Re: [Fwd: Re: [PERFORM] Outer joins and Seq scans]

Sami Dalouche wrote: > -- For some reason, my message doesn't seem to go through the mailing > list, so I am trying without any attachment FWIW, you can post EXPLAIN ANALYZE results on the web here: http://www.explain-analyze.info/ It's a pretty cool utility by Michael Glaesemann that should make

Re: [Fwd: Re: [PERFORM] Outer joins and Seq scans]

Thanks for your answer. So, basically, what you are saying is that there is nothing particularly wrong with the query, nor with its optimization ? So if I need performance for this query, I should just revert to other techniques (giving more memory to postgres, caching outside postgres, etc..) ?

Re: [Fwd: Re: [PERFORM] Outer joins and Seq scans]

Sami Dalouche <[EMAIL PROTECTED]> writes: > Compare that to the following query, that is exactly the same except > that the City table is inner'joined instead of outer joined > ... > the explain analyze is available at : > http://www.photosdesami.com/temp/exp6.txt AFAICS it's just absolutely blind

Re: [Fwd: Re: [PERFORM] Outer joins and Seq scans]

Sami Dalouche <[EMAIL PROTECTED]> writes: > -- For some reason, my message doesn't seem to go through the mailing > list, so I am trying without any attachment Please don't do that, at least not that way. These explain outputs have been line-wrapped to the point of utter unreadability. The main

[Fwd: Re: [PERFORM] Outer joins and Seq scans]

-- For some reason, my message doesn't seem to go through the mailing list, so I am trying without any attachment Hi, Thank you Tom and Dimitri for your precious help. So, I applied the patch at http://archives.postgresql.org/pgsql-committers/2007-10/msg00374.php The good news is that with the

Re: [PERFORM] Outer joins and Seq scans

Richard Huxton a écrit : Dimitri Fontaine wrote: Hi, Le lundi 29 octobre 2007, Tom Lane a écrit : Is there any chance you can apply the one-line patch shown here: http://archives.postgresql.org/pgsql-committers/2007-10/msg00374.php If rebuilding packages is not to your taste, possibly a down-

Re: [PERFORM] Outer joins and Seq scans

Dimitri Fontaine wrote: Hi, Le lundi 29 octobre 2007, Tom Lane a écrit : Is there any chance you can apply the one-line patch shown here: http://archives.postgresql.org/pgsql-committers/2007-10/msg00374.php If rebuilding packages is not to your taste, possibly a down-rev to 8.2.4 would be the

Re: [PERFORM] Outer joins and Seq scans

Hi, Le lundi 29 octobre 2007, Tom Lane a écrit : > Is there any chance you can apply the one-line > patch shown here: > http://archives.postgresql.org/pgsql-committers/2007-10/msg00374.php > > If rebuilding packages is not to your taste, possibly a down-rev to > 8.2.4 would be the easiest solution

Re: [PERFORM] Outer joins and Seq scans

Sami Dalouche <[EMAIL PROTECTED]> writes: > So, the version of postgres I use is : > [EMAIL PROTECTED]:~/Desktop $ dpkg -l | grep postgres > ii postgresql-8.2 8.2.5-1.1 OK. I think you have run afoul of a bug that was introduced in 8.2.5 that causes it not to realize

Re: [PERFORM] Outer joins and Seq scans

Hi, Sorry for not giving enough information.. I didn't want to pollute you with too much detail... So, the version of postgres I use is : [EMAIL PROTECTED]:~/Desktop $ dpkg -l | grep postgres ii postgresql-8.2 8.2.5-1.1 object-relational SQL database, version 8.2 ii

Re: [PERFORM] Outer joins and Seq scans

Sami Dalouche <[EMAIL PROTECTED]> writes: > So, what could prevent postgrs from using the index ? You've carefully withheld all the details that might let us guess. If I had to guess anyway, I'd guess this is a pre-8.2 PG release that doesn't know how to rearrange outer joins, but there are any nu

[PERFORM] Outer joins and Seq scans

Hi, I have a query that uses left outer join, and this seems to prevent the index on the right column to be used. I couldn't really trim down the query without having the index used normally.. So, I have the following tables that join : Offer -> AdCreatedEvent -> Account -> ContactInformation