Cisco CCNA ICND2 640-816

IPv6: Understanding Basic Concepts and Addressing

by Jeremy Cioara

Start your 7-day free trial today.

This video is only available to subscribers.

A free trial includes:

  • Unlimited 24/7 access to our entire IT training video library.
  • Ability to train on the go with our mobile website and iOS/Android apps.
  • Note-taking, bookmarking, speed control, and closed captioning features.
Video Title Duration

Review: Rebuilding the Small Office Network, Part 1

Review: Rebuilding the Small Office Network, Part 2

Review: Rebuilding the Small Office Network, Part 3

Switch VLANs: Understanding VLANs

Switch VLANs: Understanding Trunks and VTP

Switch VLANs: Configuring VLANs and VTP, Part 1

Switch VLANs: Configuring VLANs and VTP, Part 2

Switch STP: Understanding the Spanning-Tree Protocol

Switch STP: Configuring Basic STP

Switch STP: Enhancements to STP

General Switching: Troubleshooting and Security Best Practices

Subnetting: Understanding VLSM

Routing Protocols: Distance Vector vs. Link State

Routing Protocols: OSPF Concepts

Routing Protocols: OSPF Configuration and Troubleshooting

Routing Protocols: EIGRP Concepts and Configuration

Access-Lists: The Rules of the ACL

Access-Lists: Configuring ACLs

Access-Lists: Configuring ACLs, Part 2

NAT: Understanding the Three Styles of NAT

NAT: Command-line NAT Configuration

WAN Connections: Concepts of VPN Technology

WAN Connections: Implementing PPP Authentication

WAN Connections: Understanding Frame Relay

WAN Connections: Configuring Frame Relay

IPv6: Understanding Basic Concepts and Addressing

00:00:00 - It's the final topic of the CCNA series, and I'd like to say
00:00:04 - we saved the best for last because it's a cool one,
00:00:08 - TCP/IPv6.
00:00:10 - This marks a pretty monumental day because TCP/IPv6
00:00:14 - has actually been out for a long time, more than
00:00:18 - a decade. I remember teaching back in 1997-98
00:00:22 - timeframe. One of, one of the first areas
00:00:26 - I taught in was Microsoft and I was teaching Microsoft Internet
00:00:29 - Information Server 3, and I remember saying, I believe that
00:00:34 - the whole Internet will be TCP/IPv6 by the year 2003
00:00:37 - because I just, I pulled the data out of the
00:00:41 - year because I figured it's gotta be by then, right?
00:00:45 - And well 2003 came and went and my prediction went unheard until
00:00:49 - now, until 2008 timeframe and we're now looking at TCP/IPv6
00:00:54 - as a very viable alternative and it's starting to move into all
00:00:59 - CISCO curriculums. So what we're going to talk about as we get into
00:01:02 - here is understanding the basic concepts and addressing.
00:01:05 - We're not going to be able to go in depth into every, everything
00:01:08 - TCP/IPv6, but this will give you a very good idea
00:01:13 - of where this technology is going and what it's going to look
00:01:16 - like. First question I want to ask is will we ever need to upgrade
00:01:20 - to IPv6? Meaning we've survived this long in IPv4,
00:01:23 - is there really a need? We'll then look at the IPv6
00:01:27 - addressing format and what the new kinds of addressing
00:01:30 - look like. They're big. We'll then look at the headers and address
00:01:34 - types because there's many different kinds of IP addresses
00:01:37 - in IPv6, and we'll do some in-depth exploration
00:01:40 - into what these new addresses look like and even basics of
00:01:45 - how we can start subnetting.
00:01:47 - Do we really need IPv6? The answer is yes, but it's
00:01:53 - more of a hidden thing, kind of like spam,
00:01:57 - email spam. Email spam in recent years has reached all new
00:02:02 - levels in that there are literally hundreds of thousands of
00:02:05 - spam emails being sent out every single minute all around the
00:02:09 - globe. We know there's a problem but we've got spam filters
00:02:13 - that are good enough to detect most of the spam and filter it
00:02:16 - out of our corporate email accounts. So we just kind of are
00:02:19 - turning a blind eye, but we need to fix the spam problem. We
00:02:23 - need a solution. It's just as of right now we kind of have this
00:02:26 - band-aid in place of filtering that people are just like, okay, we'll just,
00:02:29 - we'll pretend the spam is not really there. In the same sense
00:02:33 - we've got IP addresses on the Internet, and we've had
00:02:37 - TCP/IPv4 for years that has been running NAT.
00:02:40 - We have NAT happening to where we can have hundreds of
00:02:45 - clients share the same public IP address on the Internet
00:02:48 - as they go out. So there's a problem, there's an IP address
00:02:52 - shortage but it's so hidden. It's kind of like the spam filter.
00:02:55 - It's like, well we'll just pretend that there's not really a shortage
00:02:59 - because NAT fixes most of that.
00:03:01 - The U.S. Have, yes, Virginia, there is an IP address shortage.
00:03:05 - It's kind of like the, the famous Santa Claus editorial, but the U.S.A.,
00:03:10 - the United States who invented the Internet is still sitting
00:03:12 - pretty in the sense that they have tons of public addresses
00:03:16 - left, a lot of them just sitting there idle and unused. It's
00:03:20 - the rest of the world where the problem is. Until very recently,
00:03:23 - how generous, right, the United States gave Asia and Africa
