The df was made while the error messages occured and the scripts were
running. So it is not a temporary 'mammothean' table.
When I thought of this, two possible elads came to my mind :
- A mistake when I tuned the postgresql.conf file (resource is scarce, as
this test machine is a PIII 700 with 256 MB RAM dating from Q4 2000)
- The default for this particular DB has been set to keep OIDs (I "pour"
several INSERTs sets arising from different files in an auxiliary table, and
keep the last entered entry for any relevant keys, Those entries then go to
a production table).
----- Original Message -----
From: "Stefan Kaltenbrunner" <stefan@kaltenbrunner.cc>
To: "J6M" <j6m@adm.estp.fr>
Cc: "Tom Lane" <tgl@sss.pgh.pa.us>; <pgsql-bugs@postgresql.org>
Sent: Monday, April 07, 2008 8:31 PM
Subject: Re: [BUGS] BUG #4096: PG 8.3.1. confused about remaining disk space
> J6M wrote:
>> OK. So why did this not occur when I was running 8.2.6 ?
>
> I would advise to install some very detailed monitoring on your diskspace
> usage and look for spikes that correlate with your database errors -. I
> have seen this issue with bad queries that are resulting in enormous
> on-disk sorts and running the box temporary out of diskspace.
>
>
> Stefan
>
>
> --
> Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-bugs