Date: Tue, 6 Jul 2010 16:26:23 -0400
From: Stephen Frost <sfrost@snowman.net>
To: Ranga Gopalan <ranga_gopalan@hotmail.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Question about partitioned query behavior
Message-ID: <20100706202623.GM21875@tamriel.snowman.net>
References: <SNT129-W528E68327857A0FB9BB59A91B20@phx.gbl>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="Tz2v/xvbm3T/MeTx"
Content-Disposition: inline
In-Reply-To: <SNT129-W528E68327857A0FB9BB59A91B20@phx.gbl>
User-Agent: Mutt/1.5.18 (2008-05-17)


--Tz2v/xvbm3T/MeTx
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Ranga,

* Ranga Gopalan (ranga_gopalan@hotmail.com) wrote:
> It seems that this is an issue faced by others as well - Please see this =
link: http://stackoverflow.com/questions/2236776/efficient-querying-of-mult=
i-partition-postgres-table
>=20
> Is this a known bug? Is this something that someone is working on or is t=
here a known work around?

Actually, if you look at that, the problem the original poster had was
that they didn't have constraint_exclusion turned on.  Then they were
complaining about having the (empty) master table and the needed
partition included (which, really, shouldn't be that big a deal).

Did you look at what the other reply suggested?  Do you have
constraint_exclusion =3D 'on' in your postgresql.conf?

	Thanks,

		Stephen

--Tz2v/xvbm3T/MeTx
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkwzkW8ACgkQrzgMPqB3kii8GACeNHTVHZke6hPvaEiZSdvs5+6f
2NUAnAgJJE1koJHofsYx/X9LHN6SiFa/
=VFPz
-----END PGP SIGNATURE-----

--Tz2v/xvbm3T/MeTx--