Skip to main content

Hierarchical Join/Prune Attributes
RFC 7887

Yes

Alvaro Retana

No Objection

(Alexey Melnikov)
(Alissa Cooper)
(Ben Campbell)
(Benoît Claise)
(Deborah Brungard)
(Jari Arkko)
(Joel Jaeggli)
(Mirja Kühlewind)
(Terry Manderson)

Note: This ballot was opened for revision 07 and is now closed.

Alvaro Retana Yes

(Alexey Melnikov; former steering group member) No Objection

No Objection (for -07)

                            

(Alissa Cooper; former steering group member) No Objection

No Objection (for -07)

                            

(Ben Campbell; former steering group member) No Objection

No Objection (for -07)

                            

(Benoît Claise; former steering group member) No Objection

No Objection (for -07)

                            

(Deborah Brungard; former steering group member) No Objection

No Objection (for -07)

                            

(Jari Arkko; former steering group member) No Objection

No Objection (for -07)

                            

(Joel Jaeggli; former steering group member) No Objection

No Objection (for -07)

                            

(Kathleen Moriarty; former steering group member) No Objection

No Objection (2016-04-19 for -07)
I support Stephen's comments.

(Mirja Kühlewind; former steering group member) No Objection

No Objection (for -07)

                            

(Stephen Farrell; former steering group member) No Objection

No Objection (2016-04-19 for -07)

Two related but minor points:

- section 5, last para: can a router always know for sure that
everyone who'll get a message sent on an interface knows about
this new encoding? If not, the "MUST NOT" here is incorrect.
One could implement a MUST NOT that said to only send this to
routers who'd expressed the hello option, but saying that the
MUST NOT applies to everyone accessible from the sender's
interface is arguably not implementable. The real-world result
would be the same though, even if one's code would better match
the latter way of describing the MUST NOT, so not that big a
deal.

- section 6: If a router had a bug that caused it to crash (or
do bad stuff) when it unexpectedly receives a message with this
new encoding, then if I faked a message with the new hello
option to a peer of that router, I might be able to cause
someone else to crash the victim/target.  There are probably
other bad things I can do if I can fake a hello like that, but
this is perhaps a new one. I'm not claiming that's worth a
mention though, unless the authors/chairs/shepherd want to add
it.

(Suresh Krishnan; former steering group member) No Objection

No Objection (2016-04-19 for -07)
Section 5 contains the following text 

"Note that it also needs to include the Join-Attribute Hello option as specified in [RFC5384]."

but it does not talk about what exactly happens if a message is received with the Hierarchical Join/Prune Attribute but without the Join-Attribute. Can you clarify what happens in this case? I would also prefer it if this can be reworded as a MUST.

(Terry Manderson; former steering group member) No Objection

No Objection (for -07)