"Philippe Lang" writes:
> A small question here: solde_po is an SQL function (not PLPGSQL). Is it
> "inlined" in the parent query before the whole query execution plan is
> calculated?
You should be able to tell that by inspecting the filter conditions
in the ANALYZE output. Do you see solde_po(
In response to "Philippe Lang" :
[snip]
> > Anyway, the real reason I posted -- I doubt if anyone will be able to
> > make sense of a query plan that complex without the actual query, so
> > you'll probably want to post it as well.
>
> :) What? I thought you would read that like Neo was readin
pgsql-general-ow...@postgresql.org wrote:
> In response to "Philippe Lang" :
>>
>> I'm using Postgresql 8.3.6 under Freebsd 7.1.
>>
>> After a fresh restore of a customer dump (running version 8.2.7 at
>> the moment), a rather big query executes in about 30 seconds. As
>> soon as I run ANALYZE, i
hi
it's look better with explain.depesz.com :) you can easily find that
in "before" it uses index scan and in "after" it uses seq
before:
http://explain.depesz.com/s/RC
after:
http://explain.depesz.com/s/nm
try as mentioned before to change costs
Best regards,
Sebastian Pawłowski
Wia
Oh, the other thing you could try experimenting with are these two parameters.
Your query has *way* more tables than the default values for these so you
would have to raise them substantially. Given that the query was running in
30s you may find that this increases the planning time by more time
> In response to "Philippe Lang" :
>>
>> I'm using Postgresql 8.3.6 under Freebsd 7.1.
>>
>> After a fresh restore of a customer dump (running version 8.2.7 at the
>> moment), a rather big query executes in about 30 seconds. As soon as I
>> run ANALYZE, it is instantly 4-5 times slower. I could che
Bill Moran writes:
> I opened one of those links figuring I'd take a few minutes to see if I could
> muster up some advice ... and just started laughing ... definitely not the
> type of query that one can even understand in just a few minutes!
You might consider setting default_statistics_target
In response to "Philippe Lang" :
>
> I'm using Postgresql 8.3.6 under Freebsd 7.1.
>
> After a fresh restore of a customer dump (running version 8.2.7 at the
> moment), a rather big query executes in about 30 seconds. As soon as I
> run ANALYZE, it is instantly 4-5 times slower. I could check tha
Hi,
I'm using Postgresql 8.3.6 under Freebsd 7.1.
After a fresh restore of a customer dump (running version 8.2.7 at the
moment), a rather big query executes in about 30 seconds. As soon as I
run ANALYZE, it is instantly 4-5 times slower. I could check that
multiples times.
Here is the EXPLAIN A
