On 22 May 2012 23:51, Brendon Stanton <[email protected]> wrote:
> I am having this same problem and wondered if anyone had discovered a
> solution.

The cause for this occurring in some cases has been hard to nail down.
Problem is that to sort it out may well need hacking up mod_wsgi to
get additional information out of it about what is going on. This is
something that has usually been beyond the people having the problem.

I note you have also posted at:

http://stackoverflow.com/questions/10695302/two-django-projects-conflicting-with-one-another

We can try and work through it here if still interested, but be warned
that am quite overloaded at the moment, so may be slow process.

To start with, ensure that have:

LogLevel info

at the minimum, rather than 'warn'.

Then monitoring messages from mod_wsgi about what process/application
context WSGI scripts are being loaded in.

A further step one could take is to ramp up to:

LogLevel debug

and add:

WSGIVerboseLogging On

You will need mod_wsgi 3.X at least from memory.

Finally, rather than sub domains I would have used separate daemon
process groups and delegated each web application to their own. That
ensures process level separation and avoids problems with environment
variable leakage.

If the issue is wrong interpreter selection within the process though,
then could see more memory used that necessary if not also forced to
use main interpreter of respect daemon processes.

Graham

-- 
You received this message because you are subscribed to the Google Groups 
"modwsgi" 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/modwsgi?hl=en.

Reply via email to