On Wed, 2015-11-18 at 15:56 +0100, Hannes Frederic Sowa wrote: > On Wed, Nov 18, 2015, at 15:45, Lorenzo Colitti wrote: > > On Wed, Nov 18, 2015 at 10:31 PM, Hannes Frederic Sowa > > <han...@stressinduktion.org> wrote: > > > I was not saying using tcp_close literally, sorry for not making that > > > clear, but just model the state transitions after tcp_close. At least it > > > seems like a normal close to me. > > > > But it shouldn't be a normal close. Consider calling SOCK_DESTROY on a > > socket that is streaming data to a peer. If SOCK_DESTROY results in > > the kernel sending a FIN, the remote side might think that the sender > > closed the connection gracefully, even though the local side aborted > > the connection. > > Oh, yes, I understand. The connection wasn't closed by the application > but by the administrator forcefully. So we should never indicate a > successful TCP shutdown with FIN but with RST. A TIME_WAIT period > actuallty still seems useful to me, maybe with different semantics, only > RST incoming data?
There is some confusion. TIME_WAIT are used to be able to send ACK packets to incoming valid packets. To send RST, you need nothing at all. TIME_WAIT are hints for normal ending connections, to handle old packets (and or duplicates) -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
