Re: 9.0.4 Data corruption issue

2011/7/20 Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
> Ken Caruso <ken(at)ipl31(dot)net> writes:
>> On Sun, Jul 17, 2011 at 3:04 AM, Cédric Villemain <
>> cedric(dot)villemain(dot)debian(at)gmail(dot)com> wrote:
>>>> Block number 12125253 is bigger that any block we can find in
>>>> base/2651908/652397108.1
>
>>> Should the table size be in the 100GB range or 2-3 GB range ?
>
>> The DB was in the 200GB-300GB range when this happened.
>
> Cédric was asking about the particular table's size, not the whole DB...
> the table in question is the one with relfilenode = 652397108.

Yes, the error origin is probably due to faillure outside PostgreSQL.
It should not happen but... well I was wondering how a table can be,
by error, partially truncated (truncation of segment, not the sql
TRUNCATE) by PostgreSQL...

--
Cédric Villemain +33 (0)6 20 30 22 52
http://2ndQuadrant.fr/
PostgreSQL: Support 24x7 - Développement, Expertise et Formation