WEBVTT

0:00:08.460000 --> 0:00:10.420000
 Okay, welcome back.

0:00:10.420000 --> 0:00:14.960000
 So we're going to keep going here with
 the second part of our introduction

0:00:14.960000 --> 0:00:21.800000
 and we're going to go over IPv6,
 solicited node, multicast.

0:00:21.800000 --> 0:00:31.700000
 And this all comes down to a basis of the
 problem that we don't have broadcast.

0:00:31.700000 --> 0:00:36.460000
 So earlier in this lesson I mentioned
 how that was a good thing and yes

0:00:36.460000 --> 0:00:42.600000
 it is. However, as much of a good thing
 as it is, it does bring certain

0:00:42.600000 --> 0:00:43.920000
 problems to the table.

0:00:43.920000 --> 0:00:49.480000
 Okay, the biggest one of course is
 since there's no broadcast, there's

0:00:49.480000 --> 0:00:56.940000
 no ARP. So you know with IPv4 typically
 what you're going to have, machine

0:00:56.940000 --> 0:01:04.840000
 comes online, you need to get your
 address from DHCP, that's going to

0:01:04.840000 --> 0:01:08.720000
 be a whole other discussion but that's
 also part of this whole thing.

0:01:08.720000 --> 0:01:14.920000
 You know I need to get an address from
 DHCP or statically set, doesn't

0:01:14.920000 --> 0:01:18.880000
 matter. Will you get an address and if
 you need to get off your own segment

0:01:18.880000 --> 0:01:22.280000
 that means you're going to get
 a default gateway as well.

0:01:22.280000 --> 0:01:25.240000
 And the next thing you're going to
 have to do is you're going to have

0:01:25.240000 --> 0:01:30.980000
 to find out the layer two address
 of your default gateway.

0:01:30.980000 --> 0:01:35.220000
 So on IPv4 we use of course ARP,
 address resolution protocol.

0:01:35.220000 --> 0:01:42.540000
 For that to work you send out a broadcast
 saying hey, who has this IP

0:01:42.540000 --> 0:01:47.000000
 address, I need to talk to this IP address
 and of course the router will

0:01:47.000000 --> 0:01:51.860000
 answer. That's great except for that
 stupid little word I used in there,

0:01:51.860000 --> 0:01:55.100000
 oh yeah, broadcast.

0:01:55.100000 --> 0:02:00.240000
 See problem. IPv6 does
 not support broadcast.

0:02:00.240000 --> 0:02:02.860000
 So there's no ARP.

0:02:02.860000 --> 0:02:06.180000
 Okay, so this brings
 up the next problem.

0:02:06.180000 --> 0:02:09.500000
 So instead of ARP, since we don't have
 it, we have to use something called

0:02:09.500000 --> 0:02:10.840000
 neighbor discovery.

0:02:10.840000 --> 0:02:15.680000
 So you can pretty much look at neighbor
 discovery as the new ARP.

0:02:15.680000 --> 0:02:20.740000
 Okay, now a couple fundamental differences
 to be aware of though and we'll

0:02:20.740000 --> 0:02:24.460000
 talk more about these as we go through
 different lessons and such.

0:02:24.460000 --> 0:02:28.920000
 ARP was technically
 a separate protocol.

0:02:28.920000 --> 0:02:31.420000
 It was not part of IP.

0:02:31.420000 --> 0:02:32.540000
 It ran outside of it.

0:02:32.540000 --> 0:02:37.040000
 In fact, we were just talking
 about protocol identifiers.

0:02:37.040000 --> 0:02:40.180000
 Well this would have been
 the layer two header.

0:02:40.180000 --> 0:02:47.040000
 Had a different protocol identifier
 for IPv4 versus ARP.

0:02:47.040000 --> 0:02:51.160000
 Well with neighbor discovery, neighbor
 discovery is actually part of ICMP

0:02:51.160000 --> 0:02:58.460000
 version 6 now. So is ICMP still used
 for pings and traces and all that?

0:02:58.460000 --> 0:03:01.660000
 Yes, but it's used for
 a whole lot more now.

0:03:01.660000 --> 0:03:08.840000
 I always like to make the joke that
 IPv6, ICMPV6 is all grown up now.

0:03:08.840000 --> 0:03:14.260000
 It has so much more functionality than
 we had in IPv4 and one of those

0:03:14.260000 --> 0:03:16.480000
 things is neighbor discovery.

