most windows fans don't
> see that "their" OS is pretty instable. So it's not a question if the
> community can do anything to make PostgreSQL look deficient - it's a
> question
> of what people do with it on Win. I had a similar case recently with a
> customer: His MS Office suite crashed at least 3 times a day. So I switched
> him to OpenOffice. Now OO crashed once after a month of perfect operation -
> guess what, the customer is back to MS Office because OO crashed on him and
> MS has this new version that's sooo much better. Call it dumb - but that's
> how a lot of people are. Well, he paid a couple $k to get new licenses and
> is
> back where he was a month ago.
>
> > I am not holding anybody responsible, but I think we need to do a massive
> > re-orientation of the community not to carry the Linux-Windows game too
> > far.
>
> It's just a fact: any unix is a better platform for databases than windows.
> Windows was designed (and mostly still is) as a Desktop operating system -
> and it's fairly good on the desktop. Never trust a server that needs a
> mouse
>
> attached to operate properly. Unix was designed with scalability, stability
> and multiuser-operation in mind - and that's what it's good at. I wouldn't
> want my payroll on a windows box - much less my company data.
>
> UC

- --
UC

- --
Open Source Solutions 4U, LLC 2570 Fleetwood Drive
Phone: +1 650 872 2425 San Bruno, CA 94066
Cell: +1 650 302 2405 United States
Fax: +1 650 872 2417
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFCLpKajqGXBvRToM4RAjb7AJ96fllQAqY6g6y3XxBzRi682+BvAgCg0XWx
/a9Y4VNCmPUlZQ+xlj1ZmJw=
=cHVW
-----END PGP SIGNATURE-----

On Wed, 9 Mar 2005 16:02:46 +1100, Edwin New <edwin_new(at)toll(dot)com(dot)au> wrote:
>
>
> I don't want to split hairs, but wasn't Firebird originally Interbase? If
> so, you'll find it was originally a *nix product before it was a Windows
> database (back in the Ashton-Tate days for those with long memories).

"InterBase started on Apollo Domain, a spectacularly wonderful
workstation with terrific networking. The initial release supported
Apollo, Sun, HP/UX, VAX/VMS, Ultrix, and something else that escapes
me. So, if you wonder 'was InterBase originally a Windows/DOS
system?', the answer is 'no'."

From: http://firebird.sourceforge.net/index.php?op=history&id=ann_2

(This page: http://firebird.sourceforge.net/index.php?op=history&id=ann_1
says also: "InterBase started in the shower." Maybe the "something
else that escapes me" was NetBSD? ;-)

Ian Barwick

--- "Uwe C. Schroeder" <uwe(at)oss4u(dot)com> wrote:

> > The problem is, that it's a question of
> perception. Most windows fans don't
> > see that "their" OS is pretty instable.

That may have been true in 1995. However, in this day
and age most Windows fans don't see that their OS as
unstable because it isn't - unless of course you are
referring to the non-NT variations.

Regards,

Shelby Cain

__________________________________
Celebrate Yahoo!'s 10th Birthday!
Yahoo! Netrospective: 100 Moments of the Web
http://birthday.yahoo.com/netrospective/

Personally, I find the anti-windows bias that has been shown in this
thread by some developers to be disappointing. Maybe it sucks to program
in, and maybe it's not as stable as unix (though I don't put much water
in that argument anymore), but the fact is there's still a LOT of places
that are windows shops and a LOT of people who use windows more heavily
than *nix. More important, the egotism of "If you want to use PostgreSQL
you better run it on what we tell you to run it on" is certain to turn
people off of PostgreSQL. It will certainly turn off windows developers
who might have been interested in working to improve PostgreSQL now that
it runs on windows.
--
Jim C. Nasby, Database Consultant decibel(at)decibel(dot)org
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"

Shelby Cain wrote:

>--- "Uwe C. Schroeder" <uwe(at)oss4u(dot)com> wrote:
>
>
>
>>>The problem is, that it's a question of
>>>
>>>
>>perception. Most windows fans don't
>>
>>
>>>see that "their" OS is pretty instable.
>>>
>>>
>
>That may have been true in 1995. However, in this day
>and age most Windows fans don't see that their OS as
>unstable because it isn't - unless of course you are
>referring to the non-NT variations.
>
>
O.k. I don't want to start an OS war here. However
there are a couple of things I know.

1. As of Windows 2000, Windows is reasonably stable.
However there is a caveat, it still can not perform
under load (read slowness, possible crash) like Linux
or other UNIX variants can.

2. As of Windows 2003, Windows is very stable and
performs fairly well under load. However it still
can not keep up with Linux or other UNIX variants.

The majority of the problem with Windows in these
days is people who hire other people with little
pieces of paper that say they are knowledgeable.

A properly managed Windows server can be reliable,
can perform reasonably well, if you have the expertise
to do so. This is not that much unlike UNIX. The difference
is that UNIX requires the expertise, Windows makes you
feel like you have it when you don't.

Sincerely,

Joshua D. Drake

>Regards,
>
>Shelby Cain
>
>
>
>
>__________________________________
>Celebrate Yahoo!'s 10th Birthday!
>Yahoo! Netrospective: 100 Moments of the Web
>http://birthday.yahoo.com/netrospective/
>
>---------------------------(end of broadcast)---------------------------
>TIP 5: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/docs/faq
>
>

--
Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
Postgresql support, programming shared hosting and dedicated hosting.
+1-503-667-4564 - jd(at)commandprompt(dot)com - http://www.commandprompt.com
PostgreSQL Replicator -- production quality replication for PostgreSQL

On Wed, 9 Mar 2005 11:02:10 -0600, Jim C. Nasby <decibel(at)decibel(dot)org> wrote:
>... but the fact is there's still a LOT of places
> that are windows shops and a LOT of people who use windows more heavily
> than *nix. More important, the egotism of "If you want to use PostgreSQL
> you better run it on what we tell you to run it on" is certain to turn
> people off of PostgreSQL.

Perhaps someone on the list who knows and uses the different operating
systems could set up a lab, to compare PostgreSQL between them.
Perhaps the latest Windows Server, a popular distribution of Linux,
and Mac OS X?

Has this already been done, with regard to performance?

Doug

Ho ho, flame on! :)

My completely annecodal experience with devs which prefer windows over
posix is that the former prods things until they seem to work and accepts
unexplained behavior far more readily than the latter. Do I *really* want
that kind of mentality in my database devs?

Anyway, I think you have the focus wrong. It's not: "run our software on
what we tell you to".... it's more: "we believe this platform is better
than others, so we'll write our free software for that. But if you want to
port it over to the platform of your choice, have fun doing that."

On Wed, 9 Mar 2005, Jim C. Nasby wrote:

> Personally, I find the anti-windows bias that has been shown in this
> thread by some developers to be disappointing. Maybe it sucks to program
> in, and maybe it's not as stable as unix (though I don't put much water
> in that argument anymore), but the fact is there's still a LOT of places
> that are windows shops and a LOT of people who use windows more heavily
> than *nix. More important, the egotism of "If you want to use PostgreSQL
> you better run it on what we tell you to run it on" is certain to turn
> people off of PostgreSQL. It will certainly turn off windows developers
> who might have been interested in working to improve PostgreSQL now that
> it runs on windows.
> --
> Jim C. Nasby, Database Consultant decibel(at)decibel(dot)org
> Give your computer some brain candy! www.distributed.net Team #1828
>
> Windows: "Where do you want to go today?"
> Linux: "Where do you want to go tomorrow?"
> FreeBSD: "Are you guys coming, or what?"
>
> ---------------------------(end of broadcast)---------------------------
> TIP 7: don't forget to increase your free space map settings
>

Le mercredi 09 mars 2005 à 09:47 -0800, Ben a écrit :
> Ho ho, flame on! :)

Hear hear!!! This man is a troll if ever we have seen one.

> > Personally, I find the anti-windows bias that has been shown in this
> > thread by some developers to be disappointing. Maybe it sucks to program
> > in, and maybe it's not as stable as unix (though I don't put much water
> > in that argument anymore), but the fact is there's still a LOT of places
> > that are windows shops and a LOT of people who use windows more heavily
> > than *nix. More important, the egotism of "If you want to use PostgreSQL
> > you better run it on what we tell you to run it on" is certain to turn
> > people off of PostgreSQL. It will certainly turn off windows developers
> > who might have been interested in working to improve PostgreSQL now that
> > it runs on windows.

Excuse me dear sir. There seems to be about 97% of the world that runs
Windows that does not give you permission to be rude to a tiny minority
who just happen to have written an insanely great database that runs
quite nicely on their "hobby" OSs as well as the crap you call home. If
you aren't pleased with the postgresql support on Windows don't use
it!!! That is your freedom. Ours is to think (maybe wrongly) that it is
much better running it on the BSDs and Linux of our choice. That is our
freedom.

There is nothing egoist about developing a great database server on an
OS with a tiny user base. The egoists are elsewhere dear sir, far from
the free software developers, in the closed source world. The code is
there, it is free - go and improve it. Maybe you need a dictionary to
look up the word egoist?

Please go and troll over at MySQL. They have a Windows version too and
maybe a lot more time and patience for rude people such as yourself.

Tony

On Wed, 9 Mar 2005, tony wrote:

> Le mercredi 09 mars 2005 à 09:47 -0800, Ben a écrit :
> > Ho ho, flame on! :)
>
> Hear hear!!! This man is a troll if ever we have seen one.

Who? Jim Nasby? He's made several helpful posts to this list in my
memory, and I'm sure an archive search would turn up a lot more.

tony wrote:
> Excuse me dear sir. There seems to be about 97% of the world that runs
> Windows that does not give you permission to be rude to a tiny minority
> who just happen to have written an insanely great database that runs
> quite nicely on their "hobby" OSs as well as the crap you call home. If
> you aren't pleased with the postgresql support on Windows don't use
> it!!! That is your freedom. Ours is to think (maybe wrongly) that it is
> much better running it on the BSDs and Linux of our choice. That is our
> freedom.
>
> There is nothing egoist about developing a great database server on an
> OS with a tiny user base. The egoists are elsewhere dear sir, far from
> the free software developers, in the closed source world. The code is
> there, it is free - go and improve it. Maybe you need a dictionary to
> look up the word egoist?
>
> Please go and troll over at MySQL. They have a Windows version too and
> maybe a lot more time and patience for rude people such as yourself.
>
> Tony

This thread is getting a bit carried away, don't you think? If this keeps up, these fora run the
risk of turning into the gong show that the c.d.oracle.* newsgroup frequently becomes.
If you think it's a legitimate flame, why not ignore it, instead of adding to the noise?

Joshua,

Very well put !

Cheers,

Aly.

Joshua D. Drake wrote:
> Shelby Cain wrote:
>
>> --- "Uwe C. Schroeder" <uwe(at)oss4u(dot)com> wrote:
>>
>>
>>
>>>> The problem is, that it's a question of
>>>>
>>>
>>> perception. Most windows fans don't
>>>
>>>
>>>> see that "their" OS is pretty instable.
>>>>
>>
>>
>> That may have been true in 1995. However, in this day
>> and age most Windows fans don't see that their OS as
>> unstable because it isn't - unless of course you are
>> referring to the non-NT variations.
>>
>>
> O.k. I don't want to start an OS war here. However
> there are a couple of things I know.
>
> 1. As of Windows 2000, Windows is reasonably stable.
> However there is a caveat, it still can not perform
> under load (read slowness, possible crash) like Linux
> or other UNIX variants can.
>
> 2. As of Windows 2003, Windows is very stable and
> performs fairly well under load. However it still
> can not keep up with Linux or other UNIX variants.
>
> The majority of the problem with Windows in these
> days is people who hire other people with little
> pieces of paper that say they are knowledgeable.
>
> A properly managed Windows server can be reliable,
> can perform reasonably well, if you have the expertise
> to do so. This is not that much unlike UNIX. The difference
> is that UNIX requires the expertise, Windows makes you
> feel like you have it when you don't.
>
> Sincerely,
>
> Joshua D. Drake
>
>
>
>
>
>> Regards,
>>
>> Shelby Cain
>>
>>
>>
>>
>> __________________________________ Celebrate Yahoo!'s 10th Birthday!
>> Yahoo! Netrospective: 100 Moments of the Web
>> http://birthday.yahoo.com/netrospective/
>>
>> ---------------------------(end of broadcast)---------------------------
>> TIP 5: Have you checked our extensive FAQ?
>>
>> http://www.postgresql.org/docs/faq
>>
>>
>
>
>
> ------------------------------------------------------------------------
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster

--
Aly Dharshi
aly(dot)dharshi(at)telus(dot)net

"A good speech is like a good dress
that's short enough to be interesting
and long enough to cover the subject"

Quoting "Joshua D. Drake" <jd(at)commandprompt(dot)com>:

