Re: Help writing a query to predict auto analyze - Mailing list pgsql-general

From Alvaro Herrera
Subject Re: Help writing a query to predict auto analyze
Date
Msg-id 1274326933-sup-9169@alvh.no-ip.org
Whole thread Raw
In response to Re: Help writing a query to predict auto analyze  (Gordon Shannon <gordo169@gmail.com>)
List pgsql-general
Excerpts from Gordon Shannon's message of mié may 19 23:32:07 -0400 2010:
>
> alvherre wrote:
> >
> > n_live_tup and n_dead_tup corresponds to the current numbers,
> > whereas "last analysis tuples" are the values from back when the
> > previous analyze ran.  These counters keep moving per updates, deletes,
> > inserts, they are not static.
>
> OK.  Do you know how can I get the values from back when the previous
> analyze ran?

I don't think we expose those values to SQL.  I guess you could create a
C function to get it, modelled after pg_stat_get_live_tuples and friends
(see src/backend/utils/adt/pgstatfuncs.c).
--

pgsql-general by date:

Previous
From: Gordon Shannon
Date:
Subject: Re: Help writing a query to predict auto analyze
Next
From: Richard Walker
Date:
Subject: How to lose transaction history (xmin values, WAL, etc.)?