Re: postgres FROM clause problem

Поиск
Список
Период
Сортировка
От Paolo Tavalazzi
Тема Re: postgres FROM clause problem
Дата
Msg-id 200403101833.41692.ptavalazzi@charta.it
обсуждение исходный текст
Ответ на Re: postgres FROM clause problem  (Tom Lane <tgl@sss.pgh.pa.us>)
Ответы Re: postgres FROM clause problem  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: postgres FROM clause problem (GROUP BY subquestion)  (Bruno Wolff III <bruno@wolff.to>)
Re: postgres FROM clause problem  (Martijn van Oosterhout <kleptog@svana.org>)
Список pgsql-general
Alle 16:54, mercoledì 10 marzo 2004, hai scritto:
> Paolo Tavalazzi <ptavalazzi@charta.it> writes:
> > I have two query that they are different only for order of the tables
> > in FROM lclause , but give back different query plan :
>
> Hm, seems like the planner is making wacko estimates in the second case.
> You didn't say what data types are involved in this query --- are any of
> the join columns int8 or float8 or timestamp?  If so you might be
> getting bitten by the 7.4 pg_statistic alignment bug.  Please follow the
> repair procedures suggested in the 7.4.2 release notes:
> http://developer.postgresql.org/docs/postgres/release.html#RELEASE-7-4-2
> and see if that improves matters.
>
>             regards, tom lane


I  have applied the procedure for fixing pg_statistic as you had said, but  the result is the same!
Only tran.time in the query is a timestamp , no int8 or float8.
The OS is FEDORA 0.96 x86_64 and the flag --enable-integer-datetimes is false,it could be a problem??

I don't know what to make, help me please!

Another question, in 7.4 GROUP BY clause  not guarantee the ordering of result.
Can I desable the new group by algorithm to maintain the coherence whit the programs that I have in production with
pg_7.3

Thank you !!!

В списке pgsql-general по дате отправления:

Предыдущее
От: Bruno Wolff III
Дата:
Сообщение: Re: Moving from 7.3.4 to 7.4.x?
Следующее
От: Frank van Vugt
Дата:
Сообщение: Re: does this look more like a possible bug or more like a possible hardware problem...? (long)