tcp/dccp: drop SYN packets if accept queue is full
authorEric Dumazet <edumazet@google.com>
Wed, 26 Oct 2016 16:27:57 +0000 (09:27 -0700)
committerDavid S. Miller <davem@davemloft.net>
Sat, 29 Oct 2016 19:09:21 +0000 (15:09 -0400)
commit5ea8ea2cb7f1d0db15762c9b0bb9e7330425a071
tree46fb718cf8da30f0cbb023e19081921cac3832a0
parent58effd7168e9b74252650a008b5229c7cf224c7b
tcp/dccp: drop SYN packets if accept queue is full

Per listen(fd, backlog) rules, there is really no point accepting a SYN,
sending a SYNACK, and dropping the following ACK packet if accept queue
is full, because application is not draining accept queue fast enough.

This behavior is fooling TCP clients that believe they established a
flow, while there is nothing at server side. They might then send about
10 MSS (if using IW10) that will be dropped anyway while server is under
stress.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
include/net/inet_connection_sock.h
net/dccp/ipv4.c
net/dccp/ipv6.c
net/ipv4/tcp_input.c