Re: Really dumb planner decision - Mailing list pgsql-performance

From Robert Haas
Subject Re: Really dumb planner decision
Date
Msg-id 603c8f070904160442l361465b7i6c3a168956eb5c24@mail.gmail.com
Whole thread Raw
In response to Re: Really dumb planner decision  (Matthew Wakeling <matthew@flymine.org>)
Responses Re: Really dumb planner decision
List pgsql-performance
2009/4/16 Matthew Wakeling <matthew@flymine.org>:
> On Thu, 16 Apr 2009, Grzegorz Jaśkiewicz wrote:
>>
>> On Thu, Apr 16, 2009 at 11:37 AM, Matthew Wakeling <matthew@flymine.org>
>> wrote:
>>>
>>> I have a query that is executed really badly by Postgres. It is a nine
>>> table
>>> join, where two of the tables are represented in a view. If I remove one
>>> of
>>> the tables from the query, then the query runs very quickly using a
>>> completely different plan.
>>
>> And what happens if you execute that view alone, with WHERE .. just
>> like it would be a part of the whole query? ((id = 1267676))
>
> Really quick, just like the query that works in my email.

What happens if you change join_collapse_limit and from_collapse_limit
to some huge number?

http://www.postgresql.org/docs/current/static/runtime-config-query.html#GUC-FROM-COLLAPSE-LIMIT

...Robert

pgsql-performance by date:

Previous
From: Matthew Wakeling
Date:
Subject: Re: Really dumb planner decision
Next
From: Matthew Wakeling
Date:
Subject: Re: Really dumb planner decision