On Feb 11, 2008 9:24 PM, James <[EMAIL PROTECTED]> wrote: > > Hi again, > So a poster on the CherryPy list had a similar problem to the one I > described below, and he never found a solution. He suggested that > moving to using mod_wsgi rather than mod_rewrite might help, however. > > Documentation on running TG via mod_wsgi on Apache seems a little > light - is this a common setup, and can anyone give me any hints/tips/ > pointers? > I would read: http://docs.turbogears.org/1.0/Deployment or http://lucasmanual.com/mywiki/TurboGears#head-36b7eef1526da4fe58c73738c925f34f6bc93c1d
then refer to mod_wsgi page to tailor your setup to your needs. Lucas > As it stands, performance is so poor that we're actually considering > porting to a different framework, maybe even different language, > unless we can make some progress - at least identify where the > problems are. I'd really like to stick with TurboGears if at all > possible, but maxing out at 2 requests / second is not really > practical.. > > Any help appreciated! > James > > On Feb 6, 5:48 pm, James <[EMAIL PROTECTED]> wrote: > > Hi all, > > I'm trying to track down aperformanceproblem with my app. This is my > > > setup: > > Apache 2.2 (using mod_rewrite to proxy onto port 8080) > > TurboGears 1.0.4.2 > > CherryPy 2.3.0 > > SQLAlchemy 0.4.2p3 > > MySQLdb 1.2.2 > > > > I've instrumented my code to show where the time's being spent and my > > controllers and database are actually doing pretty well - no > > controller takes more than 0.5 seconds from entry to exit. However, > > there is a latency 5 to 10 seconds for the page loads. The > > throughput's pretty low and I'm not moving much data at all (a whole > > page is <400k, without clientside caching), so the problem is not with > > Apache. > > > > Therefore, the problem seems to be somewhere in between the requests > > hitting port 8080 and my controller method being invoked. > > > > My threadpool configuration: > > sqlalchemy.pool_size=50 > > sqlalchemy.max_overflow=50 > > server.thread_pool = 100 > > > > Firstly, it would be great to know in general terms what happens in > > between the request reaching port 8080 and hitting my controller. > > Also, I'd really like some tracing information out of CherryPy seeing > > as the delay seems to be in CherryPy or in the identity framework - > > setting up a DEBUG logger with a 'cherrypy' qualname barely produced > > any output at all... > > > > Also posting to CherryPy list. > > > > Thanks a lot! > > James > > > -- Vehicle Information Number. Check for Authentic VIN http://lucasmanual.com/vin TurboGears Documentation: http://www.lucasmanual.com/mywiki/TurboGears --~--~---------~--~----~------------~-------~--~----~ 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?hl=en -~----------~----~----~----~------~----~------~--~---
