On 11 March 2010 23:43, Ken Elkabany <[email protected]> wrote: > Hello, > > My client program is sending a gzip-ed http request, which is > decompressed by the mod-deflate InputFilter in Apache. This works when > I am using modwsgi (without any custom configurations) to serve a > django site that is receiving compressed requests. However, when I > turn on daemon mode via the WSGIDaemonProcess directive, the django > site begins receiving garbled requests which I interpret as the mod- > deflate not being active. The error is 100% reproducible, and simply > commenting out the WSGIDaemonProcess line makes everything work again > (I'm leaving in the WSGIScriptAlias directive). Any ideas? > > Side Note: If you're wondering how I compress http requests for wsgi: > My wsgi script scans through the wsgi.input (environ dictionary), > calculates the decompressed content length, and then changes CONTENT- > LENGTH appropriately before the request enters django.
Can you post the WSGI wrapper you are using to recalculate CONTENT_LENGTH and supply alternate wsgi.input to inner application? Have you use something like second recipe in: http://code.google.com/p/modwsgi/wiki/DebuggingTechniques#Tracking_Request_and_Response to record the actual request headers and request body which are going in to that wrapper and verified that the content is actually the same as the original compressed content? Can you supply a standalone Python script which acts as a HTTP client to send the compressed content as otherwise don't have a ready way of doing testing as don't have anything myself which does that? Finally, exactly what version of mod_wsgi are you using, what Apache version and what platform? 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.
