Re: [DNA] Model: how to treat "link down" events

[Erik Nordmark <erik.nordmark@sun.com>]

Michael.G.Williams@nokia.com wrote:
>  
> The very last statement:
> 
>> We can't rely on timely 'link down' events for DNA, hence it makes
> sense to only rely on 'link up'
>> events, which is what CPL and the DNA solution do.
>>
>>    Erik
> 
> If timely link down events were available (say from IEEE 802.21), would
> that help things?

I don't think they would help with DNA, since DNA on a single-homed host 
can't do much useful from knowing that the network disappeared. The 
upper layers, like application protocols, could stop waiting for the 
network to respond, but perhaps they should be patient and wait for it 
to come back.

For multihomed hosts, in particular those having protocols like MOBIKE 
or SHIM6, getting a timely link down event helps to have it quickly 
switch to using an alternate network interface.

    Erik