Re: Non-Last Small IPv6 Fragments

[Christian Huitema <[e-mail]>]

On 1/10/2019 5:02 PM, Brian E Carpenter wrote:

> If an RFC7195 translator receives an IPv4 fragment, it will translate it into an IPv6 fragment. So if the IPv4 fragment is "small" by some definition, so is the IPv6 fragment.
>
> I think that's a legitimate route to non-last fragments shorter than 1280, regardless of what is considered normal for IPv6.


There is that, and there is also a semi-legit other case that involves
middleboxes. Suppose the end system "knows" that the PMTU is 1400 bytes,
and splits the original UDP datagram, insert fragment headers, etc.. The
fragments are 1400 bytes. And then, somewhere along the path, some
helpful middlebox realizes that the PMTU is now 1280 bytes, so splits
each of the 1400 bytes fragments into one 1280 bytes fragment and
another 120 bytes long.

-- Christian Huitema