Hm - So could this be some kind of ubuntu bug, or some problem with
minor version releases in upstream [apache|ctypes]. On my servers the
problem manifests never in Ubuntu Hardy (apache 2.2.8, python 2.5.2),
and always in Ubuntu Intrepid (apache 2.2.9, python 2.5.2). Tested
mod_wsgi versions: 2.0, 2.1, 2.3 on both
Rob, have you tried other WSGI hosts than apache/mod_wsgi?
On Nov 11, 7:38 pm, "Graham Dumpleton" <[EMAIL PROTECTED]>
wrote:
> 2008/11/11 Robert Coup <[EMAIL PROTECTED]>:
>
>
>
>
>
> > On Tue, Nov 11, 2008 at 10:56 PM, Robert Coup <[EMAIL PROTECTED]>
> > wrote:
>
> >> Maybe i should run the geos tests inside a wsgi request...
>
> > They fail inside mod-wsgi. In fact, so does a wsgi application that does:
>
> > from django.contrib.gis.geos import GEOSGeometry
> > g = GEOSGeometry("POINT(0 0)")
> > s = g.wkt # boom!
>
> > So either the pointer being returned from the C constructor is rubbish, and
> > therefore any methods being called with that explode - or the ctypes
> > interface is broken somehow. Another function in the interface -
> > geos_version() - works ok though fwiw.
>
> > /me thinks the answer is getting closer, but why only under apache/mod-wsgi?
>
> Possibly because memory layout is simply different to normal Python
> interpreter.
>
> I heard of another case once where code under mod_wsgi wasn't working
> but would work under fastcgi. Turned out to be a bug in Python cPickle
> that only manifested under mod_wsgi.
>
> 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
-~----------~----~----~----~------~----~------~--~---
