DSM> Ok, cool, keep me informed. Another bogon I know about are
DSM> Solaris-2.5 and older machines without all the correct patches
DSM> installed, these can output bad checksums too.
I get (way too many) bad TCP Checksums between a RedHat 4.2 machine
running ppp-2.3.3 on the latest kernel from RedHat (2.0.32) and
a RedHat 5.0 machine running ppp-2.3.3 on the 2.1.95 kernel (and also
a lot of kernels before that).
2.0.32 to 2.0.32 works perfectly in the same setup.
When I check /proc/net/dev on both machines, only the 2.1.95 one
reports errors. Apparently all frames received by the 2.0.32 machine
are fine.
If the checksum algorithm used is the same for both transmit and
receive, it would be expected that _both_ hosts would complain about
bad checksums.
Also, if the problem is in a lower layer (such as PPP), it would be
expected that both directions suffer equally.
Any ideas?
Benny
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu