Thank you, Greg! I tweaked your function to use recursion to search all
inherited tables; my inheritance structure is two levels deep.
This function is for integers only; I will copy/waste to create one for
timestamps. Extra credit for anyone who can rewrite it to be polymorphic.
-- Same
On Wed, Sep 2, 2009 at 4:15 PM, Kenneth Cox wrote:
> With postgresql-8.3.6, I have many partitions inheriting a table. SELECT
> min() on the parent performs a Seq Scan, but SELECT min() on a child uses
> the index. Is this another case where the planner is not aware enough to
> come up with the b
Yep I ran into the exact same problem.
My solution was to create a pl/pgsql function to query the child tables: (
http://archives.postgresql.org/pgsql-performance/2008-11/msg00284.php)
If you find a better solution please share.
-Greg Jaman
On Wed, Sep 2, 2009 at 1:15 PM, Kenneth Cox wrote:
With postgresql-8.3.6, I have many partitions inheriting a table. SELECT
min() on the parent performs a Seq Scan, but SELECT min() on a child uses
the index. Is this another case where the planner is not aware enough to
come up with the best plan? I tried creating an index on the parent ta