0:03:16.480000 --> 0:03:22.420000
 Okay, now for the actual communication,
 this is where this whole solicited

0:03:22.420000 --> 0:03:24.760000
 node multicast address comes in.

0:03:24.760000 --> 0:03:29.760000
 Since we don't have broadcast and I
 don't know who you are, then clearly

0:03:29.760000 --> 0:03:34.660000
 we need some other way to communicate
 initially and that's what this is

0:03:34.660000 --> 0:03:38.740000
 all about. How do we start that initial
 communication between two devices

0:03:38.740000 --> 0:03:43.600000
 when I don't know your layer two address,
 you don't know my layer two

0:03:43.600000 --> 0:03:48.900000
 address, all I know is your layer three
 address but I can't broadcast.

0:03:48.900000 --> 0:03:51.140000
 That's the dilemma.

0:03:51.140000 --> 0:03:53.080000
 That's what this is all about.

0:03:53.080000 --> 0:03:59.980000
 To solve that, we have the solicited node
 multicast address in this format.

0:03:59.980000 --> 0:04:04.160000
 Now we'll get into multicast
 a little bit later.

0:04:04.160000 --> 0:04:07.380000
 Okay, and I'll just go ahead and bring
 up this next point because it specifies

0:04:07.380000 --> 0:04:14.140000
 that I'm about to tell you, FF as I
 already told you means multicast in

0:04:14.140000 --> 0:04:17.220000
 IPv6. So that's the first byte.

0:04:17.220000 --> 0:04:22.840000
 The first byte all
 ones FF multicast.

0:04:22.840000 --> 0:04:32.260000
 The zero is four bits of flag bits
 and they indicate various things.

0:04:32.260000 --> 0:04:37.640000
 They indicate whether this packet contains
 rendezvous point information,

0:04:37.640000 --> 0:04:40.440000
 whether it contains
 prefix information.

0:04:40.440000 --> 0:04:45.180000
 There's four flags of course
 that can go into that field.

0:04:45.180000 --> 0:04:53.180000
 The last byte there,
 the two is the scope.

0:04:53.180000 --> 0:04:56.280000
 Okay, two means link local.

0:04:56.280000 --> 0:05:02.520000
 Okay, one is sort of like a multicast
 loop back and never leaves the device

0:05:02.520000 --> 0:05:06.960000
 and we can talk more about scopes later
 but the fact of the matter is

0:05:06.960000 --> 0:05:16.940000
 the two there simply means do not
 go to, don't leave this segment.

0:05:16.940000 --> 0:05:21.420000
 Okay, so if you're familiar with IPv4
 multicast, it's the equivalent of

0:05:21.420000 --> 0:05:26.520000
 2240, 0, slash 24.

0:05:26.520000 --> 0:05:30.260000
 So it's link local multicast,
 non-routable.

0:05:30.260000 --> 0:05:35.320000
 Okay, now as far as the rest of it goes,
 back up on the first bullet point

0:05:35.320000 --> 0:05:44.460000
 there, the colon colon in that case
 represents four fields of zeros.

0:05:44.460000 --> 0:05:47.900000
 You can tell this because four other
 fields are shown and there's eight

0:05:47.900000 --> 0:05:52.880000
 fields total so the colon colon obviously
 is four fields of zeros.

0:05:52.880000 --> 0:06:02.640000
 This is then followed by one
 colon FF00 colon 0 slash 104.

0:06:02.640000 --> 0:06:07.380000
 Now, if you do a little bit of math
 on that one, since you know we have

0:06:07.380000 --> 0:06:14.640000
 128 bits and I'm specifying 104 here,
 that means that there's 24 bits

0:06:14.640000 --> 0:06:19.280000
 that are not covered at
 the end of this address.

0:06:19.280000 --> 0:06:26.140000
 Okay, now those bits are in fact the
 last two zeros from the FF field.

0:06:26.140000 --> 0:06:31.760000
 Remember that that represents
 eight bits right there.

0:06:31.760000 --> 0:06:38.920000
 And then the last field is 16.

0:06:38.920000 --> 0:06:44.860000
 So between all of those, the last digits,
 remember that those three zeros

0:06:44.860000 --> 0:06:55.980000
 at the end technically represent
 actually six eights, right?

