On Nov 6, 2007 1:02 PM, Norman Rasmussen <[EMAIL PROTECTED]> wrote: > mmm, I think the timer code is more broken than I originally thought. > This means that the ping generation code has, um, never worked.
(which means that if the yahoo transport was behind certain firewalls, user connections would timeout after ~5 minutes of in-activity) FYI: I've just fixed this in CVS. CPU load should be less, but obviously network load will go up a bit. Depending on feedback, I might add an option to 'reset' the connection idle counter each time a packet is sent. (The keep-alive duration is suggested dynamically by the yahoo server, default is a keep-alive per-user every 4 minutes. I've made it randomize the point at which keep-alives are sent so that the network usage doesn't peak because of them too) -- - Norman Rasmussen - Email: [EMAIL PROTECTED] - Home page: http://norman.rasmussen.co.za/ _______________________________________________ py-transports mailing list py-transports@blathersource.org http://lists.modevia.com/cgi-bin/mailman/listinfo/py-transports
