Re: 2.2 TCP resets
Andi Kleen (ak@muc.de)
26 Jul 1999 15:37:05 +0200
shevek@tirin.openworld.co.uk (Shevek) writes:
> K6-2/450 CPU.
> gcc-2.7.2.3 or egcs-1.1.2
> Slackware-4.0 reasonably unmodified.
> Also running appletalk, smbfs but nothing else 'weird'.
> Standard 10Mb hubs running into the Portmaster, and from there into a BT
> NTU at 64KBps and down a leased line to VBC (UK)
>
> When I send weird packets to the Portmaster, I get back a lot of packets
> with a RST bit set in the header. (Things like nmap -O) I'm pretty sure
> it's not a firewall problem in the portmaster, as the firewalling is done
> by the Linux boxes themselves, and it should return an ICMP not a TCP RST.
>
> I have seen no trouble over the local ethernet, or between 2.0.x, 2.2.x,
> 95, NT or MacOS, and I don't seem to have trouble jumping from one half of
> the subnet to the other over the portmaster, it's only down the leased
> line.
>
> Has anyone else seen this, and has anyone any suggestions for fixing it? I
> think we're about to upgrade the ComOS version on the IRX and replace the
> leased line.
Sounds like broken VJ compression.
Try
echo 0 > /proc/sys/net/ipv4/tcp_timestamps
echo 0 > /proc/sys/net/ipv4/tcp_sack
echo 0 > /proc/sys/net/ipv4/tcp_window_scaling
-Andi
--
This is like TV. I don't like TV.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/