WEBVTT

0:00:08.200000 --> 0:00:10.800000
 Now, as far as the actual
 addressing goes.

0:00:10.800000 --> 0:00:16.360000
 So for IPv6 addressing, okay, I think
 most people probably know this,

0:00:16.360000 --> 0:00:19.700000
 but it does create a little
 bit of a problem, okay?

0:00:19.700000 --> 0:00:23.040000
 We're using 816.

0:00:23.040000 --> 0:00:25.820000
 I wrote byte there.

0:00:25.820000 --> 0:00:27.200000
 That's not right.

0:00:27.200000 --> 0:00:27.980000
 Sorry about that.

0:00:27.980000 --> 0:00:29.000000
 That is a mistake.

0:00:29.000000 --> 0:00:32.160000
 That should be 16-bit fields.

0:00:32.160000 --> 0:00:34.800000
 That would make it one
 heck of a long address.

0:00:34.800000 --> 0:00:41.500000
 But yeah, it's 816-bit fields or 2
 bytes on each field, separated by a

0:00:41.500000 --> 0:00:48.500000
 colon. Okay, now the colon, you know,
 I always joke about this when I'm

0:00:48.500000 --> 0:00:50.420000
 teaching classes and stuff.

0:00:50.420000 --> 0:00:53.260000
 You know, look down
 at your keyboard.

0:00:53.260000 --> 0:00:57.120000
 Couldn't they have come up
 with some other character?

0:00:57.120000 --> 0:00:59.260000
 Really? I mean, we
 have a lot of them.

0:00:59.260000 --> 0:01:01.860000
 There's a lot of keys
 on the keyboard.

0:01:01.860000 --> 0:01:04.500000
 Couldn't we have picked something
 other than the colon?

0:01:04.500000 --> 0:01:07.360000
 And I'll tell you two problems
 I have with it.

0:01:07.360000 --> 0:01:12.220000
 One of them is, well, quite frankly,
 you have to hit shift every time.

0:01:12.220000 --> 0:01:15.240000
 So, I mean, first off,
 in convenience.

0:01:15.240000 --> 0:01:16.960000
 Now you can do keyboard
 maps and stuff.

0:01:16.960000 --> 0:01:17.940000
 There's a way to fix that.

0:01:17.940000 --> 0:01:22.320000
 But the second problem I have with
 it, I think I have outlined on the

0:01:22.320000 --> 0:01:25.020000
 screen here for you, okay?

0:01:25.020000 --> 0:01:29.440000
 Is if you just typed into your browser
 to go to http colon forward slash

0:01:29.440000 --> 0:01:34.600000
 forward slash two thousand one DB eight
 one hundred colon colon ten colon

0:01:34.600000 --> 0:01:39.420000
 eighty eighty. I would have to
 ask you, oh, that's great.

0:01:39.420000 --> 0:01:46.120000
 That's cute. Now, the 8080, is that
 part of the address or is that the

0:01:46.120000 --> 0:01:55.500000
 port number? Now, this is a little bit
 of a problem with, you know, teaching

0:01:55.500000 --> 0:01:57.160000
 and explanations.

0:01:57.160000 --> 0:02:01.260000
 I haven't gotten into the rules
 of shortening the addresses yet.

0:02:01.260000 --> 0:02:04.420000
 But I wanted to talk about the colon,
 you know, sort of one of those.

0:02:04.420000 --> 0:02:08.520000
 So hopefully everybody's at least semi
 familiar with the double colons.

0:02:08.520000 --> 0:02:09.620000
 If not, you will be in a minute.

0:02:09.620000 --> 0:02:14.080000
 But the problem with the double colons
 is it eliminates a whole field

0:02:14.080000 --> 0:02:16.880000
 or fields of zeroes.

0:02:16.880000 --> 0:02:19.680000
 That's great. How many fields?

0:02:19.680000 --> 0:02:21.600000
 See, you don't know.

0:02:21.600000 --> 0:02:26.680000
 Was that address supposed to be four
 fields and then the 8080 is a port

0:02:26.680000 --> 0:02:33.400000
 number or is that just a five field address
 and the colons are eliminating

0:02:33.400000 --> 0:02:38.600000
 three fields, not four
 fields of zeroes?

0:02:38.600000 --> 0:02:40.460000
 See, that's the question.

0:02:40.460000 --> 0:02:43.340000
 And it's not really a question
 that I can tell you.

0:02:43.340000 --> 0:02:48.440000
 That top address, the 8080,
 is part of the host address.

0:02:48.440000 --> 0:02:50.640000
 Wow, I'm assuming host.

0:02:50.640000 --> 0:02:54.640000
 I'm assuming they're following
 the RFCs and doing a slash 64.

0:02:54.640000 --> 0:02:56.720000
 But we'll get to that later.

0:02:56.720000 --> 0:03:00.920000
 The 8080, as it's listed right there,
 is part of the host address.

0:03:00.920000 --> 0:03:05.560000
 If that was supposed to be a port number,
 then you need to look at the

0:03:05.560000 --> 0:03:07.480000
 address right below it.

0:03:07.480000 --> 0:03:10.500000
 That's using it as a port number.

0:03:10.500000 --> 0:03:15.460000
 That format is referred to
 as a literal IPv6 address.

0:03:15.460000 --> 0:03:18.720000
 If you ever want to look that up and
 give it a little more detail on it.

0:03:18.720000 --> 0:03:27.140000
 But that's using the brackets around
 it to contain the IPv6 address so

0:03:27.140000 --> 0:03:34.060000
 it becomes really clear what part is
 IPv6 address and which part is port

0:03:34.060000 --> 0:03:38.080000
 number. It's actually really funny,
 by the way, if you read up on some

0:03:38.080000 --> 0:03:39.860000
 of the documentation and such.

0:03:39.860000 --> 0:03:42.420000
 And I've seen it a couple of
 various different places.

0:03:42.420000 --> 0:03:47.700000
 People make comments like, well, this
 isn't new when I'm referring to

0:03:47.700000 --> 0:03:52.180000
 this. I mean, the use of the brackets
 to contain the address.

