Tom Lane wrote:
> Another possibility (though not a back-patchable solution) is that
> we could just dispense with the heuristic size estimate and trust a
> zero-sized table to stay zero-sized. This would be relying on the
> assumption that autovacuum will kick in and update the stats, leading
> to invalidation of any existing plans that assume the table is small.
> I don't feel very comfortable about that though --- throwing a few
> hundred tuples into a table might not be enough to draw autovacuum's
> attention, but it could surely be enough to create a performance
> disaster for nestloop plans.
FWIW autovacuum fires an analyze with the 51st tuple inserted on a
table on 8.3's default configuration.
--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.