Re: HELP: Urgent, Vacuum problem - Mailing list pgsql-general

From John Sidney-Woollett
Subject Re: HELP: Urgent, Vacuum problem
Date
Msg-id 4575C7FA.8080500@wardbrook.com
Whole thread Raw
In response to HELP: Urgent, Vacuum problem  ("Schwenker, Stephen" <SSchwenker@thestar.ca>)
Responses Re: HELP: Urgent, Vacuum problem  (Scott Marlowe <smarlowe@g2switchworks.com>)
List pgsql-general
To recover disk space, reindex the heavily updated tables. You can do
this while the database is in production.

Check the REINDEX command.

John

Schwenker, Stephen wrote:
> Hello,
>
> I'm having a major Vacuuming problem.  I used to do a full vacuum every
> morning on my postgres database to clean up empty space on a table but
> because of it's size, the locking of the database causes my application
> server to max out the database connections and causes database errors.
> To fix that problem, I have turned off the full vacuum and are just
> doing a standard analyze vacuum.  No I'm getting very close to running
> out of space on my disks because the table keeps on growing and the
> database is not re-using deleted record space.  I know this because I
> delete 99% of the records from the table after I have exported them but
> the size of the database tables are not decreasing.  Now I can't shrink
> the size of the tables because the full vacuum takes too long to run
> Over 2 hours and locks the table for too long.
>
> Can anyone help me with fixing my problem with vacuuming and disk space?
>
> I'm using version 7.4.2 on solaris.
>
> Thank you,
>
>
> Steve.
>

pgsql-general by date:

Previous
From: "Alejandro Michelin Salomon \( Adinet \)"
Date:
Subject: RES: Problem working with dates and times.
Next
From: Tom Lane
Date:
Subject: Re: Unable to restart postgres - database system was interrupted