Pygresql / Postgresql Problem

[Rod MacNeil]

Hi,

I am fairly new to Python, Linux and Postgresql
so I'm hoping this problem is a simple mistake
on my part. I have been experimenting with some database
programming using Python 1.5.2 and Postgresql 7.
The Postgresql server is running on Redhat 7.
I have been running the Python program on another
Redhat 7 box.

The Python program uses Pygresql 3.1 to make a
lot (many thousands) of simple queries to lookup values.
I had this running on my desktop machine and noticed that
as the program runs, my machine got slower and slower.
When I looked into this I found, at least I think I found
that the amount of memory allocated to the python process
climbed steadily as it ran.

This is a simple program I wrote to demonstrate this:

-----------------------------------------------------------
import pgdb

conx =
pgdb.connect(host='192.168.1.101',user='blahblah',password='blahblah')
cur1 = conx.cursor()

for iter in range(500000):
 sql="select * from mucw where word='%s'" % 'BOWL'
 cur1.execute(sql)
 data=cur1.fetchall()

cur1.close()
conx.close()
-------------------------------------------------------------

I also coded this to delete and recreate the cursor object
each time, but I got the same result.

This is process information I gathered at 1 minute intervals
using the command ps -el | grep python:


  F S   UID   PID  PPID  C PRI  NI ADDR    SZ WCHAN  TTY          TIME CMD
000 R   600  1786  1580 27  74   0    -  2642 -      pts/4    00:00:00
python : initial
000 R   600  1786  1580 14  79   0    -  3135 -      pts/4    00:00:08
python : after 1 minute
000 R   600  1786  1580 14  77   0    -  3613 -      pts/4    00:00:17
python : after 2 minutes
000 R   600  1786  1580 14  76   0    -  4095 -      pts/4    00:00:25
python : after 3 minutes
000 R   600  1786  1580 14  73   0    -  4574 -      pts/4    00:00:34
python : after 4 minutes
000 R   600  1786  1580 14  71   0    -  5056 -      pts/4    00:00:43
python : after 5 minutes



Has anyone else had this problem?
Is this indicative of a memory leak in Pygresql?
Am I using Pygres incorrectly in some way?


Thanks in advance for any assistance you can offer, its appreciated.

Rod MacNeil
rmacneil at interactdirect.com