0:06:55.980000 --> 0:06:57.940000
 Now here's what we do.

0:06:57.940000 --> 0:07:00.120000
 So the first part
 of that is given.

0:07:00.120000 --> 0:07:02.260000
 It's assigned by IANA.

0:07:02.260000 --> 0:07:06.800000
 That's not adjustable.

0:07:06.800000 --> 0:07:09.300000
 And by the way, just two
 quick things here.

0:07:09.300000 --> 0:07:13.660000
 This also is used for duplicate address
 detection or DAD, which I'm going

0:07:13.660000 --> 0:07:16.000000
 to talk about here
 in just a minute.

0:07:16.000000 --> 0:07:18.300000
 And this is the part
 that I just said.

0:07:18.300000 --> 0:07:22.960000
 The format leaves the last 24 bits
 of the address that are going to be

0:07:22.960000 --> 0:07:25.720000
 unique per host.

0:07:25.720000 --> 0:07:27.940000
 Okay. Now, what does that mean?

0:07:27.940000 --> 0:07:29.060000
 What do we do with it?

0:07:29.060000 --> 0:07:31.860000
 Well, it doesn't all
 fit on one slide.

0:07:31.860000 --> 0:07:34.700000
 So here's what that means.

0:07:34.700000 --> 0:07:41.440000
 Okay. It means that if your host gets
 an IP address, IPV6 address, as

0:07:41.440000 --> 0:07:45.100000
 shown here. So I'm not going to
 read that whole thing to you.

0:07:45.100000 --> 0:07:49.720000
 That is an example of an address
 sort of based off a MAC address.

0:07:49.720000 --> 0:07:51.940000
 We'll talk more about that later.

0:07:51.940000 --> 0:07:56.640000
 Okay. That means that it's going to
 have the solicited node address of

0:07:56.640000 --> 0:08:01.940000
 FFO2 colon colon 1 colon FF.

0:08:01.940000 --> 0:08:08.200000
 Now, all that part, all of that is from
 the definition that we just looked

0:08:08.200000 --> 0:08:09.740000
 at a second ago.

0:08:09.740000 --> 0:08:14.600000
 Then what we do is we take the last
 24 bits of our real address.

0:08:14.600000 --> 0:08:22.900000
 So the 2, 8, 9C, 5A, we take that and
 we put it on the end of our solicited

0:08:22.900000 --> 0:08:28.700000
 node address. That's how you form
 the solicited node address.

0:08:28.700000 --> 0:08:33.880000
 Now, that said, how does
 all this actually work?

0:08:33.880000 --> 0:08:36.900000
 What's it for? What are
 we actually doing here?

0:08:36.900000 --> 0:08:42.180000
 Okay. So, this is used
 for a couple things.

0:08:42.180000 --> 0:08:49.820000
 First, your machine, you're just booting
 up and you have an IPV6 address.

0:08:49.820000 --> 0:08:53.020000
 I don't care how you got it.

0:08:53.020000 --> 0:08:55.100000
 That's coming up next.

0:08:55.100000 --> 0:08:58.060000
 We have a lot of ways
 to get an address.

0:08:58.060000 --> 0:08:59.420000
 I don't care how you got it.

0:08:59.420000 --> 0:09:00.800000
 That's your address.

0:09:00.800000 --> 0:09:07.900000
 So the first thing you do is you take
 the solicited node address and you

0:09:07.900000 --> 0:09:14.740000
 form off of your address what your unique
 solicited node multicast address

0:09:14.740000 --> 0:09:18.240000
 should be. Now, a lot of people look
 at this and go, but you're only using

0:09:18.240000 --> 0:09:22.660000
 the last 24 bits of
 the IPV6 address.

0:09:22.660000 --> 0:09:30.240000
 Well, agreed. But 24 bits
 is still 2 to the 24th.

0:09:30.240000 --> 0:09:31.800000
 That's still what?

0:09:31.800000 --> 0:09:37.000000
 2.4 million possible combinations,
 roughly.

0:09:37.000000 --> 0:09:42.700000
 I don't think you're going to have two
 machines on the same segment with

0:09:42.700000 --> 0:09:45.560000
 the same last 24 bits.

0:09:45.560000 --> 0:09:52.760000
 See, they used the 24 least significant
 bits out of the address to try

0:09:52.760000 --> 0:09:56.540000
 to make sure that they would
 in fact be unique.

