Re: Extending a /64

[Alexandre Petrescu <[e-mail]>]

Le 16/11/2020 à 16:16, Philip Homburg a écrit :
>> I suspect one would not oppose embedding a 39bit airplane ID into
>> an ULA prefix, at which point the arguments about id/loc split
>> might not disallow airplane ID into GUA either.  Because routing
>> towards ULA and GUA addresses happen in the same way.
> 
> Using ULA with an airplane ID is technically a violation of RFC 4193,
> Section 3.2.1:
> "Locally assigned Global IDs MUST be generated with a pseudo-random
> "algorithm consistent with [RANDOM].

'pseudo-random'...

But I meant this text from RFC4193 ULAs:
>    This document defines a specific local method to allocate Global IDs,
>    indicated by setting the L bit to 1.  Another method, indicated by
>    clearing the L bit, may be defined later.
> [end of partial citation from RFC]


> (In theory there is also ULA-C, but that doesn't allow it either)

I dont know what is ULA-C... and why does it disallow airplane ID in a 
prefix.

Whereas I fully agree IPv6 addresses must be allocated by a central 
authority, I disagree with over specifying even the ULA contents.  I 
thought one was free to do whatever one wanted in ULA space.

Alex

>