Re: slaac-renum: New rev of draft-gont-6man-slaac-renum (Fwd: New Version Notification for draft-gont-6man-slaac-renum-08.txt)

[Fernando Gont <[e-mail]>]

Hello, Philip,

On 22/5/20 12:35, Philip Homburg wrote:
>> That's a bold statement.
>>
>> No, the algorithm will now fail.
>>
>> If the algorithm were to not consider ULAs separately, then:
>> 1) If the local network uses, say, ULA and non-ULA, *and*
>> 2) The router advertises prefixes in separate packets, *and*
>> 3) There's a lot of packet loss
> 
> You are fixing one broken edge case with another.

FWIW, one case is common, the other one isn't. But anyway... point taken 
--  That's not how the current algorithm is specified, anyway, so let's 
focus on what's actually in the I-D.



>>> We want something that quickly after the router reboots causes the
>>> old prefix to be dropped. Yet, the algorithm should not drop any prefixes
>>> due to modest packet loss. Currently, 2 out of 3 RAs can be lost without
>>> ill effects.
>>
>> LTA_INVALID is your friend. As noted, one might even set it to "Router
>> Lifetime".
> 
> Like I said, if an address is no longer preferred it can already break
> new connections.

That wouldn't be the case. We're talking about:

1) Current algorithm as specified in the I-D, which separates processing 
of ULA and non-ULA.
2) LTA_INVALID set to, say, "Router Lifetime"

Would that address your concern?



>> I believe this seems to me like an over statement.
>>
>> As with my description above, for this to happen:
>> 1) The same router should be advertising multiple non-ula prefixes, *and*
>> 2) the router must be advertising different prefixes in different RAs, *and*
>> 3) Multiple RAs need to be lost, for a considerable period of time, *and*
>> 4) A new type of address (?) would need to be specified
> 
> No need to specify a new type address. Today anybody can run with
> 2 different ULA prefixes that have different semantics. For example,
> one can be used to communicate purely internally, another to communicate
> with closely related organisations.
> 
>> Me, I consider this very unlikely.
> 
> I think it is very likely that people come up with complex addressing
> schemes. It is also very likely there will be packet loss. Certainly on
> radio link. And give the rate at which new stuff gets added to RA, it is
> very likely that RAs will be split.

FWIW, Section "4.4.  Conveying Information in Router Advertisement (RA) 
Messages" of the draft recommends all PIOs in the same RA.




>> That said, I don't mind setting LTA_INVALID to the "Router Lifetime",
>> and/or recommending that a host sends an unicasted RA at LTA_DEPRECATE,
>> and another one at, say, time LTA_DEPRECATE + (LTA_INVALID -
>> LTA_DEPRECATE) / 2.
> 
> In my opinion having the host send a unicast RS (I assume that's what you
> meant) and retransmitting that, just before a prefix is no longer preferred
> is a lot more complex than my algorithm.

What I was meaning was:

1) Set LTA_INVALID to "Router Lifetime", and/or,
2) Send at least one unicast message at time T= LTA_DEPRECATE + 
(LTA_INVALID - LTA_DEPRECATE) /2 . This RA, besides being an additional 
one, is sent as unicast, which tends to have a lower loss probability.

I blieve either of them should address your concern -- at the time you 
wait till "Router Lifetime", enough RAs should be getting through since 
other wise ND itself would fail (default router would be reoved, etc).

Thoughts?

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