Re: Postgres eats up memory when using cursors

On Thursday 01 March 2001 21:33, Tom Lane wrote:
> Denis Perchine <dyp@perchine.com> writes:
> > I declare a cursor on the table of approx. 1 million rows.
> > And start fetching data by 1000 rows at each fetch.
> > Data processing can take quite a long time (3-4 days)
> > Theoretically postgres process should remain the same in size.
> > But it grows... In the end of 3rd day it becames 256Mb large!!!!
>
> Query details please?  You can't expect any results from such a
> vague report.

:-)))
That's right.

declare senders_c cursor for select email, first_name, last_name from senders
order by email

fetch 1000 from senders_c

db=# explain declare senders_c cursor for select email, first_name, last_name
from senders order by email;
NOTICE:  QUERY PLAN:

Index Scan using senders_email_key on senders  (cost=0.00..197005.37
rows=928696 width=36)

db=# \d senders
          Table "senders"
 Attribute  |   Type    | Modifier
------------+-----------+----------
 email      | text      |
 first_name | text      |
 last_name  | text      |
 stamp      | timestamp |
Index: senders_email_key

db=# \d senders_email_key
Index "senders_email_key"
 Attribute | Type
-----------+------
 email     | text
unique btree

That's all. I could not imagine anything more simple...

--
Sincerely Yours,
Denis Perchine

----------------------------------
E-Mail: dyp@perchine.com
HomePage: http://www.perchine.com/dyp/
FidoNet: 2:5000/120.5
----------------------------------