On Fri, Jul 3, 2020 at 5:50 PM Tom Lane <t...@sss.pgh.pa.us> wrote: > > OK, I'll go ahead and push the patch I proposed yesterday. >
Thank you. I tested 12_STABLE with my real queries on the real data set, and the "hard coded" estimate of 200 distinct rows (when use_remote_estimte is turned back on) is enough to get rid of the worst plans I was seeing in 12.3. Cheers, Jeff