Re: [PATCHES] Win32 Event log

> -----Original Message-----
> From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
> Sent: 13 August 2004 03:59
> To: Dave Page
> Cc: Andrew Dunstan; pgsql-hackers-win32
> Subject: Re: [pgsql-hackers-win32] [PATCHES] Win32 Event log
>
> (http://www.serverwatch.com/tutorials/article.php/10825_2239651_3 for
> > further analyis). Therefore, attempts to create tablespaces will
> > probably cause a crash.
>
> Can't we fix such a _hypothetical_ crash when someone finally
> reports it?
> In fact I bet we aren't going to be able to prevent folks
> from running on NT4 anyway so we will have to fix it.

On further examination of Andreas' patch, it doesn't use an APIs that
aren't in NT4 as far as I can see, instead it uses existing functions
(such as DeviceIoControl), but with control codes like
FSCTL_SET_REPARSE_POINT which are *not* in NT. As far as I can tell
without spending time I don't have on it, symlink creation will fail
gracefully and as long as the create tablespace code can handle that, it
should be pretty clean.

> If the installer can't run on NT4 SP4, that is an installer
> issue.  I don't see why we should label the base source code
> as not working on it.

Because the vast majority of Windows users *will not* compile from
source, they will use the installer. If press releases go out saying it
supports NT4 and we do get a ton of downloads from NT users, we will end
up with a correspopnding flood of emails and bug reports. End users will
not see the difference between the source, and the precompiled binary,
even if we do see them as separate projects.

> If it doesn't have symlinks, fine, no tablespaces.   I will update the
> release notes to mention the installer doesn't run on NT4.
>
> Basically, we can't turn the PostgreSQL ship on a dime. If we
> come out and say NT4 doesn't work, the someone else appears
> and gets it working, it is harder to communicate that.  I
> don't want to be adding/removing
> NT4 based on untested ideas.

The installer issue is not untested - I got a bug report about it this
morning. The table spaces issue is, but as no-one here seems to have NT4
we can't test it. We also cannot reasonably support a platform that none
of us have access to.

By all means note that the base code will compile and run without
tablespaces on NT4, but please do not announce that it's an official
supported platform.

Regards, Dave.