0:03:52.180000 --> 0:03:55.680000
 It's part of standard URL formatting
 is what it comes down to.

0:03:55.680000 --> 0:03:59.800000
 This isn't new. You can do
 this with IPv4 addresses.

0:03:59.800000 --> 0:04:02.220000
 And when I first read that, I'm looking
 at it going, you know, that's

0:04:02.220000 --> 0:04:09.820000
 interesting. I never actually knew that
 you could do that with IPv4 addresses,

0:04:09.820000 --> 0:04:17.600000
 you know, because I never had to
 because there was no confusion.

0:04:17.600000 --> 0:04:19.060000
 You know what I'm saying?

0:04:19.060000 --> 0:04:24.000000
 So, you know, it's fun and all, but
 the fact of the matter is it does

0:04:24.000000 --> 0:04:25.300000
 create confusion.

0:04:25.300000 --> 0:04:27.640000
 Plus you got to hit the shift key.


0:04:27.640000 --> 0:04:29.780000
 I don't know. I just
 sit there and go.

0:04:29.780000 --> 0:04:33.640000
 Couldn't you have used any
 other key on the keyboard?

0:04:33.640000 --> 0:04:35.600000
 Right? Not to mention the fact.

0:04:35.600000 --> 0:04:37.720000
 I mean, I'm picking on the
 right side of this address.

0:04:37.720000 --> 0:04:41.000000
 There's a colon on the left, too.

0:04:41.000000 --> 0:04:45.840000
 HTTP colon. I mean, how many colons
 can we have in the same line that

0:04:45.840000 --> 0:04:47.600000
 mean three different things?

0:04:47.600000 --> 0:04:49.840000
 You know, it's just sort of funny.


0:04:49.840000 --> 0:04:51.160000
 But it is what it is.

0:04:51.160000 --> 0:04:52.160000
 We have to live with it.

0:04:52.160000 --> 0:04:54.980000
 So I like to make fun of things.

0:04:54.980000 --> 0:04:56.820000
 Nonetheless, it is.

0:04:56.820000 --> 0:04:58.300000
 It's there. It's not changing.

0:04:58.300000 --> 0:04:59.340000
 So let's move on.

0:04:59.340000 --> 0:05:04.820000
 Okay? So as you've gathered by now, of
 course, 8 by 16 gives us 128 bits.

0:05:04.820000 --> 0:05:07.000000
 Okay? So this is 2 to 128.

0:05:07.000000 --> 0:05:09.940000
 There's the math.

0:05:09.940000 --> 0:05:15.000000
 It's 340. I'm not 100% sure
 even how to say that.

0:05:15.000000 --> 0:05:17.920000
 Undicellian, something or other.

0:05:17.920000 --> 0:05:20.980000
 It's a lot. How about that?

0:05:20.980000 --> 0:05:26.460000
 Okay. And again, I will not be the person
 to sit around and say, you know,

0:05:26.460000 --> 0:05:31.520000
 it'll be. More addresses than we'll
 ever need in our entire lifetime.

0:05:31.520000 --> 0:05:35.480000
 Again, I have no idea what's
 happening tomorrow.

0:05:35.480000 --> 0:05:37.540000
 So I will not be the
 one to say that.

0:05:37.540000 --> 0:05:41.940000
 Does it certainly appear to be more addresses
 than we could possibly ever

0:05:41.940000 --> 0:05:44.720000
 need? Yes, it does.

0:05:44.720000 --> 0:05:47.160000
 But again, we don't know
 what's going to happen.

0:05:47.160000 --> 0:05:53.020000
 So for things the way they are right
 now, and that would be, you know,

0:05:53.020000 --> 0:05:57.880000
 every man, woman and child, alive on
 planet Earth, and all of that stuff,

0:05:57.880000 --> 0:06:02.480000
 they're estimating it's probably enough
 address base to address the entire

0:06:02.480000 --> 0:06:04.820000
 solar system. Okay?

0:06:04.820000 --> 0:06:07.480000
 So I think we're probably okay.

0:06:07.480000 --> 0:06:11.000000
 But you know, some of the jokes I always
 use, you know, we never know

0:06:11.000000 --> 0:06:16.220000
 when, you know, they come out with,
 you know, nanomite technology next

0:06:16.220000 --> 0:06:21.540000
 week, and we all get injected with
 150 billion nanomites that heal our

0:06:21.540000 --> 0:06:22.660000
 bodies from the inside.

0:06:22.660000 --> 0:06:27.260000
 And they're all networked, and they're
 all, they all need an IPV6 address,

0:06:27.260000 --> 0:06:30.840000
 you know, and all of a sudden every
 person walking on the planet needs

0:06:30.840000 --> 0:06:34.660000
 100 billion addresses just for
 them, you know, you never know.

0:06:34.660000 --> 0:06:39.180000
 I'm just saying, I won't be the one to
 say it's more addressing than we'll

0:06:39.180000 --> 0:06:42.320000
 ever need. I won't
 say it, will not.

0:06:42.320000 --> 0:06:48.240000
 But it's a lot. And the main reason
 I'm bringing this up too is as we

0:06:48.240000 --> 0:06:54.380000
 start getting into things and you're
 going to find that they recommend

0:06:54.380000 --> 0:07:02.860000
 using slash 64s for client networks,
 end devices, host subnets, things

0:07:02.860000 --> 0:07:07.900000
 like that. And for people who have
 worked with IPV4, you know, we look

0:07:07.900000 --> 0:07:12.200000
 at that, and I know, you know, I did
 as well, you know, until you really

0:07:12.200000 --> 0:07:15.320000
 start looking at the big picture, you
 look at that and go, yeah, okay,

0:07:15.320000 --> 0:07:19.320000
 I hear that it's a ton of address base,
 but then I look over here and

0:07:19.320000 --> 0:07:24.400000
 I see you telling me to use a slash
 64 for every client facing subnet.

