Re: [PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread David Miller
From: Jason Baron Date: Tue, 17 Jan 2017 13:37:19 -0500 > From: Jason Baron > > Using a Mac OSX box as a client connecting to a Linux server, we have found > that when certain applications (such as 'ab'), are abruptly terminated > (via ^C), a FIN is sent

Re: [PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread Rick Jones
On 01/17/2017 11:13 AM, Eric Dumazet wrote: On Tue, Jan 17, 2017 at 11:04 AM, Rick Jones wrote: Drifting a bit, and it doesn't change the value of dealing with it, but out of curiosity, when you say mostly in CLOSE_WAIT, why aren't the server-side applications reacting to

Re: [PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread Eric Dumazet
On Tue, Jan 17, 2017 at 11:04 AM, Rick Jones wrote: >up unneeded resources in a more expedient fashion. > > > Drifting a bit, and it doesn't change the value of dealing with it, but out > of curiosity, when you say mostly in CLOSE_WAIT, why aren't the server-side >

Re: [PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread Rick Jones
On 01/17/2017 10:37 AM, Jason Baron wrote: From: Jason Baron Using a Mac OSX box as a client connecting to a Linux server, we have found that when certain applications (such as 'ab'), are abruptly terminated (via ^C), a FIN is sent followed by a RST packet on tcp

Re: [PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread Eric Dumazet
On Tue, 2017-01-17 at 13:37 -0500, Jason Baron wrote: > From: Jason Baron > > Using a Mac OSX box as a client connecting to a Linux server, we have found > that when certain applications (such as 'ab'), are abruptly terminated > (via ^C), a FIN is sent followed by a RST packet

[PATCH net-next] tcp: accept RST for rcv_nxt - 1 after receiving a FIN

2017-01-17 Thread Jason Baron
From: Jason Baron Using a Mac OSX box as a client connecting to a Linux server, we have found that when certain applications (such as 'ab'), are abruptly terminated (via ^C), a FIN is sent followed by a RST packet on tcp connections. The FIN is accepted by the Linux stack but