Re: How to force PostgreSQL using an index

"Owen Jacobson" <ojacobson(at)osl(dot)com> writes:
> On Wed, Feb 15, 2006 at 04:58:54PM -0500, Daniel Caune wrote:
>> I see, but that doesn't explain whether it is possible to specify the
>> index to use. It seems that those options just force PostgreSQL using
>> another plan.

> Drop the simple index and re-create it when you're done?

BTW, the cute way to do that is

BEGIN;
DROP INDEX unwanted;
EXPLAIN ANALYZE whatever...;
ROLLBACK;

No need to actually rebuild the index when you are done.

This does hold an exclusive lock on the table for the duration of your
experiment, so maybe not such a good idea in a live environment ... but
then again, dropping useful indexes in a live environment isn't a good
idea either, and this at least reduces the duration of the experiment by
a good deal.

regards, tom lane