On 31-jul-2007, at 10:11, Fred Baker wrote:
here's an interesting gotcha. Macs run a 1000/100/10 Ethernet
interface and run a 1500 byte MTU/MRU regardless.
(On many Macs, you can manually set an MTU size of upto 9000 bytes
regardless of the link speed.)
The router can correctly observe that it is 1 GBPS and send the
jumbogram, the switch can support the jumbogram, and have the Mac
not accept the packet.
Hence, there are variations of this that are beyond the router's
knowledge.
Right. Hence the new mechanisms that I propose in my draft:
http://www.ietf.org/internet-drafts/draft-van-beijnum-multi-mtu-00.txt
However, it would be good if benefits could be had without
introducing such a relatively large number of new options and messages.
My current thinking is that if hosts implement RFC 4821, they can use
a larger-than-standard MTU for the transport protocols that use RFC
4821, but they'd still need to maintain the standard MTU for packets
that aren't generated by those transport protocols OR they have to
implement the neighbor MTU negotiation laid out in my draft (or
something similar).
So for full end-to-end jumbo use across the internet the additional
mechanisms need to be implemented on all intermediate ethernets (or
jumboframes must be enabled administratively) but with only RFC 4821
you can still have out-of-the-box jumboframe capability for TCP and
maybe some other transports on the local subnet.
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