0:09:56.540000 --> 0:10:03.020000
 For a lot of devices, the last 24 bits
 are going to be based off of the

0:10:03.020000 --> 0:10:06.340000
 unique portion of the MAC address.


0:10:06.340000 --> 0:10:10.220000
 Now, man, could we start getting into
 a whole lot of discussion on this

0:10:10.220000 --> 0:10:12.520000
 whole topic right here?

0:10:12.520000 --> 0:10:16.340000
 How does a host get
 his host portion?

0:10:16.340000 --> 0:10:22.080000
 Well, I'm going to talk about that on
 the next slide, but a lot of devices

0:10:22.080000 --> 0:10:34.500000
 left on their own, such as MAC, so
 OSX, Linux, Cisco routers, a lot of

0:10:34.500000 --> 0:10:41.000000
 these devices left to their own device
 by default will pull that based

0:10:41.000000 --> 0:10:43.300000
 on the MAC address.

0:10:43.300000 --> 0:10:49.980000
 Now, if we allow them to do that, that
 means that the last 24 bits of

0:10:49.980000 --> 0:10:57.120000
 the MAC address are supposed to be
 the unique portion of the address.

0:10:57.120000 --> 0:10:58.780000
 But here's the problem.

0:10:58.780000 --> 0:11:03.040000
 It's only unique per individual
 manufacturer.

0:11:03.040000 --> 0:11:09.780000
 So, in other words, if you have a 3-com
-NIC and a Broadcom-NIC, they could

0:11:09.780000 --> 0:11:17.060000
 very easily have the same last 24 bits
 because the OUI portion of the

0:11:17.060000 --> 0:11:23.900000
 MAC address is owned by 3-com or
 Broadcom or Intel or whoever.

0:11:23.900000 --> 0:11:27.440000
 So the fact of the matter is you've
 got to be really, really careful,

0:11:27.440000 --> 0:11:31.200000
 okay? But it's okay because this
 is going to check for us anyway.

0:11:31.200000 --> 0:11:34.240000
 And that's where I'm headed with
 the second part of this.

0:11:34.240000 --> 0:11:41.100000
 So you form this solicited note address
 off of the last 24 bits of your

0:11:41.100000 --> 0:11:46.160000
 address, okay? Oh, by the
 way, quick side note.

0:11:46.160000 --> 0:11:50.160000
 If you notice in that list I gave you
 there, one thing I did not mention

0:11:50.160000 --> 0:11:55.320000
 was Windows. If you're actually sitting
 on a Windows machine right now,

0:11:55.320000 --> 0:12:02.420000
 if you have not actually turned it
 off, IPv6 is enabled by default on

0:12:02.420000 --> 0:12:04.620000
 all of your latest versions
 of Windows.

0:12:04.620000 --> 0:12:09.400000
 And if you just do an IP config and
 you look at your IPv6 address, it'll

0:12:09.400000 --> 0:12:12.080000
 have created one of those linked local
 addresses I was talking about a

0:12:12.080000 --> 0:12:13.920000
 minute ago, the FE80.

0:12:13.920000 --> 0:12:18.360000
 And if you look at it, you'll notice
 it is not based off your MAC address,

0:12:18.360000 --> 0:12:23.820000
 okay? It's referred to as an IPv6 privacy
 address, at least most places

0:12:23.820000 --> 0:12:26.880000
 I've seen it listed.

0:12:26.880000 --> 0:12:32.040000
 Basically, Microsoft doesn't
 use the MAC address directly.

0:12:32.040000 --> 0:12:35.140000
 I think it actually still uses it
 as a seed or something like that.

0:12:35.140000 --> 0:12:41.820000
 I'm not 100% sure how they do
 it internally inside the OS.

0:12:41.820000 --> 0:12:45.440000
 But the fact of the matter is, it's
 not based on the MAC address.

0:12:45.440000 --> 0:12:47.660000
 And that's why I was a little hesitant
 to say, oh, we could have a whole

0:12:47.660000 --> 0:12:49.540000
 big discussion, yeah.

0:12:49.540000 --> 0:12:53.380000
 Different OSs can do it
 different ways, okay?

