[GENERAL] poor cpu utilization on dual cpu box

Folks, I’m hoping someone can give me some pointers to resolving an issue with postgres and it’s ability to utilize multiple CPUs effectively. The issue is that no matter how much query load we throw at our server it seems almost impossible to get it to utilize more than 50% cpu on a dual-cpu

Re: [GENERAL] unexpected query behavior with UTF text

Hi, Thanks for the response. I actually have tried to re initdb with locale=C and I got the same results. If my locale is ignoring Ja chars, how can I change that? I expected that UTF-8 would consider all characters, being a universal encoding. A colleague also tried the same test with a

[GENERAL] SCSI vs. IDE performance test

http://hardware.devchannel.org/hardwarechannel/03/10/20/1953249.shtml?tid=20tid=38tid=49 -- - Ron Johnson, Jr. [EMAIL PROTECTED] Jefferson, LA USA I can't make you have an abortion, but you can *make* me pay child support for 18

Re: [GENERAL] poor cpu utilization on dual cpu box

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 It is probably not really a pg issue that you don't get your CPUs fully used. I would guess that one CPU saturates the memory bus or the disk bandwidth depending on where your data is. This is the typical behavior for memory bounded applications where

Re: [GENERAL] plpgsql extract epoch problem

Sorry found the answer I was taking the docs to literal It should be extract(epoch from _test) Sorry for the email Darren Darren Ferguson wrote: Hey all I am trying to extract the epoch from a timestamp variable in plpgsql. The function is as follows: create or replace function test() returns

[GENERAL] Prepared Statements

Does anyone know the exact definition of a Prepared Statement? String sql = SELECT count(id) FROM customer; executeQuery(sql); I assume the sql string is a prepared statement? Semantic debate here. - Thx ---(end of broadcast)--- TIP

Re: [GENERAL] plpgsql extract epoch problem

On Wed, Oct 22, 2003 at 12:50:22PM -0400, Darren Ferguson wrote: create or replace function test() returns integer as ' declare _test timestamp with time zone; _f integer; begin _test := current_timestamp; _f := extract(epoch from timestamp _test); Why not simply _f := extract(epoch

Re: [GENERAL]

Andrew Overholt wrote: djee Bcc: Subject: Re: [PORTS] [GENERAL] Redhat RPMs Reply-To: Andrew Overholt [EMAIL PROTECTED] In-Reply-To: [EMAIL PROTECTED] Lamar Owen once said: On Friday 10 October 2003 08:52 pm, Christopher Browne wrote: Oops! [EMAIL PROTECTED] (Nigel J. Andrews) was seen

Re: [GENERAL] [Erserver-general] alternate eRServer build

On Wed, Oct 22, 2003 at 02:43:54PM -0400, Andrew Rawnsley wrote: This is to announce an alternate fork/build of the eRServer 1.2 code base, provided to the community for testing purposes. If everything works out fine, and the Powers That Be approve, we can work on moving some of it to the

Re: [GENERAL] Identifier max length

Bruce Momjian wrote: Birahim FALL wrote: Hi, Certainly a stupid question : What is the max length for identifiers. A test shows me 63 for a table name. I just imagine it's the same for constraint name, view name etc ... but I'm not sure. Or where can I get this info, because it's not in the

Re: [GENERAL] [Erserver-general] alternate eRServer build

On Wednesday, October 22, 2003, at 03:03 PM, Andrew Sullivan wrote: On Wed, Oct 22, 2003 at 02:43:54PM -0400, Andrew Rawnsley wrote: This is to announce an alternate fork/build of the eRServer 1.2 code base, provided to the community for testing purposes. If everything works out fine, and the

Re: [GENERAL] Trigger

On Wed, 22 Oct 2003, Peter Childs wrote: Is it possible to deferr a trigger until commit, Or to have the trigger not occur if the transaction is rolled back? Like transaction. I think its possible since constraints use triggers and if so why is this a standard feature. Also

[GENERAL] plpgsql extract epoch problem

Hey all I am trying to extract the epoch from a timestamp variable in plpgsql. The function is as follows: create or replace function test() returns integer as ' declare _test timestamp with time zone; _f integer; begin _test := current_timestamp; _f := extract(epoch from timestamp _test);

[GENERAL] ANNOUNCE: Bricolage 1.6.7

I'm pleased to announce the release of Bricolage 1.6.7. This maintenance release addresses a few issues discovered since the release of version 1.6.6. Some of the more important changes include: * Fixed bric_soap to accept a --server argument starting with https, which is more

Re: [GENERAL] Trigger

scott.marlowe writes: A trigger inside a transaction should automagically roll back should the transaction fail, shouldn't it? The database actions that the trigger possibly executed are rolled back just like any other database actions. But if your trigger does something external to the

Re: [GENERAL] database files

Gail Zacharias [EMAIL PROTECTED] writes: I am investigating the possibility of using pgsql as the database in an application. I have some unusual requirements that I'd like to ask you all about. I apologize in advance if my terminology is a little off, I'm not familiar with pgsql (yet). I

