Re: patch review, please: Autovacuum/Vacuum times via stats. - Mailing list pgsql-hackers

From Jim C. Nasby
Subject Re: patch review, please: Autovacuum/Vacuum times via stats.
Date
Msg-id 20060504173823.GL97354@pervasive.com
Whole thread Raw
In response to Re: patch review, please: Autovacuum/Vacuum times via stats.  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: patch review, please: Autovacuum/Vacuum times via stats.  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Thu, May 04, 2006 at 01:20:32PM -0400, Tom Lane wrote:
> "Jim C. Nasby" <jnasby@pervasive.com> writes:
> > On Wed, May 03, 2006 at 02:25:54PM -0400, Tom Lane wrote:
> >> Isn't this already dealt with by contrib/pg_freespacemap?
> 
> > AFAIK that does nothing to tell you how much space is desired by
> > relations.
> 
> I thought the latest patch arranged to expose per-relation lastPageCount
> values.  If it doesn't then it certainly still needs work.

After CVS-upping... yes, both lastpagecount and nextpage are now
included. But unfortunately the README says next to nothing about what
they mean...
  lastpagecount  |                      | Count of pages examined for useful                 |                      |
freespace.  nextpage       |                      | page index (from 0) to start next                  |
     | search at.
 

Perhaps if there was some discussion about how the backend used these
two values it would make more sense, but right now I'm not seeing how
lastpagecount relates to how many pages weren't included in the FSM that
should have been...
-- 
Jim C. Nasby, Sr. Engineering Consultant      jnasby@pervasive.com
Pervasive Software      http://pervasive.com    work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf       cell: 512-569-9461


pgsql-hackers by date:

Previous
From: "Jim C. Nasby"
Date:
Subject: Re: autovacuum logging, part deux.
Next
From: Martijn van Oosterhout
Date:
Subject: Re: Semi-undocumented functions in libpq