Re: ANALYZE after restore - Mailing list pgsql-hackers

From Neil Conway
Subject Re: ANALYZE after restore
Date
Msg-id 20020402205147.3960fdae.nconway@klamath.dyndns.org
Whole thread Raw
In response to ANALYZE after restore  ("Christopher Kings-Lynne" <chriskl@familyhealth.com.au>)
List pgsql-hackers
On Wed, 3 Apr 2002 09:40:13 +0800
"Christopher Kings-Lynne" <chriskl@familyhealth.com.au> wrote:
> Hi,
> 
> Would it be an idea to have pg_dump append an ANALYZE; command to the end of
> its dumps to assist newbies / inexperienced admins?

That strikes me as a good idea; a lot of the questions we get on
-general and on IRC are solved by suggesting "have you run ANALYZE?"
And that is only the sub-section of the user community that takes the
time to track down the problem and posts about it to the mailing
list -- I shudder to think how many people have never taken the time
to tune their database at all.

Given that ANALYZE is now a separate command, so there is no need to
run a VACUUM (which could be much more expensive); furthermore, since
ANALYZE now only takes a statistical sampling of the full table, it
shouldn't take very long, even on large tables. However, I'd say we
should make this behavior optional, controlled by a command-line
switch, but it should be enabled by default.

Cheers,

Neil

-- 
Neil Conway <neilconway@rogers.com>
PGP Key ID: DB3C29FC


pgsql-hackers by date:

Previous
From: "Christopher Kings-Lynne"
Date:
Subject: SHOW ALL as a query result
Next
From: Peter Eisentraut
Date:
Subject: Re: SHOW ALL as a query result