From: Scott Morris (swm@emanon.com)
Date: Tue Aug 09 2005 - 13:55:06 GMT-3
IMHO, you won't likely see them on the lab because they are on the way out.
While they are still supported as far as an address (unlike the IPv4
Compatible Addressing), I just don't see it being an important piece. Treat
them more like any other particular network where you have a starting point
(network) and subnet however they tell you.
Be aware that the IPv6 common usage guidelines are to use /64 masks all the
time. That way that "magical" eui-64 addressing works the way it's supposed
to!
HTH,
Scott
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
George Papadimitriou
Sent: Tuesday, August 09, 2005 12:49 PM
To: Scott Morris; ccielab@groupstudy.com
Subject: RE: ipv6 - subnet masks
Thanks,
I am more considered on what would be expected/implied in a scenario like
the one I described. I know that both /64 and longer subnet masks will work
with site-local ipv6 addresses. I am just not sure what would be expected to
configure in the real lab if subnet masks are not specified and the
addresses are ipv6 site-local.
In that case, given that the site-local addresses are by default /64 are we
restricted in our answer to use this subnet mask or not?
Thanks,
George
Scott Morris <swm@emanon.com> wrote:
According to the spec for Site Local addressing, they'll all work on /64
masks. There's nothing preventing you from using smaller masks like that,
but it's just not part of the spec!
If it makes you feel any better they're trying to get rid of Site Local
Addressing spec anyway! *shrug*
Scott
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
George Papadimitriou
Sent: Tuesday, August 09, 2005 11:03 AM
To: ccielab@groupstudy.com
Cc: george_papadimitriou@yahoo.com
Subject: ipv6 - subnet masks
Hi group,
A quick question, may seem obvious to many.
I'm a bit confused with what we do when we're asked to configure site-local
ipv6 addresses but the subnet mask is not specified. According to:
http://www.cisco.com/univercd/cc/td/doc/product/software/ios123/123cgcr/ipv6
_c/sa_bconn.htm#wp1027186
the 16-bit SLA field is the subnet field.
Let's say a site-local address has to be configured on an interface and the
subnet field in the ipv6 address must be the same (hex equivalent) as the
subnet field of the interface's ipv4 address. In that case, on an interface
with 172.16.15.1/24 do we have to configure FEC0::F:0:0:0:1/64 or are we
free to configure a longer subnet mask, e.g. FEC0::F:1/125 ?
In NMC scenario 1, the SLA and subnet fields are explicitly given and
site-local ipv6 subnets are configured "within" SLAs, e.g.
FEC0:0:0:A::7B:2/125, where A is the SLA and 7B is the ipv6 subnet ID.
My understanding is:
- If the SLA is explicitly given then it has to occupy the last 16-bits of
the first half of the ipv6 address, but further subnetting is possible with
longes subnet masks and subnet IDs
- If the SLA is not given and a site-local address is required with a
particular subnet ID then that subnet ID will have to go into the SLA field.
What are your views on this?
Thanks,
George
---------------------------------
Start your day with Yahoo! - make it your home page
This archive was generated by hypermail 2.1.4 : Sun Sep 04 2005 - 17:01:18 GMT-3