0:07:24.400000 --> 0:07:30.940000
 I mean, that's crazy, you know, that's
 exponentially double the size of

0:07:30.940000 --> 0:07:32.420000
 our current address space, right?

0:07:32.420000 --> 0:07:36.400000
 I mean, the entire IPV4 address space
 is only 32 bits, and you're telling

0:07:36.400000 --> 0:07:40.760000
 me to put, and mathematically, of course,
 it's way, way, way more than

0:07:40.760000 --> 0:07:47.020000
 double. It's exponentially double the
 bits just on one segment, you know,

0:07:47.020000 --> 0:07:50.360000
 and a lot of people who have been doing
 IPV4 a long time have a big problem

0:07:50.360000 --> 0:07:55.640000
 with that. And possibly the best way
 to explain it is, yes, that's true.

0:07:55.640000 --> 0:08:01.440000
 That is true. You are using way, way,
 way, way, super way more addresses

0:08:01.440000 --> 0:08:08.000000
 on a single subnet than you could possibly
 ever want to need by billions

0:08:08.000000 --> 0:08:10.040000
 and billions and billions.

0:08:10.040000 --> 0:08:13.640000
 But you know, a lot of people look
 at that, well, they're just, yeah,

0:08:13.640000 --> 0:08:16.820000
 they have a lot of space, but
 they're just thrown it away.

0:08:16.820000 --> 0:08:21.780000
 Yeah, but that's like taking a bucket
 of water out of the ocean, throwing

0:08:21.780000 --> 0:08:25.960000
 it away and saying, oh, that's it,
 you just wasted a bucket of water.

0:08:25.960000 --> 0:08:31.620000
 I think the ocean's going to be okay
 without that bucket of water.

0:08:31.620000 --> 0:08:35.280000
 And I bet I could give everybody a bucket
 of water out of the ocean, and

0:08:35.280000 --> 0:08:37.260000
 we all still have plenty.

0:08:37.260000 --> 0:08:42.360000
 So if you see the point I'm trying
 to make here, you know, with IPV4,

0:08:42.360000 --> 0:08:46.460000
 I was giving you, you know, a little,
 maybe a little shot glass or something

0:08:46.460000 --> 0:08:47.600000
 out of the ocean.

0:08:47.600000 --> 0:08:49.220000
 Now I'm giving you a bucket.

0:08:49.220000 --> 0:08:52.860000
 Maybe you want to open it up to
 a 55 gallon drum or something.

0:08:52.860000 --> 0:08:54.060000
 I don't know, whatever.

0:08:54.060000 --> 0:08:59.960000
 The point is, there's still, even with
 giving hosts slash 64s, there's

0:08:59.960000 --> 0:09:05.300000
 still such a sliver that that is
 out of the overall address space.

0:09:05.300000 --> 0:09:07.740000
 It's not even, it's
 not even funny.

0:09:07.740000 --> 0:09:11.840000
 So I know that can be tough for people
 just coming out of the IPV4 world

0:09:11.840000 --> 0:09:16.440000
 to get their head around because believe
 me, you know, I'm the same way.

0:09:16.440000 --> 0:09:19.940000
 The first glance was, oh yeah, so
 you got all this address space.

0:09:19.940000 --> 0:09:21.620000
 Why don't you just throw
 it around then?

0:09:21.620000 --> 0:09:22.760000
 You know, let's just waste it.

0:09:22.760000 --> 0:09:27.820000
 Why not? You know, sort of, you know,
 a valid point on one level.

0:09:27.820000 --> 0:09:31.300000
 But okay. Now, just some quick things
 about shortening these addresses

0:09:31.300000 --> 0:09:36.060000
 because at 128 bits, they're obviously
 going to get incredibly long.

0:09:36.060000 --> 0:09:40.940000
 So the first rule is any zeros
 out of a field can be dropped.

0:09:40.940000 --> 0:09:47.020000
 So if you look at our first example
 there, 2001 colon 0 dB8 colon 0 1

0:09:47.020000 --> 0:09:52.440000
 0 0 colon colon 1 can very easily
 just drop the leading zeros.

0:09:52.440000 --> 0:09:55.280000
 And I've already given you some examples
 like this already, right?

0:09:55.280000 --> 0:09:58.220000
 But 2001 colon dB8 colon 100.

0:09:58.220000 --> 0:10:03.400000
 The basic idea, look at it almost
 like your paycheck, right?

0:10:03.400000 --> 0:10:07.420000
 If they dropped a bunch of leading zeros
 out of your paycheck, you wouldn't

0:10:07.420000 --> 0:10:14.020000
 care. If they dropped trailing zeros,
 that would make a big difference.

0:10:14.020000 --> 0:10:16.900000
 So you know, that's how you sort
 of have to look at this too.

0:10:16.900000 --> 0:10:19.200000
 Okay, leading zeros, no problem.

0:10:19.200000 --> 0:10:22.200000
 You cannot drop trailing
 zeros, right?

0:10:22.200000 --> 0:10:27.620000
 That 100 field there, you can't
 make that one colon colon.

0:10:27.620000 --> 0:10:31.980000
 That's, you know, one is not the same
 as 100 is what it comes down to.

0:10:31.980000 --> 0:10:35.220000
 Okay. But 0 100 is
 the same as 100.

0:10:35.220000 --> 0:10:36.780000
 So it's a pretty easy one.

0:10:36.780000 --> 0:10:42.580000
 Okay. The other thing is the
 consecutive fields of zeros.

0:10:42.580000 --> 0:10:46.820000
 Okay. Now again, I addressed this a
 little bit a few minutes ago, but

0:10:46.820000 --> 0:10:49.860000
 I just want to make sure a
 couple things are clear.

0:10:49.860000 --> 0:10:54.100000
 I'm sure most of you are probably aware
 of this, but in case you're not,

0:10:54.100000 --> 0:11:01.340000
 the double colons can represent
 any number of fields of zeros.

0:11:01.340000 --> 0:11:06.300000
 The easiest way to explain this is all
 the devices will take everything

0:11:06.300000 --> 0:11:11.680000
 to the left of the double colons and move
 it to the beginning of the address.

