[v6ops] public prefixes on loopback

[Michael Richardson <[e-mail]>]

Owen DeLong <owen=[e-mail]> wrote:
    > I’m still trying to comprehend a testing scenario that would require
    > this where GUA or ULA wouldn’t be a better choice. Loopback semantics
    > imply addresses that are identical across all machines,
    > quasi-automatically configured, and not routable outside of host
    > scope.

{same mis-comprehension for me too, but:}

    > This is not to be confused with virtual interface semantics
    > (which loopback interfaces iare frequently (ab)used for in IPv4) where
    > a routable address is assigned to non-physical interface to facilitate
    > reachability regardless of which subset of physical interfaces remain
    > up. (e.g. the use of configured routable “loopback” addresses on
    > routers to terminate iBGP sessions).

1. I really wish this practice had a good name, and did not involve the word
   "loop" or "back".  Because of the confusion.

2. The IETF has no really good descriptions of the above practice.
   (Why do *I* even know about it is a question... Like I always knew this?)
   I'm guessing it must be in routing-platform vendor training materials.

Maybe someone has some pointers to good material?
Again, as I said in another email, it would really be nice to be able to cite
this exact operational "pro-tip"

Owen, you write **ab** used in that somewhat cynical way, so you think that
maybe this is sometimes a wrong practice?
One could put the address on a "dummy" interface or other such thing.  Is it really different on any platforms?

--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide