On Tue, Aug 02, 2005 at 10:41:01AM -0500, Dan Armbrust wrote:
> But it is rather easy to get into rant mode when the prevailing
> opinion is that not being able to insert rows into a table with a
> foreign key without running Analyze after X rows is a misuse of the
> DB, rather than a bug.
I havn't seen it in this thread but it reminds me of someone who had a
similar problem a long time ago. Basically, he was doing:
TRUNCATE
ANALYZE
<load database>
This screwed everything up, because the ANALYZE set the statistics to
zero size tables. The solution was: *Don't* analyze the table when it's
empty. If he left out the ANALYZE altogether it worked.
It also works because just after a CREATE TABLE it defaults to using
indexes too.
The *only* time it starts worrying about seq scans is if you run ANALYZE
on an empty table. So don't do that.
Hope this helps,
--
Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/
> Patent. n. Genius is 5% inspiration and 95% perspiration. A patent is a
> tool for doing 5% of the work and then sitting around waiting for someone
> else to do the other 95% so you can sue them.