Hi all, this is my first post here so allow me to give my compliments to Mr. Di Pierro and all the people that works on this wonderful project. I do not have a big experience in web programming, i came from "offline" :) programming and triyng to add value to the products of my software house with online capabilities so i'm in "learning mode" (about 1 month now) with python uwsgi nginx and all the web stuff in general
i have implemented a nginx+uwsgi_emperor+web2py multisite solution on a new server that will serve (in future) different sites and services for our company and customers. (it works great for now :) i'm a little worried about a sentence in this page: http://projects.unbit.it/uwsgi/wiki/ThingsToKnow I do not understand this paragraph: "If an http request has a body (like a POST request generated by a form) you have to read it in the application." [...] "If you do not do it, the communication socket with your webserver could be clobbered." i want to implement a lot of restful calls in many parts of my sites and i'm ok reading the body of a legit request but i don't know how to handle NON legit requests... do i need to read the body in any case?? suppose i want to check a POST request in code and discard the request raising HTTP(500) if i do not like the user-agent (is only an example): def call(): session.forget() if request.env.http_user_agent == "NonGoodBrowser": raise HTTP(500) return service() also if a browser send me a lot of wrong POST requests with null-filled bodies to every page of my site, do i need to read it all and use server resources? suppose i don't want to activate the --post-buffering option for the same reason of server resources drain. what you suggest me? thanks in advance, Giovanni V. --
