pg.ddx.io  pgsql-admin@postgresql.org mailing list archive  

help / color / mirror / Atom feed
From: Kenneth Marshall <ktm@rice.edu>
To: Shridhar Polas <shridharpolas@gmail.com>
Cc: pgsql-admin@postgresql.org
Subject: Re: Space occupied by Postgres index.
Date: Tue, 30 Nov 2010 11:28:41 -0600
Message-ID: <20101130172841.GK19162@aart.is.rice.edu> (raw)
In-Reply-To: <00b901cb90b3$ca9ddc40$5fd994c0$@gmail.com>
References: <00b901cb90b3$ca9ddc40$5fd994c0$@gmail.com>

On Tue, Nov 30, 2010 at 10:56:53PM +0530, Shridhar Polas wrote:
> Hi,
> 
>  
> 
> I am facing a problem where indexes creates on some tables are occupying
> huge space on disk and it seems to me that this space is not getting
> reclaimed even when there are very few record in an associated table.
> 
>  
> 
> When I ran full vacuum the disk space was reclaimed occupied by tables but
> not by disk space occupied by indexes.
> 
>  
> 
> Can somebody please tell me when disk space occupied by Postgres index is
> reclaimed, without performing re-indexing on those tables?
> 
>  
> 
> Thanks,
> 
> Shridhar
> 

VACUUM FULL will cause index bloat. You will need to REINDEX to
recover the space. Note, you should not really need to use VACUUM
FULL in a normal correctly configured system.

Cheers,
Ken



view thread (5+ messages)  latest in thread

Message-ID: <20101130172841.GK19162@aart.is.rice.edu>
Permalink:  ../20101130172841.GK19162@aart.is.rice.edu/
Also on:    postgresql.org/message-id/20101130172841.GK19162@aart.is.rice.edu

 · 

reply
Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-admin@postgresql.org
  Cc: ktm@rice.edu, shridharpolas@gmail.com
  Subject: Re: Space occupied by Postgres index.
  In-Reply-To: <20101130172841.GK19162@aart.is.rice.edu>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox