On 21-jul-2005, at 1:01, Ryota Hirose wrote:

As I read it, RFC2464 says that the maximum *default* MTU is 1500,
and, anything larger than that must be manually configured.  It
doesn't say that it is prohibited.

RFC2464 says as following:

             This size may be reduced by a Router Advertisement [DISC]
containing an MTU option which specifies a smaller MTU, or by manual
    configuration of each node.

I'm sorry for my poor English knowledge, but I think this sentence
permit to decrease MTU from default size by manual configuration.

OK, this sentence or others in the section don't prohibit increase
MTU.  Since decrease MTU is permitted and there is no mention to
increase, I understood that the increase MTU is prohibited.  Isn't it
a common sense?

Well, that's a philosophical question: is everything that isn't explicitly disallowed allowed by default, or is everything that isn't explicitly allowed disallowed by default?

My common sense tells me that the authors of RFC 2464 didn't consider the case where the MTU would legitimately be larger than 1500 bytes. They did consider the case where router advertisements contain an MTU that is apparently incorrect, because it's larger than the standard allows.

Ideally, the MTU would be a neighbor discovery option, so that the highest possible MTU can be used between two systems on a subnet, without requiring that all systems on a subnet support the same MTU size. However, this is problematic because switches and hubs also have MTU limitations that can't be discovered this way.

In any event, RFC 2464 doesn't say it's disallowed to configure an MTU larger than 1500 bytes on an ethernet subnet, and it also doesn't provide any reasons why doing so would be bad. The only thing it says is that the MTU option in router advertisements can't be used to increase the MTU beyond 1500 bytes.

So in my opinion, an implementation that supports jumboframes should use the interface MTU for IPv6 by default, and reduce this MTU for IPv6 to the one in an MTU option in router advertisements, when such an option is received.

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to