Hi All,

I can see my paster process (with a rather straightforward Pylons + SQLAlchemy + Genshi + ElementTree + psycopg2 Python 2.4 application) steadily growing in memory consumption with a mild use. Paster server is configured to use a thread pool. For some two weeks now and a total of under a hundred users (about 100 page accesses per user), it grew from 0.4% to 1.5% of server memory (it's a big server at one of commercial hosters, has 4 GB total). I think I should be concerned, but how do I even begin to find out what's going on and how to fix it? My first idea is to put the application under a profiler, run the unit tests, and see if anything's leaking. The limitation of this approach though is that paster server is not involved. Is there a "recommended way" to do such profiling in Python? Thanks in advance for help and advice!

Sergey.