On 02/17/2017 11:12 AM, otroan(_at_)employees(_dot_)org wrote:
Fernando,
It is a simple logical consequence.
Middleboxes do not exist in the IPv6 architecture.
There is no interpretation of 2460 that can lead to an implementor inserting
headers other places than at the source.
Therefore, there is no interoperability issue in RFC2460 nor any ambiguity
that needs to be resolved in RFC2460.
lAt the Last 6man meeting there was a bunch of people arguing that
inserting EHs was "ok" because "it was not explicitly banned" -- the
"interpretation" :-) at the time being that "'processed' doesn't mean
'inserted'"
I asked you whether EH insertion was allowed, and you didn't answer the
question, even after asking multiple times (it should be in the audio
recordings).
Now you say the above. When did you change your mind?
Simple:
1) It was clear to everyone that EH insertion wasn't allowed.
2) Some folks came with a funny interpretation, such that EHs could be
inserted.
3) Lots of supporters of EH-insertion (mostly from the same company)
argued that "it wasn't forbidden". And this wasted lots of people's time.
So a clarification is warranted. That's it.
P.S.: And then folks wonder why people "give up"??
Me, I'm not wasting more time on this. It should be pretty clear to the
IESG what's going on here.
Thanks,
--
Fernando Gont
SI6 Networks
e-mail: fgont(_at_)si6networks(_dot_)com
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492