On Fri, Feb 6, 2009 at 3:43 PM, Mario Splivalo <mario.splivalo@xxxxxxxxxx> wrote: > Besides PK and uq-constraint indices I have this index: > > CREATE INDEX transactions_idx__client_data ON transactions > USING btree (transaction_client_id, transaction_destination_id, > transaction_operator_id, transaction_application_id, > transaction_time_commit) I think it is because it isn't just a simple index, and for some reason planner decides - that going through every transaction_application_id, etc, etc just to find right transaction_time_commit isn't worth the hassle. Try using few more indexes, on less columns. Also, if I may - I don't think it is quite usefull to have column names that include table name in every single one of them, makes things so much less readable. -- GJ -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
