Re: seq scan on a table whereas an index exist

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Nicolas GIMMILLARO <Nicolas(dot)GIMMILLARO(at)wmi(dot)fr>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: seq scan on a table whereas an index exist
Date: 2001-03-08 02:07:04
Message-ID: 24542.984017224@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Nicolas GIMMILLARO <Nicolas(dot)GIMMILLARO(at)wmi(dot)fr> writes:
> What's wrong with my "select * from web_access where date between
> '01/01/2001' and '06/03/2001' " query ?

Nothing. The planner is estimating that this query will retrieve a
sufficiently large fraction of the rows in the table that a seqscan
will be cheaper than an indexscan. That estimate might or might not
be correct, but it's operating as designed.

The form with the subselects confuses the planner enough that it falls
back on default selectivity estimates, which do lead it to choose an
indexscan.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2001-03-08 14:59:34 Depending on system install scripts (was Re: COBOL)
Previous Message Tom Lane 2001-03-08 00:42:40 Re: Pbm with aggregates on empty output