Re: Recovering real disk space - Mailing list pgsql-general

From ptjm@interlog.com (Patrick TJ McPhee)
Subject Re: Recovering real disk space
Date
Msg-id 3b99e3F6cjupoU1@uni-berlin.de
Whole thread Raw
In response to Recovering real disk space  (Adam Siegel <adam@sycamore.us>)
List pgsql-general
In article <424AEB5D.2090204@sycamore.us>,
Adam Siegel <adam@sycamore.us> wrote:

[...]

% We perform a vacuum full after each mass delete.  This cycle can happen
% many times during over a couple of weeks.  We are in a test lab
% environment and are generating a lot of data.
%
% One of the problems we have is that the vacuum on the table can take up
% to 10 hours.  We also expect to see the physical disk space go down, but
% this does not happen.

Try vacuum full verbose next time to see what it's doing.

Try reindexing after the vacuum is done.

You may find an ordinary vacuuum is faster and just as useful as vacuum full
assuming you're filling and deleting from the same table all the time. It
won't free up space, but it will allow you to maintain a high-water mark.

Look at the relpages column in pg_class to see which relations are using
up the most space.

If you're clearing out all the data for a set of tables, drop them and
recreate them.
--

Patrick TJ McPhee
North York  Canada
ptjm@interlog.com

pgsql-general by date:

Previous
From: "roberto cisneros"
Date:
Subject: conexion de postgres con php
Next
From: "Andrus Moor"
Date:
Subject: Simple UPDATE runs forever