actually cherrypy can be faster than apache/mod_wsgi sometimes. but i often use mod_wsgi do recommend mod_wsgi as it's quite stable and flexible. i am just wondering how u crash cherrypy even if it's overloaded.
-vince On Jan 21, 3:59 pm, voltron <nhy...@googlemail.com> wrote: > Thanks Massimo, I will try that out > > On Jan 20, 8:05 pm, mdipierro <mdipie...@cs.depaul.edu> wrote: > > > Voltron, > > > if you have more than 100 cuncurrent users you should not use > > wsgiserver. You need to use mod_wsgi. I would not swear on wsgiserver > > on that heavy load. > > > You also have 100 database connections. Make sure you are suing > > pooling. > > > Also make sure your app is bytecode compiled. > > > My bet is that the database is a bottle neck. > > > You can add if statements in your model depending on request.function > > so that for example, you do not define tables when you do not need > > them. > > > Massimo > > > On Jan 19, 11:40 pm, voltron <nhy...@googlemail.com> wrote: > > > > Thanks Massimo. > > > > All I have in my Apache config are the directives for reverse > > > proxying, to my greatest dismay, it just does not scale well. I > > > noticed that anything above 100 concurrent users would just shut the > > > CherryPY server down. I have to routes activated. Another strange > > > thing are the performance dips, the response could be fast, but > > > sometimes they just grind to a halt for no reason, this is not > > > dependent of database queries oor other processes, just dynamically > > > rendered "static" pages. > > > > This is a headache > > > > On Jan 19, 11:46 am, mdipierro <mdipie...@cs.depaul.edu> wrote: > > > > > Voltron, > > > > > sorry but I do not have enough information. First of all. Is this with > > > > a specific application or can you reproduce the behavior with the > > > > welcome app as well? > > > > > Can you show us you apache configuration as well as your routes.py? > > > > > Massimo > > > > > On Jan 19, 4:29 am, voltron <nhy...@googlemail.com> wrote: > > > > > > I have been having serious performance problems. Some tracking down > > > > > led to the observation that when using Apache as a reverse proxy, > > > > > web2py runs up the CPU usage and shuts down after 100 requests. The > > > > > strangest thing is that when this happens, the CPU is only at 60% > > > > > usage. > > > > > > Is there some setting that I can tune to avod this? > > > > > > This is a root server with 2GB memory and an 2GHZ Athlon CPU > > > > > > Thanks --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "web2py Web Framework" group. To post to this group, send email to web2py@googlegroups.com To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/web2py?hl=en -~----------~----~----~----~------~----~------~--~---