00:03:28 - a single class C IP address range for their entire country.
00:03:33 - Wow. And they were, it was expected that Asia and Africa
00:03:37 - would use NAT to, to have that one class C IP address kind of
00:03:40 - last the country. Now recently they have gotten more and that
00:03:44 - the IP addresses are spreading out, but at the same time there
00:03:48 - are extreme shortages in countries outside of the United States.
00:03:52 - So what I will say, I'll add this little bonus piece right here, the
00:03:56 - the other countries are actually far ahead of the United States
00:04:00 - in implementing IPv6 because they have a desperate
00:04:03 - need for it. The United States is kind of like, ah, yeah, we know there's
00:04:07 - a problem, but it hasn't really affected us so much yet so we're
00:04:10 - very slow and migrating. But
00:04:13 - the Department of Defense, the people who created the Internet
00:04:17 - have said the year 2008 will be a big year for them because that's
00:04:20 - when they'll move all their networks to TCP/IPv6.
00:04:24 - So it's happening. The United states is just a little slow.
00:04:28 - So current IP addresses are poorly allocated. Agencies
00:04:31 - needing class C back in the day got a whole class B, and they
00:04:35 - now have all these addresses that they're just sitting on and
00:04:38 - not really using especially college universities. College
00:04:42 - universities were the first adopters of the Internet. They
00:04:45 - have tons of IP addresses that, well I won't say all of them
00:04:48 - but many of them have just, I'll, I'll throw one of them out
00:04:52 - there, University of Utah. I was actually talking with somebody
00:04:55 - who worked on the ITs out there. They said they have four unused
00:04:58 - class B public IP addresses just, just because that's what
00:05:03 - they had. Nothing, nothing against University of Utah. They
00:05:06 - just applied for that back in the day, and now they have them
00:05:08 - and they're just not using them. So estimates on IPv4
00:05:12 - exhaustion is largely debated, meaning some people say by
00:05:15 - the year 2009 we'll be out. Frankly, I really doubt that just because
00:05:20 - we've lasted this long, and I think we should be able to last
00:05:23 - a little longer. Some estimates go as far as the year 2041
00:05:27 - one where we'll be out, but hopefully by then we've migrated over
00:05:30 - anyhow. New network devices are on the rise. NAT is currently
00:05:35 - seen as a prohibitor of progress rather than a good solution
00:05:41 - anymore. That NAT is what allows our current Internet
00:05:44 - version to survive even though we've, we'd have no where near
00:05:47 - the amount of IP addresses that we need. So I guess let me
00:05:53 - give you a picture of the future,
00:05:56 - in my humble opinion. I believe that in the future when
00:06:03 - IPv6 is everywhere every single thing on the face
00:06:09 - on this planet will have an IP address. The technology to
00:06:13 - do that has been there for quite a while. We have cars that
00:06:16 - have IP addresses. We have refrigerators and microwave oven,
00:06:20 - Maytag makes them, that have IP addresses. So the technician
00:06:23 - can run diagnostics on the refrigerator remotely without having
00:06:27 - to send somebody out to figure out what's wrong with it. We
00:06:31 - have watches that have IP addresses, cell phones that have IP addresses.
00:06:34 - We have the technology to give pets, animals, dogs, cats IP
00:06:40 - addresses. All of the pets that you buy at the pet store nowadays
00:06:43 - come chipped is what they'll, they'll call it which is a chip
00:06:47 - that allows you to scan the pet and see who that pet belongs to.
00:06:51 - Well it's not too hard to modify that chip to have IP address
00:06:55 - where you could go to
00:06:57 - and find out where your dog is, you know, using
00:07:01 - some GPS signaling and stuff. There was a test done in New York City
00:07:05 - where people were volunteering to have chips the size of a
00:07:09 - grain of rice implanted in them that is biometrically powered, meaning
00:07:13 - their body energy powers these chips, and it allows them to be
00:07:17 - tracked wherever they go. It's part of an experimental program
00:07:20 - that is supposed to help with kidnappings, meaning children
00:07:24 - that are kidnapped will be able to be found much quicker
00:07:27 - because well
00:07:30 - you see where I'm going. If your children are chipped, they will
00:07:33 - be able to be tracked. Now that's somewhere in the middle
00:07:37 - of my digression there. We went from facts of cars and, and microwave
00:07:42 - oven to my theories of pets and people. But my point in all of
00:07:47 - this is the technology is there, and when IPv6 comes
00:07:51 - out you'll have the infrastructure now able to support it. Likewise,
00:07:57 - in the future, future features we're gonna see IPsec
00:08:01 - everywhere, meaning IPv6, this next version, has
00:08:06 - IPsec built in. So all network communication on every network
00:08:10 - work can be encrypted. So you'll see a big security rise. You'll
00:08:16 - see mobility where we can actually move from network to network
00:08:19 - as we're moving that will help with cars and things like that, has
00:08:23 - better mobility functions, and IPv6 has a simpler header
