Hello, all I would like to know how does HAProxy work internally in terms of http data flow?
For example, when a client connected to HAProxy server, 1. haproxy receives all HTTP request data first, then forwards the data to backend, then receives all HTTP response data from backend then forwards to client? 2. haproxy forwards every bits of data to backend as soon as it get data from client and forwards anything to client as soon as it gets from backend? Or any other way? If 1 is true, if client is to download a big iso file, does that mean haproxy will cache a whole iso file in its memory?? If 2 is true, we know that haproxy can do some `http-request deny` based on URL, if haproxy just forwards what it gets, how can haproxy do these deny actions? Or maybe haproxy first get all HTTP request header from client, then apply these filters then forward every bits of http request body to backend, and forward anything sent by backend to client? Thanks