0:12:53.380000 --> 0:12:57.120000
 And just to let you know, the reason,
 there's a little bit of concern

0:12:57.120000 --> 0:13:02.980000
 in the IPv6 field that if you're going
 to base the host portion off the

0:13:02.980000 --> 0:13:06.840000
 MAC address, then that's
 way too trackable.

0:13:06.840000 --> 0:13:10.680000
 You know, somebody could track you
 or laptop as you travel around the

0:13:10.680000 --> 0:13:13.180000
 world and go, well, you were there and
 you were there and you logged in

0:13:13.180000 --> 0:13:18.040000
 over here because your host
 portion is your MAC address.

0:13:18.040000 --> 0:13:22.140000
 So I could see everywhere that that
 host portion was coming from and see

0:13:22.140000 --> 0:13:23.320000
 everywhere you went.

0:13:23.320000 --> 0:13:27.580000
 Trace that down to your MAC address,
 down to who that was sold to, what

0:13:27.580000 --> 0:13:30.360000
 laptop, who owns the laptop.

0:13:30.360000 --> 0:13:34.540000
 Again, a little bit of
 concern over privacy.

0:13:34.540000 --> 0:13:37.920000
 And this is why, you know, like I said,
 Windows doesn't even base it on

0:13:37.920000 --> 0:13:39.400000
 the MAC address.

0:13:39.400000 --> 0:13:43.260000
 And like for all my personal devices
 and stuff, if I have IPv6 turned

0:13:43.260000 --> 0:13:45.920000
 on, I just hard code
 the addresses.

0:13:45.920000 --> 0:13:50.360000
 I don't let the host
 make up its own.

0:13:50.360000 --> 0:13:51.900000
 So, and that's easy to do.

0:13:51.900000 --> 0:13:55.660000
 I'll show you all this when we get
 to config and we'll talk more about

0:13:55.660000 --> 0:13:59.440000
 it. But right now I just
 want to focus on this.

0:13:59.440000 --> 0:14:02.900000
 So again, so back to
 the last 24 bits.

0:14:02.900000 --> 0:14:08.720000
 What if you did have two network cards
 or whatever that had the same last

0:14:08.720000 --> 0:14:13.680000
 24 bits just out of, well, you're incredibly,
 I don't know if that would

0:14:13.680000 --> 0:14:17.080000
 be considered incredibly bad
 luck or incredibly good luck.

0:14:17.080000 --> 0:14:19.660000
 I don't know if I should tell you that
 if that happens to you that you

0:14:19.660000 --> 0:14:22.600000
 should run out and play the lottery
 immediately that night.

0:14:22.600000 --> 0:14:26.540000
 Or if I should tell you to never, ever
 play the lottery again ever because

0:14:26.540000 --> 0:14:27.800000
 you have the worst
 luck in the world.

0:14:27.800000 --> 0:14:31.400000
 I'm not, I'm not really sure which
 way to go with that for you.

0:14:31.400000 --> 0:14:37.400000
 But you know, joking aside though,
 the odds are so incredibly against

0:14:37.400000 --> 0:14:41.740000
 it. I don't think you have to lose
 any sleep tonight over this.

0:14:41.740000 --> 0:14:47.600000
 But if it were to happen, the next
 thing your device is going to do is

0:14:47.600000 --> 0:14:54.600000
 it is actually going to send out a packet
 with a destination of the solicited

0:14:54.600000 --> 0:14:58.860000
 node address and it actually sends out
 what's called a neighbor discovery

0:14:58.860000 --> 0:15:03.860000
 message. It's the equivalent
 of an ARP request.

0:15:03.860000 --> 0:15:07.120000
 Okay, because I did say I would equate
 this to IPv4 as we went through.

0:15:07.120000 --> 0:15:14.260000
 So yeah, it's the equivalent of what
 we would have called on our request.

0:15:14.260000 --> 0:15:20.040000
 But of course it's not, it's
 a neighbor discovery message.

0:15:20.040000 --> 0:15:28.960000
 What we're hoping for, the cross your
 fingers and hope is nobody answers.

0:15:28.960000 --> 0:15:33.580000
 That's the address
 we intend to use.

0:15:33.580000 --> 0:15:36.720000
 Therefore nobody else
 should be using it.

0:15:36.720000 --> 0:15:39.920000
 Therefore nobody should answer.

0:15:39.920000 --> 0:15:45.960000
 This process is what we referred to on
 the previous slide there as duplicate

0:15:45.960000 --> 0:15:48.560000
 address detection or dad.