00:08:27 - than IPv4 which will improve on the processing
00:08:30 - power of all the different routers that use it.
00:08:34 - So now let's talk about some of the addressing, IPv6
00:08:38 - addresses. When, when people first started hearing about
00:08:41 - it a lot of people including myself thought it was going to
00:08:43 - be something like this. You had IPv4 so it would be something like
00:08:47 - that for version 6.
00:08:49 - They didn't do that. IPv6 addresses look like that.
00:08:55 - They are eight octets. I'll go right to left, one, two, three, four,
00:09:00 - five, six, seven, eight. With four characters each, it is now converted
00:09:05 - to hexadecimal, so A through F are valid characters
00:09:09 - along with numbers zero through nine. The point is that they
00:09:13 - don't want to do this upgrade ever again. We want to make sure
00:09:17 - that we can pick in a protocol and stay with it forever. Now
00:09:20 - I'm sure when people were creating TCP/IPv4
00:09:24 - they thought this will last forever and we ran out of addresses.
00:09:28 - And here I am IPv6 saying this will last forever, and
00:09:32 - maybe some guy like me in 50 years will be saying, oh, no,
00:09:35 - we, we thought we'd have enough addresses. But look at this, when
00:09:38 - we moved from 32-bit to 128-bit
00:09:41 - addressing we moved up to that many addresses. I say that
00:09:47 - many because I don't know how to pronounce that number. Somewhere
00:09:51 - around here we have the millions, billions, trillions, and then
00:09:55 - you lose me. I think somebody told me once it was like a
00:09:59 - quintillion or quintrillion, I don't know.
00:10:03 - It's a lot of addresses. Somebody far geekier than me figured out
00:10:07 - that with that many addresses we can give every square inch
00:10:12 - of the planet earth approximately 3.6 million addresses
00:10:17 - per square inch. Every three square feet of the Milky Way
00:10:20 - galaxy can be given an IP address with this scheme. So
00:10:25 - to tell you that's a lot of addresses. We're not going to be running
00:10:29 - out any time, any time soon. So because the addresses are so
00:10:34 - long, they decided to make them more manageable. They divided
00:10:37 - them into eight groups. So instead of dots like we have
00:10:40 - 192.168, we have colons. Each group is four
00:10:45 - hexadecimal
00:10:47 - characters each. Now you can see these are quite long to write so
00:10:51 - they came up with rules to eliminate some zeros and make them
00:10:54 - easier to manage. First off, rule number one is in an IPv6
00:10:59 - address you can eliminate groups of consecutive zeros by
00:11:03 - using a double colon,
00:11:05 - but you can only use it once per address. So you can see
00:11:10 - I had three groups of zero and I was really able to shorten that
00:11:14 - by just representing those three groups by putting double colons.
00:11:17 - But you could only use that once in an address. You should never
00:11:21 - see an address that has two sets of double colons because otherwise
00:11:24 - you wouldn't know how many zeros went in each location.
00:11:28 - Rule number two to shorten addresses is that you can drop
00:11:32 - leading zeros. So things like 0050 become 50,
00:11:36 - 0AB4 becomes AB4. So that allows you to really
00:11:42 - shorten this down to make it more manageable. Now it's still pretty
00:11:45 - long when you compared it to an IPv4 address, but at
00:11:48 - least if you're writing it, you're not going to be writing addresses
00:11:50 - like that every single time.
00:11:53 - Along with the bigger address IPv6 also provides
00:11:58 - a simpler header. I mentioned this when I was talking about
00:12:01 - the rationale for moving. In ICND1 we talked about
00:12:05 - an IPv4 four header and all these different fields that are
00:12:08 - in there like time to live, protocol, the checksum, flags,
00:12:12 - all kinds of stuff in the header that makes the, the packet harder
00:12:16 - to process for every single router.
00:12:18 - Down here is an IPv6 header. It still has some flags
00:12:22 - in there like how many hops it can go. Oops, I missed it.
00:12:25 - This one right here takes the place of time to live and, and, you know,
00:12:29 - the next header field provides a field for expanding headers
00:12:33 - and so on. And I don't mean to get into all those. The point is that it is much
00:12:37 - simpler. It's a bigger header meaning lengthwise it actually
00:12:42 - adds more data to the packet because our addresses are so big,
00:12:46 - but at the same time it's simpler for a router to process because
00:12:49 - it doesn't have to look at as many fields.
00:12:53 - Now we'll get into the real meat of the differences between
00:12:56 - IPv4 and v6 as we look at how they communicate.
00:13:01 - First off, in IPv6 there are only three types
00:13:05 - of messages, a unicast, a multicast and an anycast.
00:13:12 - Notice that one is missing from the IPv4.
00:13:16 - Which one?
00:13:18 - Broadcast. It's gone. Good riddance. Broadcasts are now a
00:13:24 - thing of yesteryear. In IPv6 there is no such
00:13:28 - thing as broadcast. It has been replaced by multicast meaning
00:13:32 - one-to-many. By using multicast I can dictate exactly where messages
00:13:38 - are sent to just a certain group of computer, all the computers.
