I've run across a strange problem with PG 8.3.3 not using indexes on a particular table after building the table during a transaction. You can see a transcript of the issue here: http://gist.github.com/21154 Interestingly, if I create another temp table 'CREATE TEMP TABLE AS SELECT * FROM act' as seen on line 107, then add the same indexes to that table, PG will use the indexes. While it's not in the gist transcript, even an extremely simple query like: SELECT * FROM act WHERE act_usr_id = 1; will not use the index on the original act table, but the jefftest and jefftest2 tables both work fine. As you can probably see in the transcript, the tables have been ANALYZEd. I even tried 'enable seqscan=0;' and that made the cost really high for the seq scan, but the planner still chose the seq scan. The issue does not affect 8.2.3 nor does it affect 8.3.4. I didn't see any mention of a fix for this sort of thing in 8.3.4's release notes. I was wondering if this is a known bug in 8.3.3 (and maybe other 8.3.x versions) and just didn't make it into the release notes of 8.3.4? -- Jeff Frost, Owner <jeff@xxxxxxxxxxxxxxxxxxxxxx> Frost Consulting, LLC http://www.frostconsultingllc.com/ Phone: 916-647-6411 FAX: 916-405-4032 -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
