Date: Mon, 15 Feb 2021 11:43:50 +0700
From: Victor Sudakov <vas@sibptus.ru>
To: Keith Fiske <keith.fiske@crunchydata.com>
Cc: Pgsql-admin <pgsql-admin@lists.postgresql.org>
Subject: Re: index bloat estimation
Message-ID: <YCn8BmEe/4mA0KIb@admin.sibptus.ru>
References: <YCY7oPIXobbY8oPu@admin.sibptus.ru>
 <CAODZiv6YZj3GJsTv5_dp07VS2FACrFu_JPWsFVi64=1mJD_t0A@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAODZiv6YZj3GJsTv5_dp07VS2FACrFu_JPWsFVi64=1mJD_t0A@mail.gmail.com>
Precedence: bulk

Keith Fiske wrote:
> >
> > What queries do you use to estimate index and table bloat?

[dd]

> Why estimate when you can get the exact amount? At least for b-tree indexes
> anyway.
> 
> https://github.com/keithf4/pg_bloat_check

Hello Keith,

Thanks for this script and for reminding about pgstattuple.

The script has the drawback of requiring r/w access to the database, so
it cannot be run on a replica. But the idea is excellent. At least it
can show the top N bloated relations, and then I can further explore with
pgstattuple(), pgstatindex() etc.


-- 
Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
2:5005/49@fidonet http://vas.tomsk.ru/