0:11:11.680000 --> 0:11:15.520000
 Everything to the right of the double
 colons move it to the far right

0:11:15.520000 --> 0:11:20.420000
 of the address and anything
 in the middle becomes zeros.

0:11:20.420000 --> 0:11:23.280000
 That's the easiest
 way to explain it.

0:11:23.280000 --> 0:11:28.180000
 Okay. So a lot of people look at it
 and go, so if I'm eliminating three

0:11:28.180000 --> 0:11:30.960000
 fields of zeros, do
 I use three colons?

0:11:30.960000 --> 0:11:33.400000
 Absolutely not. No,
 no, no, no, no.

0:11:33.400000 --> 0:11:39.480000
 It's always two colons together and
 everything in the middle pads, you

0:11:39.480000 --> 0:11:41.360000
 know what? I don't care.

0:11:41.360000 --> 0:11:44.080000
 One field, two fields,
 three fields.

0:11:44.080000 --> 0:11:50.760000
 In the top example there, we're eliminating
 five fields of zeros.

0:11:50.760000 --> 0:11:56.420000
 If you look at the second example with
 the default, just colon colon,

0:11:56.420000 --> 0:11:57.420000
 that's all zeros.

0:11:57.420000 --> 0:12:00.160000
 That's eight fields of zeros.

0:12:00.160000 --> 0:12:01.840000
 Okay, loop back.

0:12:01.840000 --> 0:12:07.280000
 Yes, they did not blow a whole class
 of addresses on a loop back address

0:12:07.280000 --> 0:12:11.920000
 this time. So with IPv6, they did learn
 their lesson, even though we have

0:12:11.920000 --> 0:12:13.780000
 tons of address space.

0:12:13.780000 --> 0:12:19.900000
 They did not waste a couple million
 like they did in IPv4 with reserving

0:12:19.900000 --> 0:12:27.540000
 the whole 127 address range
 for one address, 127.001.

0:12:27.540000 --> 0:12:30.500000
 It's real easy to look back and
 say, oh, it was really stupid.

0:12:30.500000 --> 0:12:31.640000
 Why did they do that?

0:12:31.640000 --> 0:12:34.660000
 It's easy to look
 back and say that.

0:12:34.660000 --> 0:12:39.420000
 At the time, I'm sure that they looked
 at it and said, you got to remember

0:12:39.420000 --> 0:12:44.740000
 IPv4 was developed to be an internal
 defense network originally, then

0:12:44.740000 --> 0:12:48.140000
 it expanded to colleges
 and stuff like that.

0:12:48.140000 --> 0:12:49.700000
 Nobody saw the internet coming.

0:12:49.700000 --> 0:12:55.120000
 I mean, obviously, some few people did,
 at least had an idea maybe, but

0:12:55.120000 --> 0:13:01.560000
 the point is, nobody was designing IPv4
 going, you better not waste those

0:13:01.560000 --> 0:13:02.700000
 millions of addresses.

0:13:02.700000 --> 0:13:03.880000
 We're going to need those.

0:13:03.880000 --> 0:13:08.700000
 No, no. They were looking at it going,
 there's billions of addresses usable.

0:13:08.700000 --> 0:13:12.360000
 That's more than we could possibly
 ever need, by the way.

0:13:12.360000 --> 0:13:15.700000
 See why I'm not saying
 that about IPv6?

0:13:15.700000 --> 0:13:17.320000
 That's exactly my point.

0:13:17.320000 --> 0:13:21.700000
 I'm sure they felt that way about
 the first loopback, 127.001.

0:13:21.700000 --> 0:13:26.020000
 Who cares? Just take the highest value
 in a class A address, make it a

0:13:26.020000 --> 0:13:28.400000
 loopback. What difference
 does it make?

0:13:28.400000 --> 0:13:31.300000
 I'm sure that's how it happened.

0:13:31.300000 --> 0:13:34.980000
 But now looking back,
 we go, hmm, bad idea.

0:13:34.980000 --> 0:13:38.760000
 Let's create a loopback and we'll
 just make it colon colon one.

0:13:38.760000 --> 0:13:41.180000
 And of course, we have
 another example there.

0:13:41.180000 --> 0:13:44.160000
 I'm showing you what
 you cannot do.

0:13:44.160000 --> 0:13:50.280000
 So the last one, of course, the
 one that I put first is correct.

0:13:50.280000 --> 0:13:53.740000
 You could eliminate, and again, if you
 look at this, you could theoretically

0:13:53.740000 --> 0:14:00.100000
 eliminate either the zeros before
 the 37 or the zeros after the 37.

0:14:00.100000 --> 0:14:02.680000
 Either one of those
 is actually correct.

0:14:02.680000 --> 0:14:05.900000
 What you cannot do, so in other words,
 just make sure we're clear, I could

0:14:05.900000 --> 0:14:12.200000
 have done 2001 colon zero colon zero
 colon zero colon 37 colon colon one.

0:14:12.200000 --> 0:14:13.680000
 And that would have been okay.

0:14:13.680000 --> 0:14:15.680000
 Nothing illegal about that.

0:14:15.680000 --> 0:14:17.760000
 It's not the most efficient.

0:14:17.760000 --> 0:14:22.060000
 Usually you would want to eliminate
 the more fields of zeros rather than

0:14:22.060000 --> 0:14:26.080000
 the fewer fields of zeros, but
 neither one is incorrect.

0:14:26.080000 --> 0:14:30.940000
 What you cannot do, notice the all caps,
 you cannot do the double colons

0:14:30.940000 --> 0:14:35.700000
 twice. That introduces ambiguity.

0:14:35.700000 --> 0:14:41.620000
 How many fields of zeros go before the
 37 versus how many fields go after?

0:14:41.620000 --> 0:14:44.880000
 The router would look
 at that and go okay.

0:14:44.880000 --> 0:14:48.720000
 So you've eliminated
 five fields of zeros.