> Shelby Cain wrote:
>
> >--- "Uwe C. Schroeder" <uwe(at)oss4u(dot)com> wrote:
> >
> >
> >
> >>>The problem is, that it's a question of
> >>>
> >>>
> >>perception. Most windows fans don't
> >>
> >>
> >>>see that "their" OS is pretty instable.
> >>>
> >>>
> >
> >That may have been true in 1995. However, in this day
> >and age most Windows fans don't see that their OS as
> >unstable because it isn't - unless of course you are
> >referring to the non-NT variations.
> >
> >
> O.k. I don't want to start an OS war here. However
> there are a couple of things I know.
>
> 1. As of Windows 2000, Windows is reasonably stable.
> However there is a caveat, it still can not perform
> under load (read slowness, possible crash) like Linux
> or other UNIX variants can.
>
> 2. As of Windows 2003, Windows is very stable and
> performs fairly well under load. However it still
> can not keep up with Linux or other UNIX variants.
>
> The majority of the problem with Windows in these
> days is people who hire other people with little
> pieces of paper that say they are knowledgeable.
>
> A properly managed Windows server can be reliable,
> can perform reasonably well, if you have the expertise
> to do so. This is not that much unlike UNIX. The difference
> is that UNIX requires the expertise, Windows makes you
> feel like you have it when you don't.
>
> Sincerely,
>
> Joshua D. Drake
>
>
>
>
>
> >Regards,
> >
> >Shelby Cain
> >
> >
> >
> >
> >__________________________________
> >Celebrate Yahoo!'s 10th Birthday!
> >Yahoo! Netrospective: 100 Moments of the Web
> >http://birthday.yahoo.com/netrospective/
> >
> >---------------------------(end of broadcast)---------------------------
> >TIP 5: Have you checked our extensive FAQ?
> >
> > http://www.postgresql.org/docs/faq
> >
> >
>
>
> --
> Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
> Postgresql support, programming shared hosting and dedicated hosting.
> +1-503-667-4564 - jd(at)commandprompt(dot)com - http://www.commandprompt.com
> PostgreSQL Replicator -- production quality replication for PostgreSQL
>
>

The only additional thing I would add to this if it hasn't been mentioned
already is that 2000 had/has some major security issues and even though 2003 is
more secure out of the box from what I've experienced so far, I would **never**
trust a windows box to anything other than my LAN using private IP blocks and if
it has inbound access via a public IP then it would more certainly be behind
another firewall that is NAT'ing/Port Forwarding its traffic.

--
Keith C. Perry, MS E.E.
Director of Networks & Applications
VCSN, Inc.
http://vcsn.com

____________________________________
This email account is being host by:
VCSN, Inc : http://vcsn.com

>
>
>The only additional thing I would add to this if it hasn't been mentioned
>already is that 2000 had/has some major security issues and even though 2003 is
>more secure out of the box from what I've experienced so far, I would **never**
>trust a windows box to anything other than my LAN using private IP blocks and if
>it has inbound access via a public IP then it would more certainly be behind
>another firewall that is NAT'ing/Port Forwarding its traffic.
>
>
Nobody should ever put a server regardless of OS on a public IP.
It should always be firewalled/Nat/Port Forwarding.

Sincerely,

Joshua D. Drake

--
Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
Postgresql support, programming shared hosting and dedicated hosting.
+1-503-667-4564 - jd(at)commandprompt(dot)com - http://www.commandprompt.com
PostgreSQL Replicator -- production quality replication for PostgreSQL

Quoting "Joshua D. Drake" <jd(at)commandprompt(dot)com>:

> >
> >
> >The only additional thing I would add to this if it hasn't been mentioned
> >already is that 2000 had/has some major security issues and even though 2003
> is
> >more secure out of the box from what I've experienced so far, I would
> **never**
> >trust a windows box to anything other than my LAN using private IP blocks
> and if
> >it has inbound access via a public IP then it would more certainly be
> behind
> >another firewall that is NAT'ing/Port Forwarding its traffic.
> >
> >
> Nobody should ever put a server regardless of OS on a public IP.
> It should always be firewalled/Nat/Port Forwarding.
>
> Sincerely,
>
> Joshua D. Drake
>
>
>
>
> --
> Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
> Postgresql support, programming shared hosting and dedicated hosting.
> +1-503-667-4564 - jd(at)commandprompt(dot)com - http://www.commandprompt.com
> PostgreSQL Replicator -- production quality replication for PostgreSQL
>
>

As with all things technology there is an art too it as well- several ways
to do things. I don't, for instance, NAT/Port forward public interfaces for
Linux hosts because in my experience they can be hardened without much ambiguity
to be placed there. Similarly, I don't feel the same is true with most of the
windows variants so for security sake increased an network complexity is justified.

