André Warnier a écrit : > Anthony Catel wrote: >> Eric Covener a écrit : >>> On Sat, May 23, 2009 at 10:21 AM, Anthony Catel <p@xxxxxx> wrote: >>> >>>> I don't see any reasons why Apache couldn't be aware if the client >>>> hung >>>> up gracefully. >>>> >>> Patches are always welcome. >>> >>> >> Alright ;) Just wanted to know if it was a bug or a feature. >> > It's a feature, but not only of Apache. > Apache is blocked on a read() from the proxied-to server, waiting for > an answer. When it return from that read() (if ever), it will try to > write() something to the client socket (the proper response from the > proxied server, or an error message). Only then will it receive an > error back, because the client is now gone. > You don't get an error from a socket, unless you try to do something > with it. > > > --------------------------------------------------------------------- > The official User-To-User support forum of the Apache HTTP Server > Project. > See <URL:http://httpd.apache.org/userslist.html> for more info. > To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx > " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx > For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx > Hi André, But if the client hand-up, the read() loop (of the client) will return. So Apache know that the client is disconnected. Why Apache doesn't shutdown the socket to the proxied server when this happen ? --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See <URL:http://httpd.apache.org/userslist.html> for more info. To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|