Re: [babel] Comments about rfc7298bis

[Denis Ovsienko <[e-mail]>]

---- On Wed, 02 May 2018 18:54:03 +0100 Juliusz Chroboczek  wrote ---- 
>>>> What happens if the number of neighbours is so large that the complete set 
>>>> of TC/PC echos does not fit in a full-size packet? 
> 
>>> The same thing happens as if there is no authentication at all and the 
>>> number of neighbours is so large that the complete set of IHU TLVs does 
>>> not fit. 
> 
>> I'm not sure I understand. The set of IHUs can be easily split across 
>> multiple packets -- the absence of an IHU does not imply that the 
>> neighbour has been dropped. If I understand correctly (correct me if I'm 
>> wrong), your draft requires sneding the full set of IHU+TC/PC echo in 
>> a single packet. If I'm right, then a mechanism is needed to deal with 
>> overflow, no? 
> 
>Denis, 
> 
>Did I miss an e-mail? I don't see an answer to the above. 

1. Terminology: the data unit is called a TS/PC number, this has been the term since 2012.
2. Protocol encoding: the "echo" TS/PC sub-TLV is specific to an IHU TLV, the additional requirement for IHU TLVs grouping or placement you describe does not exist.

If you go back a couple messages up this thread, hopefully you will see we had discussed both points before.

-- 

    Denis Ovsienko