00:13:41 - Now I will tell you that using multicast you can actually
00:13:45 - accomplish the same goals as broadcast, and there are some
00:13:48 - multicast messages that you'll see in IPv6 that
00:13:51 - are very similar to a broadcast. But the good old broadcast
00:13:55 - is gone. We now have unicast which is one-to-one,
00:13:59 - multicast, one-to-many or, or a group of people,
00:14:03 - and then an anycast which is one-to-closest. Anycast is going
00:14:06 - to be pretty awesome because you can, with anycast, give multiple
00:14:10 - devices the same IP address. So let's say that we have two routers
00:14:15 - right here that connect to the Internet. One of them is
00:14:19 - in a, a branch office and, you know, connects over here and maybe
00:14:23 - one is a corporate office or, I just went bad somewhere
00:14:29 - in that. We've got two routers, and we'll say they're
00:14:32 - both connected. They're, they're redundant for each other in a corporate
00:14:35 - office. We have a router here that connects to, you know, one group
00:14:38 - of users and another connection over here that's another group
00:14:42 - of users. With an anycast IP address I could actually give both
00:14:46 - of these routers the same IP address. And when the users go
00:14:50 - out to surf the Internet they'll just use whatever IP address is
00:14:53 - closest to them. Now this example that I gave is, is kind of
00:14:57 - silly. Here's a better one. Let's say that you have
00:15:03 - the world.
00:15:06 - I know, an artist at his best. This is the world and this over
00:15:11 - in North America or, you know, here's Russia, this is Australia,
00:15:14 - you're catching the drift. So we've got the world, and let's
00:15:17 - take a, a worldwide Internet site like eBay.
00:15:22 - eBay has servers all over the world, and today they actually
00:15:26 - is some pretty complex systems to make sure that they load
00:15:29 - balance correctly and are redundant. So they might have a bunch
00:15:32 - of servers in the United States. And when I go to
00:15:36 - it directs me to those servers and I access the ones closest
00:15:39 - to me. When I am in Russia, if I go to
00:15:43 - there will be a DNS infrastructure set up that will somehow get
00:15:47 - me to the server closest to me. Now to set something like
00:15:50 - that up it's very complex, and you have to use load balancers
00:15:53 - caching servers, there's a lot to it. But within an anycast address you'll
00:15:57 - be able to, with IPv6, just give all those servers
00:16:01 - the same IP address no matter where they are in the world,
00:16:04 - and the routing protocols will automatically find the closest
00:16:08 - server to you any time you're trying to communicate with any
00:16:12 - eBay website. Pretty powerful.
00:16:15 - So those are the types of messages. Now in red you can see the
00:16:19 - types of addresses. This is going to be something we have to
00:16:22 - get used to as well. In IPv6 your device can have
00:16:27 - many IP addresses and often times will have many IP addresses
00:16:31 - that it uses to communicate. There are three different addresses
00:16:35 - that are defined right now. First off is a link local address.
00:16:40 - This is something that you use to communicate in your Layer 2
00:16:43 - domain. For example, if you have people that are plugged into
00:16:46 - the same switch they will use the link local address to communicate
00:16:51 - with each other. It's just one type of address that's used for local
00:16:54 - communication. The next step up is an interesting story, it's
00:17:01 - the unique or site-local address.
00:17:05 - Now the name has changed as the IPv6 protocol
00:17:09 - has evolved, and that address was originally eliminated because
00:17:13 - people were like, we don't need this. Let me explain what it
00:17:16 - is. We today in our organizations use private addressing.
00:17:22 - We use private addressing because we have a shortage of
00:17:25 - Internet addresses, and we don't want to pay to assign public
00:17:28 - addresses to all of our clients. So we're used to this idea
00:17:31 - of private address, private address, private address. Well in IPv6
00:17:34 - we have enough IP addresses to give every device
00:17:40 - an IP address in the world for years and years and years to
00:17:43 - come. We don't have any shortage anymore. So this whole concept
00:17:47 - of private address should go away meaning we don't need
00:17:51 - private addresses anymore. But we're so used to them. We
00:17:59 - have them in the unique and site scope. It's going to be your
00:18:04 - option whether or not you would like to use a unique local
00:18:08 - or site local scope in your organization, but those fill the
00:18:11 - role of "private addresses." You don't have to use
00:18:15 - them. It's just that people are so used to using them that,
00:18:21 - you know, you just can't change. For example, if, if somebody
00:18:26 - came up with a new way of brushing teeth where I could just walk
00:18:30 - and push a button and my teeth would automatically get brushed, I
00:18:33 - would still probably instinctively go to my cabinet and grab
00:18:37 - the toothbrush and start brushing my teeth every night at least for a couple
00:18:40 - years because I've just brushed my teeth my whole life, and the fact
