Re: Vacuum Full - stops responding(?) - Mailing list pgsql-admin

From Scott Marlowe
Subject Re: Vacuum Full - stops responding(?)
Date
Msg-id dcc563d10910191937ld71f44bif644e46e4acb3d94@mail.gmail.com
Whole thread Raw
In response to Re: Vacuum Full - stops responding(?)  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Vacuum Full - stops responding(?)  (Greg Stark <gsstark@mit.edu>)
List pgsql-admin
On Mon, Oct 19, 2009 at 4:38 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Krowa Krowax <krowa333@gmail.com> writes:
>> INFO:  vacuuming "pg_toast.pg_toast_17929229"
>> INFO:  "pg_toast_17929229": found 6203 removable, 257509 nonremovable
>> row versions in 20509016 pages
>> DETAIL:  0 dead row versions cannot be removed yet.
>> Nonremovable row versions range from 45 to 2030 bytes long.
>> There were 87521183 unused item pointers.
>> Total free space (including removable row versions) is 166910212580 bytes.
>> 20384082 pages are or will become empty, including 3 at the end of the table.
>> 20507826 pages containing 166910155528 free bytes are potential move
>> destinations.
>
> That might be about the worst case of table bloat I've ever seen :-(
>
>> After 10 hours nothing new appear... What is going on with this vacuum?
>
> It's trying to repack your table in place.  On the whole, I'd suggest
> you cancel the vacuum and try a CLUSTER or some such instead.

Unless that table's already in (mostly) index order, even cluster is
gonna be pretty painful.
I'd suggest a dump and reload myself.

> After you get this cleaned up, think about enabling autovacuum ...

This.  Very much this.

pgsql-admin by date:

Previous
From: Tom Lane
Date:
Subject: Re: Vacuum Full - stops responding(?)
Next
From: Greg Stark
Date:
Subject: Re: Vacuum Full - stops responding(?)