memory error on lighttp deployment (file upload)

[arthur debert]

I have an app deployed on lighttpd, but I am getting this memory error.
My local codebase runs fine (both on the devserver and on my local
lighttpd),which leads me to think it's a server issue. I am hosted on
textdrive - we're running lighttp + fcgi.

This only happens when uploading large files (+1 MB) on the admin
interface.

This is my python stack trace:
Switch back to interactive view

Traceback (most recent call last):
File "/home/stimuli/local/django_src/django/core/handlers/base.py" in
get_response
74. response = callback(request, *callback_args, **callback_kwargs)
File
"/home/stimuli/local/django_src/django/contrib/admin/views/decorators.py"
in _checklogin
54. return view_func(request, *args, **kwargs)
File "/home/stimuli/local/django_src/django/views/decorators/cache.py"
in _wrapped_view_func
40. response = view_func(request, *args, **kwargs)
File
"/home/stimuli/local/django_src/django/contrib/admin/views/main.py" in
add_stage
253. errors = manipulator.get_validation_errors(new_data)
File "/home/stimuli/local/django_src/django/forms/__init__.py" in
get_validation_errors
58. errors.update(field.get_validation_errors(new_data))
File "/home/stimuli/local/django_src/django/forms/__init__.py" in
get_validation_errors
351. self.run_validator(new_data, validator)
File "/home/stimuli/local/django_src/django/forms/__init__.py" in
run_validator
337. if self.is_required or new_data.get(self.field_name, False) or
hasattr(validator, 'always_test'):
File "/home/stimuli/local/django_src/django/utils/datastructures.py" in
get
137. val = self[key]
File "/home/stimuli/local/django_src/django/utils/datastructures.py" in
__getitem__
114. raise MultiValueDictKeyError, "Key %r not found in %r" % (key,
self)
File "/home/stimuli/local/django_src/django/utils/datastructures.py" in
__repr__
104. return "<MultiValueDict: %s>" % dict.__repr__(self)

MemoryError at /admin/site_updater/demoreel/add/

and the lighttp error log entry:
2006-05-12 16:04:17: (mod_fastcgi.c.2851) backend died, we disable it
for a 5 seconds and send the request to another backend instead:
reconnects: 0 load: 1

any input on this would be very welcome.

thanks,
arthur

[James Bennett]

On 5/12/06, arthur debert <deb...@gmail.com> wrote:
> This only happens when uploading large files (+1 MB) on the admin
> interface.

Related:

Jason from TextDrive replied to the TXD forum thread on this issue[1]
and indicated that it's probably due to the file being stored in
memory during upload and thus running into the per-process memory
limit for their shared hosting.

[1] http://forum.textdrive.com/viewtopic.php?id=10332

--
"May the forces of evil become confused on the way to your house."
-- George Carlin