From: Zilo <ko...@gm...> - 2005-01-19 09:19:11
|
On 01/18/05 17:08:27, Don Seiler wrote: > On 17:06 Mon 17 Jan , Gary Kramlich wrote: > [...] > > As for a release, that's up to don. :) >=20 > Such power! If it were up to me, I'd say we're pretty close. There > are those two bugs that I'd like to see tracked down. Actually I =20 > think the "crazy CPU" one was addressed by evew's flood patch. Yes, we didn't care of len=3D=3D0 in bnet_conn_input_cb() after the read(),= =20 so the code wasn't catching the disconnection, falling in loop... I didn't realize I catched the bug, cool :) > That still leaves the bad login one. Basically try to log in on a =20 > server with an account that doesn't exist on that server. Like I =20 > tried logging in with 'dtseiler' on europe.battle.net, or just make =20 > up a random account name to try. >=20 > Basically it will hang on "Waiting confirmation" for a while and then > say that you are Online, even though you aren't. You _will_ get PING > from battle.net server, but you won't be able to do anything or send > any commands. Yes, I'm thinking about this... maybe we can catch it with a timer, a =20 sort of "if welcome INFO isn't received until 15 secs after login, then =20 we aren't *really* logged in" Or we can set a max user/pass attempts... I see that for a real one we =20 need to retry only one time (even if I don't know why), else it'll need =20 several attempts before the "Waiting confirmation" will come... It doesn't sounds bad to me :) So, I'll code the last one and upload a patch... notice me if you login =20 with more then two user/pass attempts! cya |