Re: [GENERAL] The planner chooses seqscan+sort when there is an index on the sort column

2006-05-03 Thread Andreas Kretschmer
Csaba Nagy <[EMAIL PROTECTED]> schrieb: > select * from table order by col_1; Without a WHERE you get the whole table. A Index-Scan is, in this case, expensive. HTH, Andreas -- Really, I'm not out to destroy Microsoft. That will just be a completely unintentional side effect.

Re: [GENERAL] The planner chooses seqscan+sort when there is an index on the sort column

2006-05-03 Thread John D. Burger
Csaba Nagy wrote: select * from table order by col_1; Isn't it supposed to choose the index scan at least when enable_seqscan=off ? Even if it is indeed not faster to do the index scan than seqscan+sort. I think because you've asked for every row, it's going to have to scan the whole table a

[GENERAL] The planner chooses seqscan+sort when there is an index on the sort column

2006-05-03 Thread Csaba Nagy
Hi all, I wonder why this happens: - postgres: 8.1.3 - the table has ~200 million rows; - there is a primary key on (col_1, col_2); - the table was ANALYZEd; - the planner chooses seqscan+sort for the following query even with enable_seqscan=off: select * from table order by col_1; Isn't i