Re: [manet] NHDP: Link quality based on seq-num

["Teco Boot" <teco at inf-net.nl>]

>-----Oorspronkelijk bericht-----
>Van: Thomas Heide Clausen [mailto:ietf at thomasclausen.org]
>Verzonden: dinsdag 10 november 2009 0:09
>Aan: Teco Boot
>CC: 'Ulrich Herberg'; 'Dearlove, Christopher (UK)'; manet at ietf.org
>Onderwerp: Re: [manet] NHDP: Link quality based on seq-num
>
>
>On Nov 10, 2009, at 12:02 AM, Teco Boot wrote:
>
>> Hi Ulrich,
>>
>>> -----Oorspronkelijk bericht-----
>>> Van: Ulrich Herberg [mailto:ulrich at herberg.name]
>>> Verzonden: maandag 9 november 2009 23:41
>>> Aan: Teco Boot
>>> CC: Dearlove, Christopher (UK); manet at ietf.org
>>> Onderwerp: Re: [manet] NHDP: Link quality based on seq-num
>>>
>>> Teco,
>>>
>>> On Mon, Nov 9, 2009 at 11:31 PM, Teco Boot <teco at inf-net.nl> wrote:
>>>> What if someone defines a reliable transport and suppress
>>>> repeating messages?
>>>
>>> In that case, that mechanism of using packet seq. numbers would not
>>> work. But I think this is okay, since the NHDP draft does not mandate
>>> this mechanism. If one decides to use packet sequence numbers for
>>> indicating lost packets, one probably has to assume a non-reliable
>>> transport transport protocol such as UDP.
>>
>> You missed my point.
>> The protocol would suppress the repeated info.
>>
>
>Then I do not understand. Your L2 (or "something-behaving-as-L2")
>below IP is reliable. That's fine, you'll get 100% of packets across.

100% delivery ratio is the goal of reliable multicast.
But at cost of delay and other annoying aspects.
It runs at layer 4 (instead of UDP or vanilla IP).


>A great L2 and a great feat if you can do reliable broadcast to an
>unknown recipient set (I'd like to learn more if you have one at hand).
>
>>>> Are we sure we want to use missing sequence numbers as
>>>> indication of missed packets?
>>>
>>> No, we don't want to mandate it. It is just *one* possible way, under
>>> certain assumptions (no reliable transport protocol).
>>
>> But it is an impossible way, as defined today in the spec.
>
>How so? Your L2 does not lend itself to "counting packets" - so you
>won't, you'll use something else (more appropriate to your L2) for
>establishing the quality.
>
>This is not only enabled by the spec, but even recommended.

Back to my first comment, the spec needs a fix. Then, we can recommend
to use it. But only when sequence numbers are used. 
We could regret this in the future, when working on optimizing TC load.


>>>> What if other reasons caused the loss, e.g. tail drop during
>>>> congestion caused by the MANET protocol itself?
>>>
>>> Well, in that case the router will face problems anyway, since it
>>> does
>>> not receive control traffic any more.
>>
>> And send more traffic. See a nearby museum for what can happen if a
>> system
>> is speeding up its own explosion.
>
>I think that's an inappropriate reference.

I've seen network meltdowns, including MANETs. This meltdown term is
borrowed
from nuclear catastrophes.


>I also think that it's an inappropriate and incorrect characterization
>of what NHDP would do....

Did I say so?

But yes, network meltdowns due to chain reaction is a real world concern.
It is definitely not specific to NHDP/OLSR. But these are also not immune.


Can we go back to my first comment, and fix this minor issue?


Teco.


>
>Thomas
>
>
>>
>> Teco.
>>
>>
>>>
>>> Ulrich
>>>
>>>>
>>>> Teco.
>>>>
>>>>> -----Oorspronkelijk bericht-----
>>>>> Van: manet-bounces at ietf.org [mailto:manet-bounces at ietf.org] Namens
>>>>> Ulrich Herberg
>>>>> Verzonden: maandag 9 november 2009 22:11
>>>>> Aan: Dearlove, Christopher (UK)
>>>>> CC: manet at ietf.org
>>>>> Onderwerp: Re: [manet] NHDP: Link quality based on seq-num
>>>>>
>>>>> On Mon, Nov 9, 2009 at 8:03 PM, Dearlove, Christopher (UK)
>>>>> <Chris.Dearlove at baesystems.com> wrote:
>>>>>> My, not fully worked out, thought is that per interface
>>>>>> packet sequence numbers should be allowed, or even
>>>>>> preferred. That leaves a procedural issue of how to get
>>>>>> from where we are to where we want to be.
>>>>>
>>>>> I agree that we should allow packet sequence numbers per
>>>>> interface. At
>>>>> the moment, I cannot think about a good reason for not preferring
>>>>> them, but maybe there is...
>>>>>
>>>>>
>>>>> Ulrich
>>>>> _______________________________________________
>>>>> manet mailing list
>>>>> manet at ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/manet
>>>>
>>>>
>>
>> _______________________________________________
>> manet mailing list
>> manet at ietf.org
>> https://www.ietf.org/mailman/listinfo/manet