My web2py application is running on Webfaction's servers over uwsgi/nginx. When the servers are started, four instances of nginx and four instances of uwsgi are spawned, with each nginx instance taking up ~18M of RAM, and each nginx instance taking up a couple of M. Accessing any part of the page (i.e. clicking a menu item) adds about 30M to one of the uwsgi instances. Clicking a menu item again can add between 2 and 30 M to any of the uwsgi instances (it will be 30 M if nothing has yet been added to the RAM usage of that particular instance). This happens until my memory limit (256 M) is exceeded. I think I've mitigated the behavior somewhat by moving some things from models to modules, and by serving static content separately, but I think my memory usage is still quite excessive. Is this normal behavior?
Should I try to move all databases from models to modules? Is it ok to have globals (i.e. settings = Storage()) declared in models? Is it alright to have import statements in models? In the controllers, is it better to have import statements inside or outside of functions? As far as uwsgi goes, would it make sense to use the 'reload-on-rss' directive to restart when excessive memory is used? Are there any major downsides to this? Any other tips for optimizing uwsgi? Any other tips for diagnosing the sources of excessive memory consumption, and diminishing this consumption? Thank you. -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