My point is that along with the performance issues this thread has point out,
data security is another reason to consider a non-windows platform to run your
production database.

--
Keith C. Perry, MS E.E.
Director of Networks & Applications
VCSN, Inc.
http://vcsn.com

____________________________________
This email account is being host by:
VCSN, Inc : http://vcsn.com

Jim C. Nasby wrote:

>Personally, I find the anti-windows bias that has been shown in this
>thread by some developers to be disappointing. Maybe it sucks to program
>in, and maybe it's not as stable as unix (though I don't put much water
>in that argument anymore), but the fact is there's still a LOT of places
>that are windows shops and a LOT of people who use windows more heavily
>than *nix. More important, the egotism of "If you want to use PostgreSQL
>you better run it on what we tell you to run it on" is certain to turn
>people off of PostgreSQL. It will certainly turn off windows developers
>who might have been interested in working to improve PostgreSQL now that
>it runs on windows.
>
>
Ok--- I will admit to a anti-Windows bias. But at least my bias is
informed. In addition to my former employment at Microsoft, I have
studies both types of OS's in detail. Here are some specific comments I
would make:

1) I do not expect PostgreSQL to *ever* perform as well on Windows as
it does on Linux. This is primarily due to the fundamentally different
emphasis in kernel architecture between UNIX-style and VMS-style
operating systems. Windows server applications which are process-based
are always likely to underperform. Windows applications ported to Linux
are similarly likely to underperform.

2) Windows stability is getting far better, but does still lag behind
that of Linux.

3) I think that it is very likely that you might be legally required to
get CAL's for Windows Server in order to allow the systems to access
PostgreSQL. While this is not enforced by the OS, I don't know whether
the EULA requires it (my guess is that it does).

PostgreSQL on Windows has 2 uses. It is for developers to play around
with, and it is for smaller businesses with few connections to use. One
you need to scale, you will probably have to go to Linux, BSD, etc.

Best Wishes,
Chris Travers

Joshua D. Drake wrote:
> Nobody should ever put a server regardless of OS on a public IP.
> It should always be firewalled/Nat/Port Forwarding.

Except for the firewall/Nat server, of course :D

--
Alban Hertroys

While we run PostgreSQL on Free-BSD for our production systems, we have
'demo' laptop windows XP systems that contain the entire server
architecture (application server, database, win32 client, etc). Sure is
handy to be able to run PostgreSQL on windows and not have to change
anything......

> PostgreSQL on Windows has 2 uses. It is for developers to play around
> with, and it is for smaller businesses with few connections to use.
> One you need to scale, you will probably have to go to Linux, BSD, etc.

On Wed, Mar 09, 2005 at 11:39:53AM -0600, Doug Hall wrote:
> On Wed, 9 Mar 2005 11:02:10 -0600, Jim C. Nasby <decibel(at)decibel(dot)org> wrote:
> >... but the fact is there's still a LOT of places
> > that are windows shops and a LOT of people who use windows more heavily
> > than *nix. More important, the egotism of "If you want to use PostgreSQL
> > you better run it on what we tell you to run it on" is certain to turn
> > people off of PostgreSQL.
>
> Perhaps someone on the list who knows and uses the different operating
> systems could set up a lab, to compare PostgreSQL between them.
> Perhaps the latest Windows Server, a popular distribution of Linux,
> and Mac OS X?
>
> Has this already been done, with regard to performance?

There is a perftest project on either pgfoundry or gborg that has been
doing performance testing. I think it's all being done on linux right
now, but it would certainly be interesting to compare linux, freebsd,
and windows. Unfortunately, there's no way to do an apples-to-apples
(pun intended) comparison with OS X, since not all of the OS's will run
on the same hardware. Linux will run on Power, though, as will OpenBSD.
I think FreeBSD's support is still pretty bare, but I'm not certain.
--
Jim C. Nasby, Database Consultant decibel(at)decibel(dot)org
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"

On Wed, Mar 09, 2005 at 09:47:17AM -0800, Ben wrote:
> Ho ho, flame on! :)
>
> My completely annecodal experience with devs which prefer windows over
> posix is that the former prods things until they seem to work and accepts
> unexplained behavior far more readily than the latter. Do I *really* want
> that kind of mentality in my database devs?

Of course not, and I don't think there's any risk of this happening. Are
you aware that every patch submitted for inclusion goes through a code
review? It's very insightful to see the discussion and mentality on the
-hackers list; data integrity is always the absolute number 1 priority.
Anyone who wants to code for PostgreSQL who doesn't share that priority
won't last long at all.

