Hi guys,

I have been given a /112 IPv6 subnet for internal use. I need to enable DHCP and when I create a scope it only allows me to create a /64 subnet.

There has to be a way to change this to give me a /112 subnet instead of the default /64.

Any ideas?

Hi,

Thanks for the post.

Please understand that the prefix length for an IPv6 subnet will always be /64; no more, no less. It allows you to place as many IPv6 devices as the underlying network medium allows. The lower-most 64 bits are called the interface ID. For Ethernet and some other interfaces, the MAC address is used as the source for interface ID.

Hope this helps.

Miles

Cool.

I see no scope options for DNS and gateways?

Unlike IPv4, IPv6 doesn't get its default gateway from DHCPv6 but from IPv6 router announcements and NDP (network discovery protocol). DHCPv6 will set your DNS server addresses just like DHCPv4 -- globally using the Server Options folder or by Scope using the Scope Options folder. The DNS server address option is listed as "DNS Recursive Name Server IPv6 Address List". Apparently the old "006 DNS Servers" was ... um... old... or something.

Actually, I'm in favor of any differentiation between v4 and v6, even when it's as arcane as the DNS server option name -- IPv6 is not an extension of IPv4, it's a completely new protocol, and anything that helps us poror network guys keep that straight as we learn all this is OK by me -- :)

Peace,

Robert

• Proposed as answer by OldRobP Monday, March 31, 2014 11:42 PM

A quick reply re: 64 bit subnet -- I spent several long days wrestling this one to the ground. The DHCPv6 scope isn't asking for a subnet, it's asking for a prefix. The full IPv6 prefix is 64 bits long and is made up of an address scope prefix, a router ID prefix and a 16-bit block for a subnet number. the remaining 64 bits are for the host address which is required to be a full 64 bits long. There are occasions where the scope or router prefixes are what's needed (6to4 address conversion, for example). The DHCPv6 server needs the full 64-bit prefix because it only manages 64-bit host (or node if you will) addresses. If you are subnetting, the subnet ID is the last 32 bits of the full 64-bit prefix, so you'll end up with scopes that are all identical except for the last block, i.e.,

Scope for subnet 4001 = 02ab:fe10:2aab 4001::/64

Scope for subnet 4001 = 02ab:fe10:2aab 4002::/64

Scope for subnet 4003 = 02ab:fe10:2aab 4003::/64

OK. Questions? I'll be more than glad to share anything I've learned getting this witchery working --

Peace,

Robert

   

• Proposed as answer by OldRobP Tuesday, April 01, 2014 12:10 AM

I realize this is old.. but if the answer is really the case then WHY does Microsoft have an option to enter the subnet prefix length?? If it has to be 64, then why give anyone the option of messing it up?