Re: 7.3.4 on Linux: UPDATE .. foo=foo+1 degrades massivly - Mailing list pgsql-general

From Philipp Buehler
Subject Re: 7.3.4 on Linux: UPDATE .. foo=foo+1 degrades massivly
Date
Msg-id 20040423053536.GA21879@srv9.de.buehler.net
Whole thread Raw
In response to Re: 7.3.4 on Linux: UPDATE .. foo=foo+1 degrades massivly  (Guy Fraser <guy@incentre.net>)
Responses Re: 7.3.4 on Linux: UPDATE .. foo=foo+1 degrades massivly
List pgsql-general
On 22/04/2004, Guy Fraser <guy@incentre.net> wrote To pgsql-general@postgresql.org:
> >Shouldn't the Database server be the entity that decides when vacuum is
> >needed?
>
> How is the database supposed to know when you want to purge records?
> Once a vacuum has been run, the table can not be rolled back or time
> traveled.

Hmm, if the UPDATE is in a transaction block. After this block is
committed, the deleted tuple could be purged if there is a flag. Like,
either in the schema 'purge-on-commit', or as an option like 'UPDATE PURGE'?

Just an idea.. (Brainfart?) :)

ciao
--
Philipp Buehler, aka fips | <double-p>

cvs -d /dev/myself commit -m "it's my life" dont/you/forget

pgsql-general by date:

Previous
From: Deepakkumar.Vasudevan@cybernetsoft.com
Date:
Subject: Downloading Windows Version of PGSQL
Next
From: "Jerry Robertson"
Date:
Subject: Question