Re: TCPv4 bad checksums

Thomas Molina (tmolina@probe.net)
Fri, 25 Dec 1998 00:20:49 -0600 (EST)


> The commonality with ippp, ppp and slip would be slhc. However
> I _only_ seem to see it on routed packets (I'm not masquerading).
>
> At home I see it on my PC talking directly via PPP to the Internet
> (via Demon) - but I've only noticed it when talking to RedHat sites.
> Could the commonality be whether the packets go through a Linux
> router or not?

my last bad "storm" of bad checksums was the night my mirror program
grabbed all the XFree 3.3.3 updates from the RedHat site. However, that
isn't the only time I get them.

I'm about half-convinced that the common thread in my case is which line
I end up on at the terminal server at my ISP. I don't have enough data
at this point to say that conclusively, nor have I had the time to make
a detailed analysis of what data I do have. It's especially aggravating
that I can't reliably trigger it yet. I went for a week with no
checksum errors, then I got them for several days in a row. I haven't
yet got to the point of logging all the TCP data and doing hand
validation of suspect packets. I probably need to be able to trigger
the effect before I go there.

One participant in this thread suspects Ascend equipment is one of the
main culprits. Just on general impressions, he could be right from my
point of view.

-
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/