Hey everyone,

We are working on a new system to help Firefox 3 and Nightly build
testers see crash trends and analyze regressions over time.

The project is here:
http://code.google.com/p/socorro/

Issue filed on the leak:
http://code.google.com/p/socorro/issues/detail?id=46

We need some help diagnosing the problem.  Do you have any tips on how
to detect memory leaks in the Pylons environment?  Recommended
profiling tools?

Information on the environment:
- Python 2.4
- Postgresql 8.1
- Paster served through mod_proxy + Apache2

Summary of the problem (pasted from issue 46):

We have a combination of problems that are causing persistent memory
leaks,
especially on the list-by-signature page:

1) whenever we attach something to "c" in a controller, it sometimes
leaks.
I'm not quite sure why.

2) whenever we perform an ORM query, the instance objects are never
cleared
from the ORM session. For the moment, I'm going to attach a patch to
explicitly clear them, but I think we probably need a more systemic
solution such as never attaching them to a session in the first place,
or
having a single session per request that is explicitly cleared at the
end
of the request.

3) Related to 2 is the caching of instance objects: when we cache them
to a
file they cause leaks... it looks like the act of pickling them is
what
causes the leaks, not unpickling them: I can reproduce the leak even
when I
never request the same list more than once

Any thoughts or suggestions?  Thanks in advance.  :)

Mike


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pylons-discuss" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/pylons-discuss?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to