0:15:48.560000 --> 0:15:50.380000
 That's that whole process.

0:15:50.380000 --> 0:15:54.880000
 If it does get an answer, it means
 somebody else is already using the

0:15:54.880000 --> 0:15:58.160000
 address and he will just
 create another one.

0:15:58.160000 --> 0:16:01.280000
 How? It depends on the
 operating system.

0:16:01.280000 --> 0:16:04.880000
 Okay, I'm not going to sit and try and
 answer for every operating system

0:16:04.880000 --> 0:16:10.540000
 out there. But he will try to create
 another address for himself then.

0:16:10.540000 --> 0:16:15.300000
 Okay? It's pseudo random anyway,
 so it really doesn't matter.

0:16:15.300000 --> 0:16:19.100000
 He'll just pick another address
 based on the operating system.

0:16:19.100000 --> 0:16:22.020000
 Now let's go the other way.

0:16:22.020000 --> 0:16:25.400000
 Let's assume that dad passed.

0:16:25.400000 --> 0:16:32.220000
 So he did not have a duplicate address
 and he goes ahead and then assigns

0:16:32.220000 --> 0:16:36.560000
 that IPv6 address
 to his interface.

0:16:36.560000 --> 0:16:38.600000
 So that's now his interface.

0:16:38.600000 --> 0:16:43.040000
 He's done duplicate address detection
 on it and everything we've just

0:16:43.040000 --> 0:16:48.600000
 discussed, the solicited node multi
-cast address has accomplished job

0:16:48.600000 --> 0:16:54.960000
 number one. That's dad
 and joining that group.

0:16:54.960000 --> 0:17:00.360000
 So the two things that this host will
 do after he does dad is he assigns

0:17:00.360000 --> 0:17:10.380000
 this IP, V6 address to his interface
 and he joins the solicited node multi

0:17:10.380000 --> 0:17:13.160000
-cast group on that interface.

0:17:13.160000 --> 0:17:19.120000
 So then instead of sending traffic to
 that address, he starts listening

0:17:19.120000 --> 0:17:25.580000
 on the FFO2 colon colon one colon
 F2 eight colon nine C five A.

0:17:25.580000 --> 0:17:27.960000
 He listens to that address.

0:17:27.960000 --> 0:17:29.640000
 What's he listening for?

0:17:29.640000 --> 0:17:32.000000
 Neighbor discovery messages.

0:17:32.000000 --> 0:17:35.420000
 He's listening for somebody
 else to ask.

0:17:35.420000 --> 0:17:38.640000
 Ask what? Well, one of two things.


0:17:38.640000 --> 0:17:42.520000
 He's looking for the other person to
 say, hey, does anybody have this

0:17:42.520000 --> 0:17:44.160000
 because I want to use it.

0:17:44.160000 --> 0:17:46.220000
 And so he can basically say no.

0:17:46.220000 --> 0:17:48.180000
 Now again, that's not exactly
 how that works.

0:17:48.180000 --> 0:17:52.340000
 I'm using a little bit of
 poetic license there.

0:17:52.340000 --> 0:17:56.380000
 Okay. Again, basically the other guy
 is going to send a solicited node

0:17:56.380000 --> 0:18:02.060000
 neighbor discovery message for our address
 and we're going to answer and

0:18:02.060000 --> 0:18:03.880000
 he's going to do the dad part.

0:18:03.880000 --> 0:18:04.600000
 He's going to go.

0:18:04.600000 --> 0:18:07.840000
 Oh, somebody else is listening
 to that group.

0:18:07.840000 --> 0:18:11.260000
 Oh darn. And he'll have
 to pick another address.

0:18:11.260000 --> 0:18:17.700000
 Okay. So this would be once we join,
 this is us doing the other side of

0:18:17.700000 --> 0:18:21.060000
 the dad. We're supplying
 the answer.

0:18:21.060000 --> 0:18:25.660000
 Okay. And of course, the other part
 of that goes hand in hand with it.

0:18:25.660000 --> 0:18:30.260000
 And once we've joined this group, we
 honestly don't know or care which

0:18:30.260000 --> 0:18:31.340000
 one we're doing.

0:18:31.340000 --> 0:18:33.060000
 It's the same message to us.

