Re: [HACKERS] Tablespaces

From: "Lawrence E(dot) Smithmier, Jr(dot)" <Larry(at)Smithmier(dot)com>
To: tswan(at)idigx(dot)com
Cc: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, tswan(at)idigx(dot)com, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Zeugswetter Andreas SB SD <zeugswettera(at)spardat(dot)at>, Greg Stark <gsstark(at)mit(dot)edu>, pgsql-hackers(at)postgresql(dot)org, PostgreSQL Win32 port list <pgsql-hackers-win32(at)postgresql(dot)org>
Subject: Re: [HACKERS] Tablespaces
Date: 2004-03-04 21:46:03
Message-ID: 1078436763.4047a39bd5861@mail.cbtree.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-hackers-win32

> "Zeugswetter Andreas SB SD" <ZeugswetterA(at)spardat(dot)at> writes:
>>>> My feeling is that we need not support tablespaces on OS's without
>>>> symlinks.
>
>> To create symlinked directories on Win2k NTFS see:
>> http://www.sysinternals.com/ntw2k/source/misc.shtml#junction
>> I think Win2000 or XP would be a reasonable restriction for Win32 PG
>> installations that want tablespaces.
>
> Oh, good --- symlinks for directories are all that we need for this
> design. I think that settles it then.
>

Er, sorry to drop into the middle of this but do you want to cripple a port
before it is even complete? Is there a compelling reason to use symlinks rather
than a flat file? If the issue is just:

> Gavin Sherry <swm(at)linuxworld(dot)com(dot)au> writes:
>how the low-level file access code finds a tablespace.

then what is wrong with using an XML file that is loaded and traversed at start
up? I agree it would be a cool to use the file system as a database, but why
place a possible limiting factor for the sake of elegance? Isn't XML a valid
and accepted way to store hierarchial data?

> Gavin Sherry <swm(at)linuxworld(dot)com(dot)au> writes:
> I am expecting to hear some bleating about this from people whose
> preferred platforms don't support symlinks ;-). However, if we don't

Well bleat I guess. Although I wouldn't exactly say preferred. I prefer to
think of myself as a realest getting paid to program on a platform. A platform
with symlinks carrying quite a bit of baggage. On NTFS they are called Junction
Points and are a special type of Reparse Point. One thing I noticed on the
Microsoft site regarding these:

(http://www.microsoft.com/whdc/DDK/IFSkit/reparse.mspx)
>Reparse Points are a powerful feature of Windows 2000 (not available on Windows
>NT® 4.0), but developers should be aware that there can only be one reparse
>point per file, and some new Windows 2000 mechanisms use reparse points (HSM,
>Native Structured Storage). Developers need to have fallback strategies for
>when the reparse point tag is already in use for a file.

makes me question their usefulness at this point. I am currently exploring
another solution to the problem that caused me to investigate them.

Well, thanks for your time. I guess I can go baaack to lurking now. ;-)

Lawrence E. Smithmier, Jr.
MCP, MCAD
(919) 522-9738
Larry(at)Smithmier(dot)com

Browse pgsql-hackers by date

  From Date Subject
Next Message Thomas Hallgren 2004-03-04 21:57:10 Minutes from Pl/Java - next step IRC
Previous Message Tom Lane 2004-03-04 21:04:05 Re: A plan to improve error messages with context, hint and details.

Browse pgsql-hackers-win32 by date

  From Date Subject
Next Message Dann Corbit 2004-03-04 22:47:24 Another crack at doing a Win32 build under MINGW
Previous Message Andrew Dunstan 2004-03-04 21:05:00 Re: [HACKERS] Tablespaces