Re: Doc limitation update proposal: include out-of-line OID usage per TOAST-ed columns - Mailing list pgsql-hackers

From Nikita Malakhov
Subject Re: Doc limitation update proposal: include out-of-line OID usage per TOAST-ed columns
Date
Msg-id CAN-LCVOKdemAwFrnntGNSpM4WiVHdYUSMcXzWuKiS5q3vvwg2Q@mail.gmail.com
Whole thread Raw
In response to Doc limitation update proposal: include out-of-line OID usage per TOAST-ed columns  (Jakub Wartak <jakub.wartak@enterprisedb.com>)
Responses Re: Doc limitation update proposal: include out-of-line OID usage per TOAST-ed columns
List pgsql-hackers
Hi!

This limitation applies not only to wide tables - it also applies to tables where TOASTed values
are updated very often. You would soon be out of available TOAST value ID because in case of
high frequency updates autovacuum cleanup rate won't keep up with the updates. It is discussed
in [1].


On Fri, Apr 21, 2023 at 9:39 AM Jakub Wartak <jakub.wartak@enterprisedb.com> wrote:
Hi -hackers,

I would like to ask if it wouldn't be good idea to copy the
https://wiki.postgresql.org/wiki/TOAST#Total_table_size_limit
discussion (out-of-line OID usage per TOAST-ed columns / potential
limitation) to the official "Appendix K. PostgreSQL Limits" with also
little bonus mentioning the "still searching for an unused OID in
relation" notice. Although it is pretty obvious information for some
and from commit 7fbcee1b2d5f1012c67942126881bd492e95077e and the
discussion in [1], I wonder if the information shouldn't be a little
more well known via the limitation (especially to steer people away
from designing very wide non-partitioned tables).

Regards,
-J.

[1] - https://www.postgresql.org/message-id/flat/16722-93043fb459a41073%40postgresql.org




--
Regards,
Nikita Malakhov
Postgres Professional
The Russian Postgres Company

pgsql-hackers by date:

Previous
From: vignesh C
Date:
Subject: Re: Non-superuser subscription owners
Next
From: Jehan-Guillaume de Rorthais
Date:
Subject: Re: Commitfest 2023-03 starting tomorrow!