To my knowledge it hasn't, at least no one has asked me for the sample
database.
Robert Treat
On Thursday 26 January 2006 17:42, Bruce Momjian wrote:
> Has this been researched. Josh posted he thought it was an optimizer
> bug, but I haven't seen anyone investigate it:
>
> http://archives.pos
Has this been researched. Josh posted he thought it was an optimizer
bug, but I haven't seen anyone investigate it:
http://archives.postgresql.org/pgsql-performance/2006-01/msg00248.php
http://archives.postgresql.org/pgsql-performance/2006-01/msg00265.php
--
Magnus's case seemed like a beast of a different animal to me, given it
was a direct index scan using a wildcard based search on a primary key
column; I'd agree I don't know exactly how it would determine a value
different that 1.
But in my example, this misestimation comes between columns that a
Magnus reported a similar problem with path names. I looked at his
statistics and found that even at 100 buckets, his LIKE 'f:/.../%" query
would never span more than one bucket, and because all the path names
were unique, there were no most common values.
In the case where the LIKE hits only on
After some extensive discussion on irc, berkus, myself and a few others
think we have uncovered a possible bug, or at the least some odd
behavior in > 8.1.1. It centers around my recent post to performance
http://archives.postgresql.org/pgsql-performance/2006-01/msg00248.php
and how I could not see
