From: | Merlin Moncure <mmoncure(at)gmail(dot)com> |
---|---|
To: | Tyler Hildebrandt <tyler(at)campbell-lange(dot)net> |
Cc: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: Query timing increased from 3s to 55s when used as a function instead of select |
Date: | 2010-05-25 14:57:42 |
Message-ID: | AANLkTilo7ySxA-Z47lF6jMdgzM7yv3s25-sYTiPWNwyl@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On Tue, May 25, 2010 at 10:55 AM, Merlin Moncure <mmoncure(at)gmail(dot)com> wrote:
> On Tue, May 25, 2010 at 9:41 AM, Tyler Hildebrandt
> <tyler(at)campbell-lange(dot)net> wrote:
>>> I think, your problem is here:
>>>
>>> SELECT INTO current_user * FROM
>>> fn_medirota_validate_rota_master(in_currentuser);
>>>
>>>
>>> The planner has no knowledge about how many rows this functions returns
>>> if he don't know the actual parameter. Because of this, this query
>>> enforce a seq-scan. Try to rewrite that to something like:
>>>
>>> execute 'select * from fn_medirota_validate_rota_master(' ||
>>> in_currentuser' || ')' into current_user
>>>
>>
>> Thanks for your response. This doesn't seem to solve our issue, unfortunately.
>>
>> As a side to that, we have the fn_medirota_validate_rota_master calls in a
>> large amount of our other functions that are running very well.
>
> any chance of seeing the function source?
oops! I missed it :-). looking at your function, what version of
postgres? have you experimented w/return query?
merlin
From | Date | Subject | |
---|---|---|---|
Next Message | Jorge Montero | 2010-05-25 16:18:11 | Re: Query timing increased from 3s to 55s when used as a function instead of select |
Previous Message | Merlin Moncure | 2010-05-25 14:55:39 | Re: Query timing increased from 3s to 55s when used as a function instead of select |