Re: [v6ops] Operational Implications of IPv6 Packets with Extension Headers - implementation-dependency

[Fernando Gont <[e-mail]>]

Hello, Eduard,

On 27/7/20 06:22, Vasilenko Eduard wrote:
> Hi Fernando, You did mention "implementation-dependent" in security
> section. But majority of people would bypass security section on the
> 1st pass of reading your draft. Except security people, others would
> look to functionality 1st, them may be security (may be not:-().
> Hence, you almost ignore "implementation-dependency" in such a way
> (putting it into security section).

Not sure what you mean...


> I propose to introduce additional
> 5.x, because (IMHO) "implementation-dependency" is the biggest
> problem in this draft - it is related to money/expenses (especially
> to replace hardware). It would be even bigger problem on next years
> with proliferation of SRv6, iFit, iOAM, BIERv6 and other abuses of
> IPv6 headers extensibility. People need to check hardware (and
> software) capability far in advance, or else could be negative
> surprise. 

What would you expect us to cover in such section, other than what we 
already cover in Section 5.1.1. (Packet Forwarding Engine Constraints)?

Thanks!

Cheers,
-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492