> Anyway, I think you have the focus wrong. It's not: "run our software on
> what we tell you to".... it's more: "we believe this platform is better
> than others, so we'll write our free software for that. But if you want to
> port it over to the platform of your choice, have fun doing that."

With the attitude of "Windows can not be made to reliably run a
database", how many developers do you think will be attracted?
--
Jim C. Nasby, Database Consultant decibel(at)decibel(dot)org
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"

On Wed, Mar 09, 2005 at 05:51:43PM -0800, Chris Travers wrote:
> Jim C. Nasby wrote:
> Ok--- I will admit to a anti-Windows bias. But at least my bias is
> informed. In addition to my former employment at Microsoft, I have
> studies both types of OS's in detail. Here are some specific comments I
> would make:
>
> 1) I do not expect PostgreSQL to *ever* perform as well on Windows as
> it does on Linux. This is primarily due to the fundamentally different
> emphasis in kernel architecture between UNIX-style and VMS-style
> operating systems. Windows server applications which are process-based
> are always likely to underperform. Windows applications ported to Linux
> are similarly likely to underperform.

This is akin to saying that an application written to use MySQL will
never perform well on PostgreSQL. It depends on *how* the code is
written. If your SQL is tuned to one database, it will likely have
performance issues on other databases. Likewise, a process-based server
will perform poorly on Windows, while a threaded server will not. This
is an implimentation choice. There's no reason why PostgreSQL on windows
*has* to be process based (though of course there would be serious
technical issues with changing it).

Of course, by simply hand waving and saying "it can never be better", it
never will be better.
--
Jim C. Nasby, Database Consultant decibel(at)decibel(dot)org
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"

* Jim C. Nasby <decibel(at)decibel(dot)org> [0336 21:36]:

> With the attitude of "Windows can not be made to reliably run a
> database", how many developers do you think will be attracted?

People are entitled to an opinion, and in many cases its formed from
experience. I think it's unrealistic to expect a large team of programmers who
have been using *NIX to think windows is equally good.
If they did, they'd run on it, right?

The process model is presumably there because for 90% of platforms it makes
sense to do it that way. No-one is going to object to a well-written thread
based postmaster, but it's expecting a bit much for it to spring into life
off the bat.

To me a database is a service, like a dns or dhcp server, and wanting to
put it on windoms is like wanting to run BIND or IPF on there.

For most people it's going to be easier to stick a linux on a dedicated box
and run postgresql on that. I don't see what the problem is with that.

Just to be clear:

I have no interest or opinion in windows, microsoft or anything else that
makes slashdotters jump up and down beyound playing civ3 on it.
You like it, that's great.

The one thing the world does'nt need is another 'my os can beat up your os'
thread.

--
'That question was less stupid; though you asked it in a profoundly stupid way.'
-- Prof. Farnsworth
Rasputin :: Jack of All Trades - Master of Nuns

Quoth decibel(at)decibel(dot)org ("Jim C. Nasby"):
> With the attitude of "Windows can not be made to reliably run a
> database", how many developers do you think will be attracted?

That remains to be seen.

I wouldn't consider it the least bit worthwhile to try to evaluate it
now, as what is happening now is that "WinFolk" are getting their very
first exposure to the software.

It would seem surprising for new developers to emerge from the
Windows(tm) population before at least 6 months have passed.

The way developers emerge is that users come along, work with the
software for a while, and discover things that "itch" them the wrong
way. They have become sufficiently committed that it is worth putting
a little effort into scratching some of the itches. That starts
getting them into understanding the code a little better, allowing
them to subsequently scratch deeper itches.
--
output = reverse("moc.liamg" "@" "enworbbc")
http://linuxdatabases.info/info/nonrdbms.html
Share and Enjoy!!

That remains to be seen.

I wouldn't consider it the least bit worthwhile to try to evaluate it
now, as what is happening now is that "WinFolk" are getting their very
first exposure to the software.

It would seem surprising for new developers to emerge from the
Windows(tm) population before at least 6 months have passed.

The way developers emerge is that users come along, work with the
software for a while, and discover things that "itch" them the wrong
way. They have become sufficiently committed that it is worth putting
a little effort into scratching some of the itches. That starts
getting them into understanding the code a little better, allowing
them to subsequently scratch deeper itches.
--
output = reverse("moc.liamg" "@" "enworbbc")
http://linuxdatabases.info/info/nonrdbms.html
Share and Enjoy!!