Hi
I can't seem to find this mentioned anywhere. My 2.4 machine
sets 2 bits in the TCP header between the 4 data offset bits and control
flags. Like so:
11/29-20:25:42.695096 24.67.152.12:62373 -> 192.18.97.241:80
TCP TTL:63 TOS:0x0 ID:0 DF
21S***** Seq: 0x74E55D1A Ack: 0x0 Win: 0x16D0
TCP Options => MSS: 1460 SackOK TS: 1767921 0 NOP WS: 0
or hex a0c2 / binary 10100000 11000010
^^
This packet is never ACK'd by http://www.sun.com and the only difference I
can see from one that is are these two bits. RFC793 says they must be zero.
Is 793 still current? Has anyone else seen this?
Mark
Date: Wed, 29 Nov 2000 20:53:32 -0800 (PST)
From: Mark Sutton <[email protected]>
^^
This packet is never ACK'd by http://www.sun.com and the only difference I
can see from one that is are these two bits. RFC793 says they must be zero.
Is 793 still current? Has anyone else seen this?
ECN, RFC2481
Turn it off with /proc/sys/net/ipv4/tcp_ecn or tell Sun to update the
software running on their firewall.
Later,
David S. Miller
[email protected]
> I can't seem to find this mentioned anywhere. My 2.4 machine
> sets 2 bits in the TCP header between the 4 data offset bits and control
> flags. Like so:
ECN
> can see from one that is are these two bits. RFC793 says they must be zero.
> Is 793 still current? Has anyone else seen this?
793 is still current but remember bits of it are superceded by a couple of
thousand other documents.
Alan