IPv6 Question - EUI-64/Auto-configuration
danb83
Member Posts: 22 ■□□□□□□□□□
in CCNA & CCENT
I have just began looking at IPv6 for the first time and have some questions around EUI-64 and auto-configuration, which are both listed on the 200-120 exam topics.
My understanding of EUI-64 so far is that this allows a device to assign itself an Interface ID, by using its own MAC and padding it with an extra 16 bits (FFFE). The 7th bit is then inverted, for what I am not sure yet?
None of the materials I have really explain what stateless/auto-configuration is, without just repeating the whole EUI-64 details. Is Auto-configuration then just the process of a device using its MAC for its Interface ID? Or is there more to it?
My understanding of EUI-64 so far is that this allows a device to assign itself an Interface ID, by using its own MAC and padding it with an extra 16 bits (FFFE). The 7th bit is then inverted, for what I am not sure yet?
None of the materials I have really explain what stateless/auto-configuration is, without just repeating the whole EUI-64 details. Is Auto-configuration then just the process of a device using its MAC for its Interface ID? Or is there more to it?
Comments
-
Magic Johnson Member Posts: 414The 7th bit is then inverted, for what I am not sure yet?
This is something to do with local/universal bit.
http://www.techexams.net/forums/ccna-ccent/93286-universal-local-bit-eui-64-format-ipv6-address.html
EUI-64 is a method of SLAAC which just removes administration from IT department's point of view. If you use DHCP with SLAAC for clients the only thing they get from the DHCP server is DNS servers iirc? The IPv6 client can pretty much give itself a unique IPv6 address and it finds the DG on it's own too.
I went over this insanely before my CCENT exam and now 3 weeks later I can only remember snippets, I would verify that info but if you are using Odom's books he has 5 chapters (comprehensive) on it. -
EdTheLad Member Posts: 2,111 ■■■■□□□□□□Auto config can be stateful or stateless, stateful is in regards to a receiving an address via dhcp, stateless is as per what you mentioned above. Basically a router on the attached subnet will advertise its 64 bit prefix address, your hosts will then create the host portion of the address using it's 48 bit mac along with fffe and the 7th bit flipped. This will give you a 128 bit ipv6 address on the interface. You could also manually configure the address on the host.
So in summary, you have a router, you configure its interface with an ipv6 address. The router should send router advertisements, "show ipv6 interface x/x" will tell you if it is, the router advertisement contains the router prefix. You can then connect 100 hosts to the subnet and they will all learn the router prefix and compute a local user address using the e164 method, this saves manually configuring 100 host machines.Networking, sometimes i love it, mostly i hate it.Its all about the $$$$ -
networker050184 Mod Posts: 11,962 ModStateless auto configuration uses RA (Router Advertisements) to get the network portion and then the device creates the host portion using it's interface identifier.
Took to long to hit submit. Ed beat me with a more through answer!An expert is a man who has made all the mistakes which can be made.