Re: [GENERAL] database files

[EMAIL PROTECTED] writes: Gail Zacharias [EMAIL PROTECTED] writes: Is the file format of the pgsql database files compatible between OS's? E.g. could I take some database files from Linux and use them on Windows? I don't know the answer to that, but I would be interested in finding out. I

[GENERAL] ANNOUNCE: Bricolage 1.6.7

I'm pleased to announce the release of Bricolage 1.6.7. This maintenance release addresses a few issues discovered since the release of version 1.6.6. Some of the more important changes include: * Fixed bric_soap to accept a --server argument starting with https, which is more

[GENERAL] ANNOUNCE: Bricolage-Devel 1.7.0

It give me great pleasure to announce the release of Bricolage-Devel 1.7.0, the first development release for what will eventually become Bricolage 1.8.0. In addition to all of the bug fixes included in the 1.6.x series, this version of the open-source content management system adds a number of

[GENERAL] Indexes and querys

Hello list, I have a table with an index in a foreing key field (int4), if I do select from pedido where prvdfk=3, explain says is using seq scan to access it. I noticed all my querys runs in that way, except in the querys where the PK field (unique type index) is included in the where

[GENERAL] Adding fields on the fly

I am investigating the possibility of using pgsql as the database in an application. I have some unusual requirements that I'd like to ask you all about. I apologize in advance if my terminology is a little off, I'm not familiar with pgsql (yet). My first requirement is that I need to be able

Re: [GENERAL] Adding fields on the fly

Gail Zacharias writes: Is this even possible? Is ALTER TABLE supported? Yes. (RTFM) If so, what performance characteristics can I expect? Compared to, say, inserting a record, can I expect this to be 10 times slower? 1000 times slower? What does the performance depend on? Number of rows?

Re: [GENERAL] plpgsql: return multiple result sets or temp table

--- Oksana Yasynska [EMAIL PROTECTED] wrote: Jeff, thank you for the time and suggestion. I'm also trying to use SETOF custom_type as a solution I don't believe that works in 7.2. But if you do make it work, be sure to tell me how ;-) That will work with 7.3, but I am assuming that, as

Re: [GENERAL] Identifier max length

Birahim FALL wrote: Hi, Certainly a stupid question : What is the max length for identifiers. A test shows me 63 for a table name. I just imagine it's the same for constraint name, view name etc ... but I'm not sure. Or where can I get this info, because it's not in the Postgresl docs !

[GENERAL] database files

I am investigating the possibility of using pgsql as the database in an application. I have some unusual requirements that I'd like to ask you all about. I apologize in advance if my terminology is a little off, I'm not familiar with pgsql (yet). I need to be able to move the database files,

Re: [GENERAL] database files

-Original Message- From: Gail Zacharias [mailto:[EMAIL PROTECTED] Sent: Wednesday, October 22, 2003 12:42 PM To: [EMAIL PROTECTED] Subject: [GENERAL] database files I am investigating the possibility of using pgsql as the database in an application. I have some unusual

Re: [GENERAL] Indexes and querys

Josué Maldonado writes: I have a table with an index in a foreing key field (int4), if I do select from pedido where prvdfk=3, explain says is using seq scan to access it. I noticed all my querys runs in that way, except in the querys where the PK field (unique type index) is included in the

Re: [GENERAL] Same conditions, different planning?

Pedro Alves [EMAIL PROTECTED] writes: As a final attempt I tryed a pg_dumpall; initdb; psql -f file . It seems to work as expected (!) It may have been a case of index bloat--did you ever REINDEX? That might have fixed it without a dump/reload. 7.4 has fixes for the index bloat

[GENERAL] Atrybutes ???

Hello !!! I have store some attrybutes to database. Attrybutes are set of integer values. {1,5,7} is valid set, but {1,3,1} is not valid set I have to build compare relation i.e. {1,5,7} is equivalent to {5,7,1} or {1,7,5} etc. but {1,5,7} is not equivalent to {5,1} or {1,4,2} What's best

[GENERAL] Identifier max length

Title: Message Hi, Certainly a stupid question : What is the max length for identifiers. A test shows me 63 for a table name. I just imagine it's the same for constraint name, view name etc ... but I'm not sure. Or where can I get this info, because it's not in the Postgresl docs ! Thanks

[GENERAL] PostgreSQL v7.4 Beta5 Available for Testing

On Oct 3rd, we released beta4 in the hopes that it would be the final beta before going into the Release phase of development ... due to several bug fixes over the past 3 weeks, we've decided to release a Beta5 in order to flesh out any bugs that may have been introduced ... Beta5 contains

Re: [GENERAL] Postgreqsl Package

On Wednesday 22 October 2003 17:11, Birahim FALL wrote: Thanks, Shridhar, I've read something like that in my search this morning! But doing that you can have only one package in the scope of a schema. Then in a multipackage project these packages cannot access directly to the same objects.

Re: [GENERAL] Postgreqsl Package

On Wednesday 22 October 2003 16:41, Birahim FALL wrote: Hi, I was an oracle dev/admin and I'm quite new to postgresql. Postgresql is great, but something bothers me. It seems that there's no concept of PACKAGE in PL/pgSQL as in Oracle PL/SQL. Is is definitely that? or did I missed something?