On 22-aug-2007, at 4:54, Jun-ichiro itojun Hagino wrote:
that's true, but when link MTU is different, it gets very nasty.
IEEE 802 standards do not permit variation in the link MTU
for Ethernet. Attempts to persuade IEEE 802 to approve use
of jumbo-MTUs (e.g. 9180 bytes + Ethernet framing) have
consistently failed within the IEEE 802.
This is strange, I never got the original message.
So I'm not sure why this point was brought up, but let me jump in and
point to my draft:
http://www.ietf.org/internet-drafts/draft-van-beijnum-multi-mtu-00.txt
We've had some discussions on the internet area mailinglist and I'm
hopeful that it will be possible to standardize a mechanism to make
nodes with different MTUs live on the same subnet and use the maximum
packet size supported by both neighbors rather than all nodes
conforming to the MTU of the least capable system.
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf