[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [DNA-BOF] Detecting Network Attachment BoF description (rev 2)



Hi Alper

> I think we need to differentiate between link/subnet/prefix changes.

Yes, 3 entities can change independently as you said below. That made 
MD work more complicated. 

> On the other hand, the link might not change (i.e., host is still connected
> to the same media), but the configuration parameters change. For example
> router renumbering. Now host has to react to this too.

Here is an another example of change with the same link. .

Assume there are two Routers R1 and R2 in a link which advertise different 
prefixes A:: and B:: respectively. Also assume there is a Host H in a link, 
which has an IP address with prefix A:: and default router R1. 

If R1 suddenly fails, H should change its address and default router. It has 
the exactly same effect as H physically moves from R1 to R2. When H has an
on going sessions, it is important to detect change as soon as possible. 

                                     |                                    |
                                 __|____                       __ |____
                                |            |                        |            |
                                | Router |                       | Router |
                                |___1__|                       |___2__|                 
                                     |  A::1                            | B::1
                           A::      |                                    |      B::  
                                 __|__________________|_____       
                                                        |  A::2
                                                    __|___                     
                                                   | Host  |                    
                                                   | H       |                       
                                                   |___ __|                        

We came across several similar problems in MD work. But I keep myself 
from delving in here (barely :-) because I thought at first we'd better make
an agenda for Vienna. 

Best regards

JinHyeock