Re: autoanalyze criteria

On 02/24/2013 12:52 PM, Alban Hertroys wrote:
> On Feb 23, 2013, at 14:11, Stefan Andreatta <s(dot)andreatta(at)synedra(dot)com
> <mailto:s(dot)andreatta(at)synedra(dot)com>> wrote:
>
>> And we are still missing a number for rows updated since the last
>> analyse.
>
> In MVCC an update is an insert + delete, so you already got those numbers.
>
Good point. But because they are an update and a delete, they cancel
each other out and do not show up in pg_stat_user_tables.n_live_tup -
and that's the only value for which we have a reference value from the
time of the last analyze (pg_class.reltuples).

On the other hand, I might again miss something out. I would be most
happy if anybody could come up with a query to estimate autoanalyze
trigger conditions from the values we have available now.

Stefan