0:18:33.060000 --> 0:18:34.720000
 And that would be
 the main purpose.

0:18:34.720000 --> 0:18:37.160000
 We went down this
 whole discussion.

0:18:37.160000 --> 0:18:42.280000
 Remember, we're trying to deal with the
 fact that we don't have ARP anymore.

0:18:42.280000 --> 0:18:48.100000
 And that was to do layer three to
 layer two address resolution.

0:18:48.100000 --> 0:18:53.780000
 So again, somebody and maybe on the
 default gateway, maybe this is, you

0:18:53.780000 --> 0:18:57.840000
 know, doesn't happen very often anymore,
 but maybe this is, you know,

0:18:57.840000 --> 0:19:02.040000
 host to host traffic on the same segment,
 maybe more in your data center

0:19:02.040000 --> 0:19:05.880000
 with multi-tiered applications than
 it happens out on the client side

0:19:05.880000 --> 0:19:10.500000
 these days. But in any case, you know,
 maybe this is just, you know, peer

0:19:10.500000 --> 0:19:14.400000
 to peer traffic on the same segment,
 not even going to the router.

0:19:14.400000 --> 0:19:15.820000
 It doesn't matter who it is.

0:19:15.820000 --> 0:19:17.660000
 They all do the same thing.

0:19:17.660000 --> 0:19:21.160000
 I need to find out what your MAC address
 is if I'm going to talk to you

0:19:21.160000 --> 0:19:23.560000
 on an ethernet segment.

0:19:23.560000 --> 0:19:31.660000
 So since I know the IPv6 address I'm
 trying to talk to, I then take the

0:19:31.660000 --> 0:19:37.380000
 IPv6 address that is my default gateway,
 the guy I'm trying to talk to,

0:19:37.380000 --> 0:19:44.440000
 whoever it is, I take his IPv6 address
 and again, form the solicited node

0:19:44.440000 --> 0:19:50.000000
 multicast address out of that going
 through this exact same method and

0:19:50.000000 --> 0:19:57.000000
 then I send my neighbor discovery solicitation
 message to that address.

0:19:57.000000 --> 0:19:58.680000
 Now here's the cool part.

0:19:58.680000 --> 0:20:05.640000
 Depending on how good your switch is,
 and I sort of mentioned this earlier,

0:20:05.640000 --> 0:20:08.980000
 you may or may not have a switch
 that can handle this.

0:20:08.980000 --> 0:20:11.760000
 This is link local multicast.

0:20:11.760000 --> 0:20:17.220000
 So it's not going to have IGMP
 joins and such for it.

0:20:17.220000 --> 0:20:22.560000
 Therefore, IGMP snooping on your
 switch is not going to help here.

0:20:22.560000 --> 0:20:26.700000
 So there's a pretty good chance that
 this is going to actually get flooded

0:20:26.700000 --> 0:20:31.820000
 by the switch to every single node on
 that switch, every single node on

0:20:31.820000 --> 0:20:37.680000
 that VLAN. Now that's okay though, because
 if you go back to our discussion

0:20:37.680000 --> 0:20:42.820000
 a little bit earlier, what I said was,
 okay, let's say your switch does

0:20:42.820000 --> 0:20:45.540000
 in fact flood this.

0:20:45.540000 --> 0:20:52.100000
 The problem with broadcast is when that
 traffic gets to your network card,

0:20:52.100000 --> 0:20:56.560000
 your network card looks at that and
 goes, oh, it's a broadcast, it's for

0:20:56.560000 --> 0:21:01.220000
 me, process it and sends
 it up to the OS.

0:21:01.220000 --> 0:21:08.540000
 Well, the problem is if it's not really
 for you, that just wasted your

0:21:08.540000 --> 0:21:11.460000
 CPU. Now here's what's
 cool about this.

0:21:11.460000 --> 0:21:14.020000
 Yes, this is not a broadcast.

0:21:14.020000 --> 0:21:19.680000
 So let's say this in fact does get flooded
 by your switch to every single

0:21:19.680000 --> 0:21:21.100000
 node on the segment.

0:21:21.100000 --> 0:21:26.260000
 Well, when the traffic comes in, you
 know, you got to remember, and maybe

0:21:26.260000 --> 0:21:29.260000
 I need to bring this up because maybe
 some of you are not aware of this,

0:21:29.260000 --> 0:21:39.980000
 every multicast address, be it IPv4
 or IPv6 gets mapped to a layer two

0:21:39.980000 --> 0:21:43.760000
 multicast MAC address.

