RE: status update for ESnet
Bob Fink <[email protected]> Tue, 16 Jul 2002 16:01:14 -0700
| Newsgroups | gmane.network.ipv6.exchanges |
|---|---|
| Message-ID | <[email protected]> |
Michel, I've asked the engineer for his reason, so you should get a reply. I don't really know the ratioanle. Bob === At 09:35 AM 7/16/2002 -0700, Michel Py wrote: >Bob, > > > Interfaces addresses are globally unique and are /64s. > > Loop back addresses are globally unique /128s from a common /64. > >I am curious about the /128 loopbacks. I typically don't treat loopbacks >differently than any other interface on a regular subnet, meaning I give >them a /64 as well. > >I understand that, contrarily to /127 commonly used on ptp links, /128 >addresses do not violate [addrarch], so doing this is ok. > >However, the use of /128s out of a "loopback pool" /64 makes loopbacks >unaggregatable, where using /64s could make them aggregatable. > >I am curious about the reasons behind the choice of /128s for loopbacks. >Is there a deliberate choice to make them unaggregatable (which has >operational advantages as well as drawbacks)? > >Thanks >Michel.