Re: [babel] Unicast Hellos

[Markus Stenberg <[e-mail]>]

On 3 Apr 2017, at 11.50, Juliusz Chroboczek <[e-mail]> wrote:
> Two kinds of Hellos, using different TLV types
> **********************************************
> 
> My suggestion was to have two Seqno counters, just like in David's
> proposal, but use a new TLV number for Unicast hellos.  This avoids the
> compatibility issue -- unicast Hellos will be ignored by legacy
> implementations, so unless an implementation periodically sends multicast
> Hellos, it will be ignored by legacy implementations.
> 
> This used to be my preferred solution.  David has convinced me to use
> a flag, but I’m willing to reconsider.

In general having both unicast and multicast hellos seems worthwhile just to estimate both types of payloads’ characteristics on the link. 

This is my preference of the listed options. Flag based correct behavior feels bit dangerous.

I am not sure I am entirely happy with any of the listed options though. I have considered slightly different scheme treating unicast and multicast as separate logical links for hello purposes (=> disjoint sequence number spaces) but that leads to issues in mixed environments with both legacy implementations and ones supporting unicast hello spec. If writing v2 version of protocol without backwards compatibility requirements, I would probably lean that way (on the wire, hellos would look same, but sequence # spaces would work bit differently based on uni<>multicast).

Cheers,

-Markus