Vishwas,

It's not a question of what's optional or not.  Erik gave a clear
example of an operational issue which is if a host sends a packet with
size equal to MTU of the link and if such a packet arrives at a router1
in such a network and router1 slaps its RH on the packet, the packet
size grows larger than the max MTU.  Now router1 forwards the packet to
router2 who issues an ICMPv6 Too Big error to the host.  The host stands
totally confused because the host did not exceed the link MTU.  

Also, tunneling used by transition mechanisms is orthogonal to this
operational issue discussion.  For tunneling with no use of RH, there is
well-defined behavior for not exceeding the MTU in RFC 2473.  

Hemant 

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of
Vishwas Manral
Sent: Monday, May 31, 2010 2:03 PM
To: Erik Nordmark
Cc: [email protected]; culler Culler
Subject: Re: Fwd: I-D Action:draft-hui-6man-rpl-routing-header-00.txt

Hi Erik,

The first thing is IPv6 MTU discovery is optional and the only
requirement is that the Minimum MTU is satisfied.

Also in cases like LLN the Routing paths may change so even after
discovery the Packet too big may come anyway. Besides PMTU issues you
mention are no different than a lot of cases such as tunnelling and
transition mechanisms.

For MTU discovery in cases when padding is done, the right MTU will be
found as the data packets will undergo the same behavior as the PMTU
discovery message.

Thanks,
Vishwas
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to