0:14:48.720000 --> 0:14:54.700000
 Great. Is that one in front, four after,
 two in front, three after, four

0:14:54.700000 --> 0:14:58.700000
 in front, one after, you know, there's
 at least, and that's just off the

0:14:58.700000 --> 0:15:03.280000
 top of my head, there's at least three
 combinations or whatever of how

0:15:03.280000 --> 0:15:09.360000
 those could be split up and you can't
 do that with an address, right?

0:15:09.360000 --> 0:15:11.800000
 Try that with the post
 office, right?

0:15:11.800000 --> 0:15:13.660000
 Just put three different
 addresses on the letter.

0:15:13.660000 --> 0:15:17.300000
 Well it could be this street address
 or maybe this one or maybe this one.

0:15:17.300000 --> 0:15:19.360000
 Yeah, that's not going to work.

0:15:19.360000 --> 0:15:22.280000
 So yeah, there's no
 way you can do that.

0:15:22.280000 --> 0:15:26.480000
 So that would just be an illegal
 address, cannot be done.

0:15:26.480000 --> 0:15:29.260000
 Just want to make sure we're
 all very clear on that.

0:15:29.260000 --> 0:15:33.780000
 The other thing to consider with IPv6
 addressing are the different address

0:15:33.780000 --> 0:15:40.840000
 types. Okay, first we have the address
 types themselves and for those

0:15:40.840000 --> 0:15:46.800000
 of you coming from IPv4, this could
 be a big one to get used to which

0:15:46.800000 --> 0:15:49.540000
 is link local addresses.

0:15:49.540000 --> 0:15:56.120000
 Okay, these are taken from the range
 FE80 colon colon slash 10.

0:15:56.120000 --> 0:15:59.900000
 Now please note that
 it's a slash 10.

0:15:59.900000 --> 0:16:09.760000
 So really that means FE8, FE9, FEA and
 FEB are all part of that because

0:16:09.760000 --> 0:16:15.620000
 the 10, the F and the E represent eight
 bits and that leaves, means there's

0:16:15.620000 --> 0:16:18.260000
 two bits left coming out
 of the eight field.

0:16:18.260000 --> 0:16:25.400000
 Well, if you take eight, it's of
 course 1, 0, 0, 0 in those bits.

0:16:25.400000 --> 0:16:28.980000
 But of course in this case it would
 just be 1, 0 which means that it can

0:16:28.980000 --> 0:16:35.240000
 be either like I said, 8, 9, A or B are
 all still technically link local.

0:16:35.240000 --> 0:16:37.180000
 I've never seen anybody use them.

0:16:37.180000 --> 0:16:41.860000
 Every link local I've ever seen has
 been FE80 but I just want to make

0:16:41.860000 --> 0:16:49.540000
 sure we're clear that that's not an
 FE80 slash 16, it's a slash 10.

0:16:49.540000 --> 0:16:54.640000
 So just being a little technical
 there but just be aware of it.

0:16:54.640000 --> 0:16:58.800000
 Okay, now the more important thing is
 what the heck's a link local address.

0:16:58.800000 --> 0:17:03.120000
 And this is where things get
 a little different with IPv6.

0:17:03.120000 --> 0:17:08.780000
 And again you can make the argument while
 we're wasting possibly you could

0:17:08.780000 --> 0:17:14.280000
 argue addresses handing out slash 64's
 to clients like we just discussed.

0:17:14.280000 --> 0:17:19.320000
 The flip side of this is this less that
 save tons of addresses elsewhere

0:17:19.320000 --> 0:17:25.560000
 in our network. The idea behind a link
 local address is it is a non-routable

0:17:25.560000 --> 0:17:32.780000
 address. This address, this communication
 can never ever leave the segment.

0:17:32.780000 --> 0:17:38.480000
 So this is perfect for sending traffic
 say between two routers.

0:17:38.480000 --> 0:17:42.320000
 So you've got a WAN link going
 out to a remote site.

0:17:42.320000 --> 0:17:50.680000
 I don't care what it is, MetroE, MPLS
 over MetroE or over PVP, I don't

0:17:50.680000 --> 0:17:51.960000
 care what it is.

0:17:51.960000 --> 0:17:56.900000
 The fact of the matter is with IPv4 you
 really sort of only have a couple

0:17:56.900000 --> 0:17:59.780000
 of choices. I mean you could
 do IP outnumbered on it.

0:17:59.780000 --> 0:18:01.780000
 That gets a little tricky.

0:18:01.780000 --> 0:18:07.120000
 You could do, you know, everybody knows
 you've been to a slash 30 or a

0:18:07.120000 --> 0:18:13.060000
 slash 31. Great, but you're still
 doing a slash 30 or a slash 31.

0:18:13.060000 --> 0:18:16.140000
 You're putting addresses
 on that WAN link.

0:18:16.140000 --> 0:18:18.720000
 And at the end of the day you have
 to ask yourself a very fundamental

0:18:18.720000 --> 0:18:25.240000
 question. Why? And I'm not saying why
 on IPv4 because on IPv4 you have

0:18:25.240000 --> 0:18:27.920000
 to. So the answer to why
 is because I have to.

0:18:27.920000 --> 0:18:33.800000
 Of course. What I'm saying is in the
 big picture why does the WAN link

0:18:33.800000 --> 0:18:42.000000
 between a remote site and your headquarters
 need to be globally addressable?

0:18:42.000000 --> 0:18:46.740000
 You don't need to be able to route
 that WAN segment from your desktop

0:18:46.740000 --> 0:18:52.720000
 machine. See the point is a lot
 of the segments on your network.

0:18:52.720000 --> 0:18:56.020000
 Okay, this can be point to point
 links between switches.

0:18:56.020000 --> 0:18:57.320000
 I mean it doesn't matter.

0:18:57.320000 --> 0:18:59.620000
 I'm just picking on WAN
 links as an example.

0:18:59.620000 --> 0:19:04.020000
 Anywhere that there are not, I
 mean just just the generalize.

0:19:04.020000 --> 0:19:09.820000
 Anywhere there's not printers,
 servers, PCs, workstations.

0:19:09.820000 --> 0:19:13.280000
 Anywhere there's not
 an end device.

0:19:13.280000 --> 0:19:16.600000
 You don't need to be able
 to route to that segment.

0:19:16.600000 --> 0:19:19.120000
 Okay, we call those
 transit segments.

0:19:19.120000 --> 0:19:21.880000
 There's no need to be
 able to get to those.

0:19:21.880000 --> 0:19:25.380000
 So why do they have
 addresses then?

0:19:25.380000 --> 0:19:29.520000
 In IPv4 they have addresses because
 that's really our only choice.

0:19:29.520000 --> 0:19:33.480000
 But in IPv6 see they looked at it and
 they said if you don't need to route

0:19:33.480000 --> 0:19:39.140000
 to that segment, if all you literally
 need to do is pass IPv6 traffic

0:19:39.140000 --> 0:19:48.980000
 through the segment, not to the segment,
 then in no way do you need globally

0:19:48.980000 --> 0:19:52.560000
 routable IPv6 addresses.

0:19:52.560000 --> 0:19:55.440000
 And we'll see these quite
 a bit as we go.

0:19:55.440000 --> 0:19:57.960000
 But that's the idea
 behind link local.

0:19:57.960000 --> 0:20:03.840000
 Okay, you don't even have to have routable
 IPv6 addresses on transit links

0:20:03.840000 --> 0:20:05.440000
 between routers.

0:20:05.440000 --> 0:20:13.400000
 All of your routing protocols, RIP,
 OSPF, EIGRP, even BGP depending on

0:20:13.400000 --> 0:20:16.420000
 the conditions. Now BGP gets a little
 different like it does for so many

0:20:16.420000 --> 0:20:21.120000
 things, right? It's always our little
 on-ball routing protocol because

0:20:21.120000 --> 0:20:23.440000
 of what it does and what it's for.


0:20:23.440000 --> 0:20:25.640000
 It does things a little
 bit different.

0:20:25.640000 --> 0:20:32.040000
 Okay, but it can, but it can also maybe
 not use link local addresses for

0:20:32.040000 --> 0:20:33.280000
 its communication.

0:20:33.280000 --> 0:20:37.020000
 When we start looking at routing tables
 and routing protocols and stuff

0:20:37.020000 --> 0:20:40.900000
 later in the course, we're going to
 discover that all of those things

0:20:40.900000 --> 0:20:43.120000
 use the link local addresses.

0:20:43.120000 --> 0:20:48.260000
 All your next tops are going to be
 linked locals until we get to BGP.

0:20:48.260000 --> 0:20:54.320000
 But they're used a ton and we really
 sort of didn't have an equivalent

0:20:54.320000 --> 0:20:59.100000
 to that in IPv4.

0:20:59.100000 --> 0:21:03.900000
 Okay, now of course there's global, okay,
 which are of course the globally

0:21:03.900000 --> 0:21:09.700000
 routable, managed by IANA, all that
 kind of stuff, just like your public

0:21:09.700000 --> 0:21:12.220000
 IPv4 addresses today.

0:21:12.220000 --> 0:21:14.960000
 Okay, now there's also
 another one here.

0:21:14.960000 --> 0:21:18.840000
 I almost hesitated even
 putting it in here.

0:21:18.840000 --> 0:21:20.880000
 It should be addressed.

0:21:20.880000 --> 0:21:27.720000
 I'm not sure why people would be using
 it, but there's always some business

0:21:27.720000 --> 0:21:30.280000
 case for something
 somewhere, right?

0:21:30.280000 --> 0:21:38.560000
 So private addressing, sort of like
 RFC 1918 addressing for IPv4.

0:21:38.560000 --> 0:21:45.580000
 Now, if anybody's been studying IPv6
 before, don't get this confused with

0:21:45.580000 --> 0:21:51.700000
 site local addressing, which
 was the FEC zero range.

0:21:51.700000 --> 0:21:55.780000
 Those have been officially deprecated,
 so you're not supposed to be using

0:21:55.780000 --> 0:22:00.060000
 those anymore. There were a
 lot of concerns with them.

0:22:00.060000 --> 0:22:03.460000
 It wasn't as clear as it should
 be and various things.

0:22:03.460000 --> 0:22:07.580000
 Unique local are the addresses that you're
 supposed to be using to replace

0:22:07.580000 --> 0:22:14.780000
 that. Now, if you read the RFCs, they
 basically state that this is for

0:22:14.780000 --> 0:22:20.800000
 use inside of a non-connected
 private network.

0:22:20.800000 --> 0:22:23.960000
 In other words, you're literally not connecting
 to the internet or anything

0:22:23.960000 --> 0:22:28.900000
 else. You just want some addressing
 to use inside for some reason.

0:22:28.900000 --> 0:22:32.660000
 A lot of people look at this and go, oh,
 so I'm supposed to use this inside

0:22:32.660000 --> 0:22:35.780000
 of NAT then? No.

0:22:35.780000 --> 0:22:37.920000
 NAT was a crutch.

0:22:37.920000 --> 0:22:43.420000
 NAT was to deal with the fact that we
 were running out of IPv4 addresses.

0:22:43.420000 --> 0:22:48.660000
 Okay? You're not supposed
 to be using NAT with IPv6.

0:22:48.660000 --> 0:22:52.280000
 And I hear people say it all the time,
 oh, we're going to keep doing NAT.

0:22:52.280000 --> 0:22:55.060000
 Okay. Let me address
 that real quick.

0:22:55.060000 --> 0:22:58.000000
 If you're going to keep doing NAT,
 then let me ask you why you're even

0:22:58.000000 --> 0:23:00.040000
 looking at IPv6 anyway.

0:23:00.040000 --> 0:23:01.900000
 We'll start with that.

0:23:01.900000 --> 0:23:05.700000
 Well, eventually they're going to
 make a code IPv6 for the internet.

0:23:05.700000 --> 0:23:10.960000
 Okay. Translate IPv4 inside of your network
 to IPv6 outside of your network.

0:23:10.960000 --> 0:23:16.740000
 Just do protocol translation
 and you're done.

0:23:16.740000 --> 0:23:18.480000
 The whole thing is this.

0:23:18.480000 --> 0:23:21.700000
 NAT has been great.

0:23:21.700000 --> 0:23:23.200000
 We all love it. We appreciate it.

0:23:23.200000 --> 0:23:27.560000
 It's wonderful. It's gotten us as far
 as we have gotten today without

0:23:27.560000 --> 0:23:31.360000
 having to make the big
 jump to IPv6 just yet.

0:23:31.360000 --> 0:23:36.200000
 But the fact of the matter is it does
 break the end-to-end connectivity

0:23:36.200000 --> 0:23:40.620000
 model. It breaks a lot of the stuff
 that we're trying to head towards

0:23:40.620000 --> 0:23:42.240000
 in this industry.

0:23:42.240000 --> 0:23:46.740000
 Okay? And at the end of the day,
 it doesn't buy you anything.

0:23:46.740000 --> 0:23:51.300000
 And you get a whole lot of, and I've
 heard a lot of pushback on this,

0:23:51.300000 --> 0:23:57.800000
 especially from security folks,
 oh, it hides my inside network.

0:23:57.800000 --> 0:24:00.900000
 To an extent, okay.

0:24:00.900000 --> 0:24:06.460000
 But on the other hand, if you're following
 RFC 1918, I can guess within

0:24:06.460000 --> 0:24:10.020000
 about three guesses what your internal
 addressing structure is.

0:24:10.020000 --> 0:24:15.860000
 If you're using some different public
 range inside of your network, then

0:24:15.860000 --> 0:24:18.200000
 you've got a whole nother level of
 NAT problems I'm not even going to

0:24:18.200000 --> 0:24:21.780000
 talk about right now.

0:24:21.780000 --> 0:24:25.820000
 And at the end of the day, I'm
 just going to be blunt here.

0:24:25.820000 --> 0:24:33.180000
 If your firewall cannot protect you from
 threats without NAT, then you've

0:24:33.180000 --> 0:24:36.820000
 got a major problem with your firewall
 and you need a new one.

0:24:36.820000 --> 0:24:39.220000
 I'm just going to throw
 it out there.

0:24:39.220000 --> 0:24:44.060000
 You should not be using NAT
 as a security mechanism.

0:24:44.060000 --> 0:24:49.260000
 You know, security through obscurity
 is never a good idea.

0:24:49.260000 --> 0:24:52.580000
 Okay? You can't get my stuff because
 you don't know what the addresses

0:24:52.580000 --> 0:24:56.800000
 are. Or you can't get my stuff because
 the traffic has to originate from

0:24:56.800000 --> 0:25:00.220000
 the inside for a NAT translation
 to be set up.

0:25:00.220000 --> 0:25:01.660000
 I've heard that one.

0:25:01.660000 --> 0:25:06.040000
 And my answer to that is
 stateful inspection.

0:25:06.040000 --> 0:25:12.000000
 Two words done. So, you know, all the
 arguments about, oh, we need to

0:25:12.000000 --> 0:25:13.840000
 keep NAT for security.

0:25:13.840000 --> 0:25:17.520000
 It's nonsense. It's just nonsense.


0:25:17.520000 --> 0:25:21.400000
 The fact of the matter is, like I said,
 your firewall needs to be able

0:25:21.400000 --> 0:25:23.300000
 to protect you either way.

0:25:23.300000 --> 0:25:27.120000
 Okay? They have provided
 this address space.

0:25:27.120000 --> 0:25:29.700000
 If you don't believe
 me, go read the RFCs.

0:25:29.700000 --> 0:25:33.020000
 They don't recommend doing it with
 NAT and all that kind of stuff.

0:25:33.020000 --> 0:25:36.260000
 So I would just not even go there.


0:25:36.260000 --> 0:25:39.580000
 Okay? That's why I said I almost considered
 not even mentioning them.

0:25:39.580000 --> 0:25:44.040000
 Okay? The other thing we need to talk
 about is the communication types.

0:25:44.040000 --> 0:25:49.840000
 Unicast versus multicast
 versus anycast.

0:25:49.840000 --> 0:25:59.160000
 Okay? Now, unicast, pretty simple,
 just like IPv4, one to one.

0:25:59.160000 --> 0:26:05.020000
 Okay? Multicast and anycast.

0:26:05.020000 --> 0:26:09.880000
 Okay. Multicast, same
 exact idea as IPv4.

0:26:09.880000 --> 0:26:14.600000
 So it's one machine sending a packet
 going to multiple destinations.

0:26:14.600000 --> 0:26:17.380000
 Okay? The addressing,
 of course, changes.

0:26:17.380000 --> 0:26:20.480000
 We'll talk a little bit
 more about this later.

0:26:20.480000 --> 0:26:25.260000
 But in IPv6, all multicast
 addresses start FF.

0:26:25.260000 --> 0:26:29.180000
 Okay? But beyond that,
 multicast is multicast.

0:26:29.180000 --> 0:26:33.440000
 Anycast. And I almost hate to put
 this under communication types.

0:26:33.440000 --> 0:26:37.760000
 This is a big pet peeve
 of mine personally.

0:26:37.760000 --> 0:26:43.860000
 But since this is how everybody lists
 it, all play along, all do it too.

0:26:43.860000 --> 0:26:47.500000
 And we'll say that anycast
 is a communications type.

0:26:47.500000 --> 0:26:55.340000
 Fine. Anycast is really nothing more
 than a specialized manner in which

0:26:55.340000 --> 0:26:59.460000
 you're going to use
 unicast addresses.

0:26:59.460000 --> 0:27:03.380000
 So there's nothing special about an
 anycast address, which is sort of

0:27:03.380000 --> 0:27:05.920000
 why I split it out under
 communication types.

0:27:05.920000 --> 0:27:09.660000
 I've seen anycast listed as address
 types before, and I'm like, oh, that

0:27:09.660000 --> 0:27:12.160000
 just bothers me so bad.

0:27:12.160000 --> 0:27:16.380000
 All anycast is, is very simply this,
 and we'll do it a couple times.

0:27:16.380000 --> 0:27:19.280000
 I do it all sorts of, for all
 sorts of different things.

0:27:19.280000 --> 0:27:25.500000
 Something as simple as take two servers
 that are offering up, I don't

0:27:25.500000 --> 0:27:28.040000
 care, a webpage, whatever.

0:27:28.040000 --> 0:27:34.560000
 The trick is they have to be offering
 identical services, okay?

0:27:34.560000 --> 0:27:40.300000
 Give them both the same address,
 either IPv6 or IPv4.

0:27:40.300000 --> 0:27:45.640000
 I don't care. Give them the same address
 and then advertise that address

0:27:45.640000 --> 0:27:48.480000
 into your routing protocol.

0:27:48.480000 --> 0:27:54.620000
 So OSPF, RIP, EIGRP, BGP,
 whatever you're running.

0:27:54.620000 --> 0:27:58.480000
 What everybody on your network is going
 to get is they're then going to

0:27:58.480000 --> 0:28:02.460000
 get two routes to the
 same destination.

0:28:02.460000 --> 0:28:07.540000
 Now, do they know that that's two separate
 devices offering the same service?

0:28:07.540000 --> 0:28:12.000000
 No. They just know that they have two
 routes to get to what they perceive

0:28:12.000000 --> 0:28:14.480000
 to be the same destination.

0:28:14.480000 --> 0:28:17.740000
 So of course, if you're not advertising
 this as a host route, you're going

0:28:17.740000 --> 0:28:21.820000
 to have to advertise this as like a slash
 24 subnet, something like that.

0:28:21.820000 --> 0:28:27.160000
 Just make sure it's being advertised
 the same on both routers advertising

0:28:27.160000 --> 0:28:31.880000
 the subnet. Everybody else on the network
 should then see two routes to

0:28:31.880000 --> 0:28:37.720000
 get there. According to any routing
 protocol I've ever seen, that means

0:28:37.720000 --> 0:28:44.560000
 that we then follow whatever that
 routing protocol's method.

0:28:44.560000 --> 0:28:49.060000
 Of course, you know, if it's RIP at
 top count, if it's EIGRP by default,

0:28:49.060000 --> 0:28:50.720000
 it's bandwidth in delay.

0:28:50.720000 --> 0:28:56.100000
 With OSPF, well, there's a lot of variables,
 but to keep things simple,

0:28:56.100000 --> 0:29:00.060000
 for right now, we'll just say that
 it generically comes down to lowest

0:29:00.060000 --> 0:29:03.900000
 cost. There's a lot more to
 OSPF than that, of course.

0:29:03.900000 --> 0:29:08.520000
 We're not going to get into that right
 now, but it always prefers internal

0:29:08.520000 --> 0:29:13.440000
 routes over external routes and routes
 from, you know, area zero and prefer

0:29:13.440000 --> 0:29:15.640000
 type one over type two externals.

0:29:15.640000 --> 0:29:20.480000
 We could go on, but the point is just
 to be very generic right now, we'll

0:29:20.480000 --> 0:29:22.340000
 just say it comes down to cost.

0:29:22.340000 --> 0:29:27.080000
 All the routers look at those two routes
 to get to that any cast destination,

0:29:27.080000 --> 0:29:31.540000
 and they simply follow whatever that
 routing protocol says is the closest

0:29:31.540000 --> 0:29:36.820000
 target. That by definition right
 there is what any cast is.

0:29:36.820000 --> 0:29:42.800000
 It's one, the guy making the decision
 of how to go, to closest.

0:29:42.800000 --> 0:29:47.180000
 Closest as determined by whatever
 the routing protocol is.

0:29:47.180000 --> 0:29:53.880000
 RIP, OSPF, EIGRP, BGP, doesn't matter,
 static routes, I don't care.

0:29:53.880000 --> 0:29:56.600000
 That's by definition
 what any cast is.

0:29:56.600000 --> 0:30:00.840000
 Now there is a big difference, well,
 one big difference, between any cast

0:30:00.840000 --> 0:30:05.240000
 for IPv4 and any cast for IPv6.

0:30:05.240000 --> 0:30:11.640000
 Any cast on IPv4, those two servers
 that you just gave the same address

0:30:11.640000 --> 0:30:15.140000
 would have to be on two
 different segments.

0:30:15.140000 --> 0:30:21.500000
 So you would have to say I don't care,
 have VLAN 10 and VLAN 20, have

0:30:21.500000 --> 0:30:26.980000
 both of them with the same subnet and
 put your servers on each of those

0:30:26.980000 --> 0:30:34.820000
 separate VLANs. Then have two separate
 routers advertise those VLANs into

0:30:34.820000 --> 0:30:37.940000
 OSPF or whatever.

0:30:37.940000 --> 0:30:40.060000
 So they would have to be separate.


0:30:40.060000 --> 0:30:43.060000
 Otherwise what do you end up with?


0:30:43.060000 --> 0:30:47.380000
 Duplicate address exists on the network
 and you have that whole overlapping

0:30:47.380000 --> 0:30:49.940000
 IPv4 address problem.

0:30:49.940000 --> 0:30:55.820000
 With IPv6 you can actually do
 any cast on the same segment.

0:30:55.820000 --> 0:31:00.180000
 You can actually put them literally on
 the same VLAN, on the same segment

0:31:00.180000 --> 0:31:06.000000
 and you'll go to, well frankly, which
 everyone answers, the neighbour

0:31:06.000000 --> 0:31:07.120000
 discovery first.

0:31:07.120000 --> 0:31:13.360000
 So even going for a while here, we're
 going to take a quick break and

0:31:13.360000 --> 0:31:22.000000
 we'll start up in the next section here
 with wrapping up the rest of the

0:31:22.000000 --> 0:31:23.280000
 introduction lesson.