00:18:44 - that I push this button doesn't make me feel like my teeth got
00:18:47 - brushed. That's, that's a really weird example, but that's kind
00:18:50 - of the example of this. For years and years to come we'll probably
00:18:53 - use unique and site-local scope addresses in IPv6
00:18:57 - because it's just what people are used to. They can't fathom.
00:19:01 - It's even difficult for me to fathom a network without private
00:19:04 - addresses, but I have a feeling in maybe five to 10 years
00:19:08 - after IPv6 is adopted we'll be talking about this
00:19:12 - as like, oh, yeah, we used to have these things called unique and
00:19:15 - site-local addresses.
00:19:18 - Those were for the people that were just so stuck in their ways, they
00:19:20 - could handle a global scope, you know. That's, that's probably the way people
00:19:24 - will talk about it, and they'll be describing people like
00:19:26 - you and I because we're so used to private addresses. But the
00:19:29 - global scope this is the Internet or what people are now
00:19:34 - calling the Internet 2. Global scope are public addresses
00:19:39 - or addresses that are alive on the Internet. Now the good news
00:19:43 - is that your organization will be able to have Internet addresses
00:19:47 - or global addresses for every device that is available within
00:19:51 - them. Let's look at those addresses in detail starting off with
00:19:55 - the link local address. That's the one that's auto-generated just
00:20:00 - like the PCs nowadays that can't find a DHCP server and they
00:20:04 - auto-generate that 169.254 address. But
00:20:07 - the difference is this is auto-generated regardless of DHCP
00:20:11 - server or not. Every device will have a link local address. Now
00:20:15 - this is where we get into a little of the technicalities of
00:20:18 - our IP addressing. The RFC has specified that these addresses
00:20:23 - will always begin with FE80. Now that is because
00:20:28 - the first 10 bits must be 1111111010.
00:20:33 - Now when I get into IPv6 addressing I needed
00:20:37 - a little hexadecimal review just because I'm so used to decimal and binary
00:20:41 - from IPv4. Moving into the hex world was a little
00:20:44 - tough. But the first thing you have to remember is that every
00:20:47 - single one of these digits are represented by four binary bits,
00:20:52 - one, two, three, four. Now the way it works is very similar
00:20:55 - to decimal. You have every character in the, in the address has
00:20:58 - four bits. All zeros is really zero. 0001 is really
00:21:03 - one. 0010 is two. We all know this. This is, this
00:21:08 - is basic binary, and you can keep going with that all the way up
00:21:11 - to just nine, you know, that would be 1001 equals
00:21:16 - nine. Now this is where the hexadecimal gets a little
00:21:19 - bit different because we also have up to 16 values. It's
00:21:23 - actually zero through 15. That's possible with four bits.
00:21:26 - So when we go to 1010 zero which would typically
00:21:31 - be 10, we then go to an A. 1011 ends up being B.
00:21:40 - We keep going, 10 or actually, hang on, 1100
00:21:46 - would end up being C and so on. You have 1101 is
00:21:51 - D. 1110 is E. And then finally, all 1's is how
00:21:56 - we end up with the F. So when you see that the first 10
00:21:59 - bits must be 1111
00:22:03 - for the first bits, it means that we're going to have F
00:22:05 - as the first one. You can see the E because that's the second
00:22:09 - set of four bits. Now you may be wondering well how did you get eight with
00:22:12 - only two bits? Well remember that the RFC specifies that the first
00:22:16 - 10 bits must be this. However, every single character in hex
00:22:22 - requires four bits to take place. So we have eight coming because
00:22:27 - it's 10 and not specified here, but the others are going
00:22:30 - to be zero to make up that character because these link local
00:22:34 - addresses always begin with these first 10 bits and are followed
00:22:38 - with 54 bits of zero. So you can assume that every single
00:22:42 - one of these characters is going to take four bits and if you
00:22:45 - ever see two bits, you only need two more to make a character. So that's
00:22:48 - where we get the eight from. And of course the zero bops in because
00:22:52 - every IPv6 address has four characters, four hexadecimal
00:23:00 - characters per octets separated by colon. So that's why we
00:23:03 - end up having that FE80. Technically speaking most of the
00:23:06 - time you'll see it written FE8 in the RFC standards,
00:23:10 - but zero is what you'll always have on your address because
00:23:13 - of the following 54 bits of zeros. So that's our first
00:23:16 - octet. All of these right here, first 64 bits is what we
00:23:20 - just talked about. Now the last 64 bit is where it gets
00:23:25 - a little bit weird. Let me clear this off here. The last 64
00:23:28 - bits is the 48-bit MAC address from the host, whatever
00:23:33 - host this is being generated on followed by or I should say squished
00:23:38 - between FFFE or that is squeezed in the middle.
00:23:42 - It's hard to say that because here's the idea. Let's say we've
00:23:46 - got host with this MAC address. Well that's 48 bits,
00:23:49 - but we need 64 to complete the 128-bit IP
00:23:52 - address. So what the last 64 bits will be is first four
