Re: [BUGS] Postgres not using indexes

2011-04-21 Thread Lawrence Cohan
Many thanks for all your advice and we will use Greg Smith's book on performance to incrementaly tune our environment. Regards, Lawrence Cohan. -Original Message- From: Kevin Grittner [mailto:kevin.gritt...@wicourts.gov] Sent: April-21-11 3:38 PM To: pgsql-bugs@postgresql.org; Lawrence C

Re: [BUGS] Postgres not using indexes

2011-04-21 Thread Kevin Grittner
Lawrence Cohan wrote: > We managed to put together a new test server running PG 9.0.2 on > 2socketsx6cores = 12CPU with 64 GB RAM against a 3PAR 10TB SAN. We > kept the settings I submitted already (and enclosed below) and > after 12 hours of pounding the box with PGBENCH running 8 scripts > to

Re: [BUGS] Postgres not using indexes

2011-04-21 Thread Lawrence Cohan
Hello Kevin, We managed to put together a new test server running PG 9.0.2 on 2socketsx6cores = 12CPU with 64 GB RAM against a 3PAR 10TB SAN. We kept the settings I submitted already (and enclosed below) and after 12 hours of pounding the box with PGBENCH running 8 scripts to perform all of IN

Re: [BUGS] BUG #5297: Add XATMI C API

2011-04-21 Thread Heikki Linnakangas
On 21.04.2011 20:04, Michael Musgrove wrote: I am referring to the X/Open specification Distributed Transaction Processing: The XA Specification (pubs.opengroup.org/onlinepubs/009680699/toc.pdf). If postgresql provides a conforming Resource Manager switch then we can include postgresql work along

Re: [BUGS] BUG #5297: Add XATMI C API

2011-04-21 Thread Michael Musgrove
On 04/21/2011 05:43 PM, Tom Lane wrote: Michael Musgrove writes: The response to this issue says it is "Not a bug". Would it be possible for someone to provide an indication of how much effort is involved in providing XA support in the C API library (libpq)? What do you consider to be "providi

Re: [BUGS] BUG #5297: Add XATMI C API

2011-04-21 Thread Tom Lane
Michael Musgrove writes: > The response to this issue says it is "Not a bug". Would it be possible > for someone to provide an indication of how much effort is involved in > providing XA support in the C API library (libpq)? What do you consider to be "providing XA support"? XA requires there

Re: [BUGS] database introspection error

2011-04-21 Thread Tom Lane
Jon Nelson writes: > SQLAlchemy encountered an error introspecting the tables. After > inspecting the SQL that it was running, I boiled it down to this: > SELECT c.relname, a.attname > FROM pg_index i, pg_class c, pg_attribute a > WHERE i.indrelid = '16684' AND i.indexrelid = c.oid > AND a.att

Re: [BUGS] BUG #5297: Add XATMI C API

2011-04-21 Thread Michael Musgrove
The response to this issue says it is "Not a bug". Would it be possible for someone to provide an indication of how much effort is involved in providing XA support in the C API library (libpq)?

Re: [BUGS] BUG #5974: UNION construct type cast gives poor error message

2011-04-21 Thread Tom Lane
"Kevin Grittner" writes: > Jeff Wu wrote: >> what needs to happen to get this fix in? > Well, "fix" implies that there is a bug, which there isn't. I think what Jeff was actually suggesting was that we commit the proposed added HINT. I didn't like the hint patch at all --- as given, the hint

[BUGS] database introspection error

2011-04-21 Thread Jon Nelson
I'm using SQLAlchemy which has database introspection smarts. Basically, you point it at a database and tell it to find out what tables are there and introspect them, the table indices, etc... It works super. Until today. Yesterday, I restructured one of my tables (specifically, I dropped one col

Re: [BUGS] BUG #5974: UNION construct type cast gives poor error message

2011-04-21 Thread Kevin Grittner
Jeff Wu wrote: > what needs to happen to get this fix in? Well, "fix" implies that there is a bug, which there isn't. The current behavior doesn't violate the requirements of the standard, nor is it a regression from the behavior of any previous PostgreSQL release. What we're talking about i

Re: [BUGS] BUG #5992: The database cluster initialisation failed

2011-04-21 Thread Nathan M. Davalos
I've had the same problem with server 2003. I usually create the folder where PostgreSQL is going to install into first, then change the permissions for "\users" to full control before attempting to install. -Original Message- From: pgsql-bugs-ow...@postgresql.org [mailto:pgsql-bugs-ow..

[BUGS] BUG #5992: The database cluster initialisation failed

2011-04-21 Thread Rakshit
The following bug has been logged online: Bug reference: 5992 Logged by: Rakshit Email address: r.a.ra...@gmail.com PostgreSQL version: postgresql-9.0. Operating system: Win 7 Description:The database cluster initialisation failed Details: While installing postgresq