Joseph Tate wrote: > I'm using Turbogears SVN and trying to get a deployment working behind > lighttpd. Everything seems to be fine, except when I send a largish > post request (>32K). These seem to timeout and give me a protocol > error on whatever client is connecting. > > These same post requests work fine if using the standalone cherrypy > server. The requests also never seem to make it into the turbogears > layer, so I think it's an issue in one of the middleware layers. > Lighttpd's documentation says it's default POST limit is 2GB, but the > requests don't show up in lighttpd's logs nor does an error show up in > any logs (lighttpd or cherrypy). I've tried with both the flup/scgi > and paste/scgi pairings. > > Has anyone seen this issue?
Here's an update on this issue. The error condition only exists with OpenSSL enabled lighttpd, and happens every 16K with POST data between 8317+(n*16384) and 16382+(n*16384) bytes. This happened because the buffer size used to read the SSL data and the buffer size used in lighttpd was different, and lighttpd never finished reading the POST data. This bug is fixed here: http://trac.lighttpd.net/trac/changeset/1095 --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" 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/turbogears -~----------~----~----~----~------~----~------~--~---
