For the first case where we would allow multiple Rule IPv4 prefixes per single 
Rule IPv6 prefix, I agree that downstream lookup would be problematic.
But don't we even have a problem even before we get to that point...The whole 
scenario presuppose that the two (or more) CEs would need the same delegated 
IPv6 prefix - this process can be independent of MAP and the IA-PD is assigned 
for example via DHCP server  (plain IPv6 address assignment).
It is simply not doable that a DHCP server hands out the same DHCP leases to 
two different clients. It seemed to me that this was not that big of a deal in 
this discussion and I just wanted to make sure I'm not missing anything.

For the overlapping cases I think as long as the bits from the shortest Rule 
IPv6 prefix length all the way to the PD length are unique, things should work 
but carving out non-overlapping subnets,ports  may be bigger trouble for the 
operators than not .

why /56 instead of /52? :P     => yes I agree it would be less wasteful.

Kris


From: Senthil Sivakumar (ssenthil) [mailto:[email protected]]
Sent: Tuesday, December 11, 2012 11:34 AM
To: Maoke
Cc: Poscic, Kristian (Kristian); [email protected]
Subject: Re: [Softwires] MAP-E/T overlaping ranges and domains

Hi Maoke,
Please see inline for Senthil2

From: Maoke <[email protected]<mailto:[email protected]>>
Date: Tuesday, December 11, 2012 5:21 AM
To: Senthil Sivakumar <[email protected]<mailto:[email protected]>>
Cc: "Poscic, Kristian (Kristian)" 
<[email protected]<mailto:[email protected]>>,
 "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: [Softwires] MAP-E/T overlaping ranges and domains

hi Senthil,

interesting discussion. see inline.
2012/12/11 Senthil Sivakumar (ssenthil) 
<[email protected]<mailto:[email protected]>>

_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires

Reply via email to