Re: big tables with lots-o-rows

From: Sam Barnett-Cormack <s(dot)barnett-cormack(at)lancaster(dot)ac(dot)uk>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Sam Barnett-Cormack <s(dot)barnett-cormack(at)lancaster(dot)ac(dot)uk>, Michiel Lange <michiel(at)minas(dot)demon(dot)nl>, "Chris G(dot) Nicholas" <cgn(at)globexplorer(dot)com>, "" <pgsql-admin(at)postgresql(dot)org>
Subject: Re: big tables with lots-o-rows
Date: 2003-07-01 14:27:27
Message-ID: Pine.LNX.4.50.0307011526160.3744-100000@short.lancs.ac.uk
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

On Tue, 1 Jul 2003, Tom Lane wrote:

> Sam Barnett-Cormack <s(dot)barnett-cormack(at)lancaster(dot)ac(dot)uk> writes:
> > On Tue, 1 Jul 2003, Michiel Lange wrote:
> >> _Could_ it be that you're hitting a filesystem limit here? I am not 100%
> >> certain, but I believe ext2 by default supports only files of 2.? GB at
> >> most... Yet I am not certain if this is really what's wrong, but I would
> >> look in that direction.
>
> > IT's a kernel-level limit, for single files. A kernel recompile with
> > large file support will eliminate that, if it is the problem.
>
> None of that has the slightest relevance to Postgres, however, since we
> always split large tables into gigabyte-sized segment files. (The only
> reason there's a largefile compilation option is so you can work with
> greater-than-2GB dump scripts in pg_dump and pg_restore; the backend
> does not need it.)

I was under the impression that largefile would make the DB store data
in larger files, where appropriate, so using largefile when your kernel
does not support large files could create such an error

> My guess is that the OP ran into an actual out-of-space situation,
> or possibly a disk quota or ulimit limitation that was reported as
> out-of-space.

For example, running out of inodes is also reported as out-of-space. Is
anything else on the same partition?

--

Sam Barnett-Cormack
Software Developer | Student of Physics & Maths
UK Mirror Service (http://www.mirror.ac.uk) | Lancaster University

In response to

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message Derek Main 2003-07-01 14:33:12 zero (o) return code on failure of pg_dump
Previous Message Tom Lane 2003-07-01 14:11:24 Re: big tables with lots-o-rows