00:23:57 - of the MAC address 0019, 0019,
00:24:02 - second D1, D1, and then for some reason the designers
00:24:07 - and the powers that be decided to squeeze FFFE
00:24:09 - right in the middle. So in every single link local address
00:24:13 - you'll always see the MAC address with this kind of sandwiched
00:24:16 - between the two ends and then you can see the rest of the
00:24:18 - MAC address, 22, 22, DCF3, DCF3. So
00:24:24 - that will end up comprising the last 64 bits of that
00:24:27 - IP address that is only used for link local communication.
00:24:31 - Now it doesn't have to be used, for instance, if you had other IP
00:24:35 - addresses that were being used to communicate outside. Those
00:24:38 - would be used but if you're speaking to somebody on the same
00:24:41 - link and you realize that you have a source address coming from
00:24:44 - that link then you can use your link local address to communicate
00:24:47 - with them.
00:24:49 - Now let's move into the second debated IP address type,
00:24:53 - the unique-local or site-local addresses. I think I mentioned this
00:24:57 - backwards before and corrected myself. Unique-local is the new name, RFC
00:25:02 - 4193. Site local is the old name that was an
00:25:05 - RFC 3513. So we're supposed to be calling
00:25:09 - these unique-local addresses which I actually like the old
00:25:13 - name better, site-local, because it really describes what
00:25:16 - they do. They are used within enterprise networks to identify the
00:25:20 - boundary of their networks. So
00:25:23 - you can expect that these addresses will be relatable to the
00:25:28 - private addresses of IPv4, for the 10 range 192.168,
00:25:33 - all that. This is kind of the same thing. Now as
00:25:36 - it is specified in the RFC, they will use the following format.
00:25:40 - You can see that the first seven bits must be all 1's for
00:25:44 - the first one, that's our F. You can see, you can see C
00:25:48 - as the second one and then 00. But the RFC specifies
00:25:52 - that only those seven bits must be that way. So that's why
00:25:57 - we have 11, so there's four bits there, five, six, seven.
00:26:02 - The RFC mentions that this last bit right here, the one with the L,
00:26:07 - is going to be up to you.
00:26:10 - But it's kind of funny because they say all locally assigned
00:26:13 - addresses, meaning assigned by you, you should set the L to 1. So
00:26:19 - our real first eight bits are typically going to be
00:26:23 - 11111101. Zero is currently reside, reserved
00:26:28 - for future use by setting the L to zero. So
00:26:33 - currently the site addresses will all begin with FD00::8
00:26:37 - because even though the RFC says only
00:26:41 - this first seven bits have to be set, they say, oh, yeah, by the
00:26:44 - way, that last little L bit on the end should be set to 1 which
00:26:48 - means it's locally assigned by an administrator thus making
00:26:51 - all enterprise addresses or all private, I guess you could
00:26:56 - say, IPv6 address starting with FD00/8.
00:27:00 - Now it seems funny when you look at something like
00:27:03 - that to be like, oh, no, you're locking it down. There is not enough
00:27:06 - private IP addresses. But remember, we've got seven full octets
00:27:11 - of four hexadecimals each that, I mean you're, that's a gazillion
00:27:17 - different Ip addresses that you could use within your organization
00:27:20 - and likewise that's why they split up these separate sections.
00:27:23 - You have 40 bits which represent the global ID. That's intended
00:27:27 - to be your company like everybody in your company will have
00:27:32 - the same 40 bits starting their IP address or I should
00:27:35 - say following the FD00. The next 16 bits
00:27:39 - here are going to be the subnet ID because you're going to have
00:27:42 - VLANs, you're going to have subnets within your company's
00:27:45 - WAN links and so on. So this will identify the specific subnets.
00:27:48 - And then finally the last 64 bits just as we saw
00:27:52 - with the link local address will be the interface ID, it
00:27:57 - will be spliced into a MAC address style format like the other
00:28:01 - one was or you could come up with your own interface ID depending
00:28:04 - on the host that you're working with. It could be a DCHP pool.
00:28:07 - There is the DHCPv6, whatever, whatever way
00:28:10 - you're assigning interface IDs that is what this one will
00:28:14 - be and that will result in a global unique at least within your
00:28:17 - your enterprise IPv6 address.
00:28:21 - So we've seen the unique-local addresses or site-local. We've
00:28:25 - seen the link local. Next up is the global, not so local addresses.
00:28:30 - These are going to be the new pool of IP addresses that
00:28:34 - will build the IPv6 Internet. As of right now the
00:28:38 - only thing the standard will say is that they have to have
00:28:41 - their first three bits, high level three bits set to 001.
00:28:44 - And what that comes out to be is 2000 or
00:28:49 - 2 something ::3. You can just see the, the first
00:28:53 - three bits are what make that number two show up, but everything
00:28:56 - else is fair game. Now you can see the global running prefix.
00:29:00 - This address is divided into three major sections. The global
00:29:03 - routing prefix is 48 bits or less. They could be, could