0:21:43.760000 --> 0:21:49.960000
 And when your machine joins this solicited
 node multicast group, he creates

0:21:49.960000 --> 0:21:57.060000
 the corresponding multicast MAC address
 and installs that onto his network

0:21:57.060000 --> 0:22:02.500000
 card. So that when this frame shows
 up on the wire, you know, just bits

0:22:02.500000 --> 0:22:07.640000
 coming in on the wire, he looks at the
 destination address, he sees the

0:22:07.640000 --> 0:22:12.460000
 FFO2, colon colon one colon FFF, you
 know, he sees this solicited node

0:22:12.460000 --> 0:22:18.140000
 multicast address, he looks at the
 multicast group and he looks at his

0:22:18.140000 --> 0:22:23.960000
 network card and the MAC addresses on
 the network card and he goes, this

0:22:23.960000 --> 0:22:26.920000
 isn't for me, throw it away.

0:22:26.920000 --> 0:22:31.700000
 Right there in hardware, right there
 in the network card, never hits the

0:22:31.700000 --> 0:22:36.500000
 CPU of the device, never hits
 the operating system, why?

0:22:36.500000 --> 0:22:43.540000
 Because that network card doesn't own
 the corresponding multicast MAC

0:22:43.540000 --> 0:22:51.760000
 address. So he's going to have his unicast
 MAC address and his solicited

0:22:51.760000 --> 0:23:01.020000
 node multicast MAC address, but he will
 not have somebody else's multicast

0:23:01.020000 --> 0:23:05.840000
 MAC address for their
 solicited node.

0:23:05.840000 --> 0:23:10.860000
 So it means that everybody on the segment
 is going to have joined and

0:23:10.860000 --> 0:23:18.660000
 have a unique, that's the key, a unique
 multicast MAC address added to

0:23:18.660000 --> 0:23:23.640000
 their neck. So when that frame comes
 in and I look at that MAC address

0:23:23.640000 --> 0:23:29.020000
 right there in hardware, I go, this
 is not for me, this is not my MAC

0:23:29.020000 --> 0:23:31.100000
 address, throw it away.

0:23:31.100000 --> 0:23:37.100000
 Now of course, the guy who I am trying
 to contact, of course, the opposite

0:23:37.100000 --> 0:23:42.640000
 happens, he goes, he looks at that,
 he is joined to that multicast MAC

0:23:42.640000 --> 0:23:47.520000
 address. He looks at it and goes, yep,
 this is for my MAC address, bring

0:23:47.520000 --> 0:23:53.880000
 it in, send it on up and he sends it
 up to the CPU, the operating system,

0:23:53.880000 --> 0:23:58.500000
 they of course look at it and see that this
 is a neighbor discovery solicitation

0:23:58.500000 --> 0:24:04.760000
 message and go, somebody's trying to get
 our info and he will send a neighbor

0:24:04.760000 --> 0:24:06.480000
 discovery response.

0:24:06.480000 --> 0:24:13.580000
 And he will then answer and whoever
 was trying to talk to me now has my

0:24:13.580000 --> 0:24:19.980000
 MAC address and of course just like
 ARP, his MAC address was included

0:24:19.980000 --> 0:24:23.240000
 in the original communication
 to me.

0:24:23.240000 --> 0:24:28.840000
 So there you go, that, and we'll see
 all this work as we start getting

0:24:28.840000 --> 0:24:32.260000
 into configuration and setting
 this all up, not to worry.

0:24:32.260000 --> 0:24:35.660000
 Like I said, this first lesson, I want
 to get all the theory out of the

0:24:35.660000 --> 0:24:40.720000
 way and we'll see it as we go, never
 hurts to see things twice anyway.

0:24:40.720000 --> 0:24:44.020000
 So this is all our theory and I'll
 show you all of this stuff working

0:24:44.020000 --> 0:24:50.160000
 as we go. But that's how soliciting
 a multicast works.

0:24:50.160000 --> 0:24:54.540000
 So yeah, you're not going to be going
 to your devices at looking at ARP

0:24:54.540000 --> 0:24:57.720000
 anymore. Now you're going to be looking
 at neighbor discovery, that's

