Re: 2.2.x: TCP lockups with tcp_timestamps

kuznet@ms2.inr.ac.ru
Sun, 18 Feb 2001 21:27:30 +0300 (MSK)


Hello!

> Yes. The 5.6.7.8 machine is connected to the Internet via a Linksys
> "router" that is also performing masquerade.
>
> I will be very angry if this turns out to be the culprit.

I am afraid it is. It corrupts packets preserving their checksum.

Look:

> Trace taken from 1.2.3.4 machine
...
> 20:29:35.179648 1.2.3.4.379 > 5.6.7.8.1053: P 16468:16528(60) ack 65211 win 31856 <nop,nop,timestamp 3397515 1679644> (DF)

and

> Trace taken from 5.6.7.8 (aka 192.168.1.102) machine
...
> 19:18:26.808820 < 1.2.3.4.379 > 192.168.1.102.1053: P 16468:16528(60) ack 65211 win 31856 <nop,nop,timestamp 3643345 3395672641> (DF)

5.6.7.8 received packet with garbage in timestamp area:

<3643345 3395672641> instead of <3397515 1679644>

And checksum is "corrupted" so that it remains right,
which is impossible to make occasionally.

Alexey
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/