Subject: Re: Peculiar ICMP6 redirect rejection
To: der Mouse <mouse@Rodents.Montreal.QC.CA>
From: William Waites <>
List: tech-net
Date: 08/19/2002 10:41:53
>>>>> "derMouse" == der Mouse <mouse@Rodents.Montreal.QC.CA> writes:

    >> to be clear:  i can't make any advice as you  are using way too
    >> strange configuration  (trying to make  address portable across
    >> subnets using RIPng, nonstandard netmasks, and more).

    derMouse> As  for  nonstandard   netmasks,  I'm  not  sure  what's
    derMouse> "nonstandard"  about  any  of  the netmasks  I'm  using,
    derMouse> possibly   excepting  using  a   /128  on   a  broadcast
    derMouse> interface;

Might I suggest  putting that /128 on the  loopback interface and then
advertise it as reachable through the LL address of your ethernet NIC? 

    derMouse> if it  makes you happier, I can  go to a /127
    derMouse> or  /126  there.   It's  not  terribly  helpful  to  say
    derMouse> something like that without indicating what a "standard"
    derMouse> netmask is,  or at least a  pointer to where  I can find
    derMouse> that information.  

I believe  itojun was referring to  the standard practice  of making a
subnet /64. 

    derMouse> a /112  or /128  netmask in some  sense, then I  have to
    derMouse> ask, why  on earth not??   What does it break,  and what
    derMouse> should I do instead?

It would certainly break the  eui64 calculation, but you are assigning
addresses manually  so that  shouldn't be a  problem. It seems  like a
/112 should  work, but it  may cause some  subtle issues that  I don't
properly understand. I too would be interested to know exactly what it
does break...

    >> i  don't  understand  why   you  cannot  following  the  normal
    >> configuration practices.

    derMouse> Principally, because  I have  no idea what  you consider
    derMouse> "normal  configuration practices".  I've  basically just
    derMouse> been mapping my  v4 knowledge over in what  appear to me
    derMouse> to be the  obvious ways, adjusting it as  I run into the
    derMouse> places where v4 and v6 differ.

Standard subnet  size is  /64. Wasteful of  bits? Maybe. It  does work
though. If you are trying to make sure your address is portable across
subnets, the  address doesn't  belong on a  physical subnet,  IMO. You
want to use  whatever is the local network on  your ethernet, and then
advertise  reachability  to an  address  that  lives  on the  loopback
interface -- the global, or  portable one. Except for the subnet size,
the same strategy works well with v4. I see no real difference.