MTU problems with GRE tunnels (fwd)

Jens Schweikhardt schweikh at
Wed Jul 8 07:36:33 UTC 1998

Dave & Dennis,

thanks a lot for your valuable insights. And of course,
apologies for making the cisco guys look like they didn't read
the RFC. Thank God this was written as a question with a smiley...

# Dennis Ferguson
# > I can tell you for sure that the Cisco routers do send the packets (GRE
# > or IP protocol 4) with a length which includes the IP header, just like
# > the RFC.  If you look I think you'll find that it is your kernel which is
# > subtracting out the IP header length before it hands the packet to you on
# > the raw socket.
# BSD Unix converts the length to host order and subtracts the IP header 
#    length.

Ugh! The platform in question is Solaris 2.5.1. So we have one more
datapoint. I'd consider this a bug unless it's documented somewhere
outside of the kernel sources. What do you think? Should I harass
the Solaris developers?

# Linux leaves the length in net order (don't know about subtracting).
# Windows leaves the length in net order and does not subtract the IP 
#    header length.
# -Dave


Jens Schweikhardt
SIGSIG -- signature too long (core dumped)

More information about the NANOG mailing list