00:29:07 - be smaller, could be larger. The subnet ID is going to be comprised
00:29:11 - of whatever bits are left over after you have this global routing
00:29:14 - prefix. Here's the idea that the powers that be have decided
00:29:20 - on for the Internet addresses. We've got this pool starting
00:29:23 - with 2xxx. You know, anything after that is fair game.
00:29:27 - So we can go ahead and assign, you know, maybe 2000:0,
00:29:33 - you know, da da da da da, some 40, you know, up to 48 bits
00:29:36 - right here for this global routing. And that'll go to Asia and 2,
00:29:40 - you know, da da da da da down the line, that will go to Egypt. Where did that
00:29:44 - come from? Egypt. How about like Australia or Egypt. Why not?
00:29:48 - Or these will go to Iran. These will go to Canada, you know. You get
00:29:52 - the idea. They're going to be chopping these up into giganto
00:29:55 - blocks and assigning them to the nations which they already
00:29:58 - have done. Many assignments have already been made. You can look
00:30:01 - at the RFC and they'll show you, you know, who's got what
00:30:05 - blocks. All these different unions have been assigned different
00:30:07 - blocks. But the primary addresses expected to comprise the
00:30:12 - IPv6 Internet
00:30:14 - are these. The ones that are coming from 2001::/16
00:30:19 - subnet, that is the IA and A's block.
00:30:23 - That is their range that they've chosen to use and,
00:30:27 - and start assigning. So you're going to see ISPs everywhere start getting
00:30:31 - blocks of those. As a matter of fact you probably are able
00:30:33 - to go and apply for your own block right now. Who knows? Maybe
00:30:37 - you'll be one of the early comers to the IPv6 address
00:30:40 - space and like 20 years from now you'll be looking
00:30:43 - back saying, I'm glad I got that block. Who would have known we'd run out, you know.
00:30:47 - Who knows? But that's, that's how they splice up all of those
00:30:51 - different global addresses and those will be public on the Internet.
00:30:56 - There is the idea behind the new IPv6 addressing.
00:31:00 - Pretty different, huh? There's going to be a huge learning
00:31:03 - curve when we really started seeing this move all way down
00:31:06 - to the desktops because remember this not only affects us as
00:31:10 - CISCO people but also affects everybody dealing with operating
00:31:14 - systems. Microsoft is going to use this. Linux is going to use this.
00:31:17 - Apple computers are all going to use this. So everybody is going to have
00:31:21 - to learn at least that are in the IT field. They're going to have to learn how
00:31:25 - IPv6 addresses work. So hopefully, I answered
00:31:29 - the question, will we need to upgrade? Yes, we will and the upgrade
00:31:33 - has already begun. As a matter of fact, oh, I just thought of this.
00:31:37 - Great, great thing I want to show you. If you go to Google and search for
00:31:42 - a BGP Looking Glass, that will let you see the routing table
00:31:46 - of the Internet. And this first link right here is just kind of, if you
00:31:50 - go there, it gives you a massive list of all kinds of websites
00:31:54 - that lets you look at Internet routing tables. And
00:31:59 - see the one I'm looking for is actually in Hawaii. Where
00:32:04 - did they go?
00:32:06 - LavaNet. That's it. LavaNet Looking Glass. If
00:32:10 - you go to the LavaNet Looking Glass and say, I would like,
00:32:13 - they actually have, if you look right here, the LavaNet IPv6
00:32:16 - Looking Glass. This shows the current routing table of
00:32:21 - the Internet 2, the new Internet that is existing. It says, what
00:32:25 - would you like to see? You can actually filter it. I'll just hit
00:32:27 - submit. It will display the whole IP routing table.
00:32:32 - Oh, wait a sec. That's not what I want to see. That's a show version. Give me the prefix-list.
00:32:37 - That's what I want to see. Submit. This will be the, my goodness.
00:32:42 - Oh, show IP BGP. That's what I want. There we go.
00:32:46 - I'm going to hit submit. This is going to show me the, there we go, the whole
00:32:50 - Internet routing table right now that is currently running
00:32:53 - the Internet 2. If you look, all the prefixes start with
00:32:56 - 2001 just like I was saying. And look at this. I'm going to scroll. These
00:33:03 - are all Internet locations on the Internet 2. And you know,
00:33:07 - I'm scrolling but the scroll bar is getting smaller and smaller
00:33:11 - over here on the right. This is just how many
00:33:15 - networks are already existing on the Internet 2.
00:33:19 - Five we go. This, this is to show you that the Internet 2
00:33:23 - exists. It is currently being built primarily in areas outside
00:33:27 - the United States. Will we need to upgrade to IPv6? Yes.
00:33:31 - It's already happening. We then saw the IPv6 address
00:33:34 - format, the eight octets of hexadecimal, 128-bit address.
00:33:38 - The headers are simpler and we saw
00:33:41 - the three different address types that exists, link local, unique-local,
00:33:44 - and the global addresses.
00:33:48 - Finally, we did an in-depth exploration of each one of those
00:33:52 - new addresses and how they're going to be used in the future.
00:33:55 - I hope this has been informative for you, and I'd like to thank
00:33:58 - you for viewing.

IPv6: Configuring, Routing, and Interoperating

Certification: Some Last Words for Test Takers

Advanced TCP/IP: Working with Binary

Advanced TCP/IP: IP Subnetting, Part 1

Advanced TCP/IP: IP Subnetting, Part 2

Advanced TCP/IP: IP Subnetting, Part 3

Please help us improve by sharing your feedback on training courses and videos. For customer service questions, please contact our support team. The views expressed in comments reflect those of the author and not of CBT Nuggets. We reserve the right to remove comments that do not adhere to our community standards.

comments powered by Disqus

Course Features

Speed Control

Play videos at a faster or slower pace.


Pick up where you left off watching a video.


Jot down information to refer back to at a later time.

Closed Captions

Follow what the trainers are saying with ease.

Offline Training

Our mobile apps offer the ability to download videos and train anytime, anywhere offline.

Accountability Coaching

Develop and maintain a study plan with assistance from coaches.
Jeremy Cioara

Jeremy Cioara

CBT Nuggets Trainer

Cisco CCNA, CCDA, CCNA Security, CCNA Voice, CCNP, CCSP, CCVP, CCDP, CCIE R&S; Amazon Web Services CSA; Microsoft MCP, MCSE, Novell CNA, CNE; CompTIA A+, Network+, iNet+

Area Of Expertise:
Cisco network administration and development. Author or coauthor of numerous books, including: CCNA Voice 640-461 Official Cert Guide; CCNA Voice Official Exam Certification Guide (640-460 IIUC); CCENT Exam Prep (Exam 640-822); CCNA Exam Cram (Exam 640-802) 3rd Edition; and CCNA Voice 640-461 Official Cert Guide.

Stay Connected

Get the latest updates on the subjects you choose.

  © 2015 CBT Nuggets. All rights reserved. Licensing Agreement | Billing Agreement | Privacy Policy | RSS