Date: Mon, 20 Mar 2017 17:31:08 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-net@FreeBSD.org Subject: [Bug 217637] One TCP connection accepted TWO times Message-ID: <bug-217637-2472-5I0muHRCbu@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-217637-2472@https.bugs.freebsd.org/bugzilla/> References: <bug-217637-2472@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D217637 --- Comment #66 from Michael Tuexen <tuexen@freebsd.org> --- (In reply to slw from comment #65) > This is wrong behaviour. This is cause lost of server data. No, the loss of data is caused by the application calling close() *before* incoming user data arrived. So the TCP stack on the server has to drop that user data. > pwrite(); > close(); > This is graceful termination. Sure. This is what the application triggers. However, when user data arrives after the close call, this gets ungraceful, since this user data can't be deliver= ed to the user anymore. To avoid this, the application could call shutdown(..., SHUT_WR) to trigger the sending of the FIN, then process incoming data until a FIN f= rom the peer arrives and then calling close(). But the application didn't. This wou= ld be more in line of how RFC 793 describes a connection termination. Please note that the CLOSE primitive in RFC 793 maps to a shutdown(..., SHUT_WR) system call, not to the close() system call. Bad naming...=20 I don't see text in RFC 793, where it is required that you continue to proc= ess a connection after you know that it failed. I think the RFC doesn't cover t= he case where the application says "I don't want to receive anymore". --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-217637-2472-5I0muHRCbu>