Re: Tablespaces

From: Marko Karppinen <marko(at)karppinen(dot)fi>
To: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Tablespaces
Date: 2004-03-08 00:07:35
Message-ID: 9A79CD2E-7094-11D8-8995-000A95C56374@karppinen.fi
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-hackers-win32

On 3 March 2004, at 19:52, Bruce Momjian wrote:
> The advantage of symlinks is that an administrator could see how things
> are laid out from the command line.

One thing to keep in mind is that system administrators don't see
symlinks as being informational -- they see them as the actual UI
for the redirection in question. So their expectation is that they'll
be able to move the actual directory around at will (as long as they
update the symlink to match).

If symlinks are used, the rule of least surprise would mean that
no information whatsoever about the physical location of a tablespace
should be stored in the system catalogs. Otherwise their relationship
with the information stored in the symlink is ambiguous.

mk

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Matthew T. O'Connor 2004-03-08 00:48:01 Re: 7.4.2 ... all commits in?
Previous Message Jan Wieck 2004-03-07 23:48:37 Re: Slony-I makes progress

Browse pgsql-hackers-win32 by date

  From Date Subject
Next Message Claudio Natoli 2004-03-08 09:52:11 Re: Win32 regression test status
Previous Message Claudio Natoli 2004-03-07 23:28:29 Re: APC + socket restrictions under Win32?