Re: Planner performance extremely affected by an hanging transaction (20-30 times)?

[Jesper Krogh <jesper@xxxxxxxx>]

On 21/09/2013, at 00.01, Jeff Janes <jeff.janes@xxxxxxxxx> wrote:

See "In progress INSERT wrecks plans on table" and "Performance bug in prepared statement binding in 9.2" also on this list

This feels like the same

http://postgresql.1045698.n5.nabble.com/Slow-query-plan-generation-fast-query-PG-9-2-td5769363.html

The issues are:

1) The planner actually queries the relation to find the end points of the variable ranges, rather than using potentially out-of-date statistics.

In my app i would prefer potentially out-of-date statistics instead.

Jesper