1 00:00:02,936 --> 00:00:05,772 [music] 2 00:00:09,009 --> 00:00:11,311 Okay, so for this next section we're going to take a look 3 00:00:11,311 --> 00:00:19,920 at MPLS and a technology called 6PE which basically means that 4 00:00:19,920 --> 00:00:25,659 we're going to be running IPv6 between-- in this example we're 5 00:00:25,659 --> 00:00:30,030 going to use switch 1 and router 1, and then IPv6 between 6 00:00:30,030 --> 00:00:33,233 switch 2 and router 4. And again, we're just going to 7 00:00:33,233 --> 00:00:35,502 do the top site again for this one. I'm not even going 8 00:00:35,502 --> 00:00:40,607 to bring in router 4 here - we don't really need to. But 9 00:00:40,607 --> 00:00:45,312 we're going to be running IPv6 there and in order 10 00:00:45,312 --> 00:00:48,815 to do this we have to fire up some things in the core, 11 00:00:48,815 --> 00:00:52,986 and for some of this I have already done it. So let 12 00:00:52,986 --> 00:00:57,090 me switch over here to a diagram that I can draw on a little 13 00:00:57,090 --> 00:01:04,564 bit easier here, and we're going to be running BGP in the corner, 14 00:01:04,564 --> 00:01:08,235 I already have this part set up. So I have BGP set up from 15 00:01:08,235 --> 00:01:16,677 R1 to R4 and from our R5 to R1. In this case, router 16 00:01:16,677 --> 00:01:25,586 1 is acting like a route reflector for iBGP. This 17 00:01:25,586 --> 00:01:31,558 multiprotocol BGP session is actually running address 18 00:01:31,558 --> 00:01:37,764 families IPv6. That's what we're going to use for this first part 19 00:01:37,764 --> 00:01:43,570 and then when we get into 6VPE in the next lesson, 20 00:01:43,570 --> 00:01:45,505 we're going to be taking advantage of the fact that it's 21 00:01:45,505 --> 00:01:53,580 also running VPNv6 address family. So we're going to 22 00:01:53,580 --> 00:01:58,719 be using one of those to start with just the IPv6. 23 00:01:58,719 --> 00:02:04,458 The whole thing with 6PE is we're going to get IPv6 through 24 00:02:04,458 --> 00:02:06,860 an IPv4 cord, just like we've been doing with these 25 00:02:06,860 --> 00:02:10,864 other technologies, but the big difference between this 26 00:02:10,864 --> 00:02:16,870 and 6VPE is with 6VPE we're actually going to be using 27 00:02:16,870 --> 00:02:21,308 VRFs and we'll be able to separate customer A 28 00:02:21,308 --> 00:02:24,678 from customer B traffic for example. And we may not 29 00:02:24,678 --> 00:02:28,281 go so far as to actually configure customer B, 30 00:02:28,281 --> 00:02:32,719 I don't think we really need to, but we will run IPv6 through a 31 00:02:32,719 --> 00:02:39,659 VRF on the PE routers. Now, for 6PE, that's the piece 32 00:02:39,659 --> 00:02:42,996 we're not doing. We're not going to use VRFs, we're 33 00:02:42,996 --> 00:02:45,465 going to run regular - in fact it's already up and 34 00:02:45,465 --> 00:02:50,404 running from our previous lesson - OSPF between switch 1 35 00:02:50,404 --> 00:02:55,375 and router 1, OSPF between switch 2 and router 4, 36 00:02:55,375 --> 00:02:58,879 I've already turned on MPLS in the core, and I'm going 37 00:02:58,879 --> 00:03:01,481 to show you some of this stuff. Then we're just going 38 00:03:01,481 --> 00:03:05,619 to take a look at how we actually get 6PE across this. 39 00:03:05,619 --> 00:03:10,490 There's honestly one command that's really sort of the magic 40 00:03:10,490 --> 00:03:13,860 here. So if we switched over to the command line, 41 00:03:13,860 --> 00:03:17,330 and we take a look at router 1. Let me show you the BGP stuff 42 00:03:17,330 --> 00:03:22,769 first. So do show run section BGP. I just want to show 43 00:03:22,769 --> 00:03:27,274 you that he has neighborship. I use the peer group here 44 00:03:27,274 --> 00:03:31,445 just to sort of shorten it up a little bit. It doesn't help 45 00:03:31,445 --> 00:03:33,814 a whole ton when you start dealing with address 46 00:03:33,814 --> 00:03:37,250 families. Because as you can see, you still have to activate 47 00:03:37,250 --> 00:03:41,455 each neighbor separately per address family even 48 00:03:41,455 --> 00:03:45,459 with peer groups. Now what it does shorten a little 49 00:03:45,459 --> 00:03:50,030 bit is the whole send community extended under the VPNv4 50 00:03:50,030 --> 00:03:54,601 and VPNv6 address families, as well as 51 00:03:54,601 --> 00:04:00,674 having the router reflector client for the IPv6 end both 52 00:04:00,674 --> 00:04:03,810 VPN address families. So what we're going to be 53 00:04:03,810 --> 00:04:08,381 starting with is the address family IPv6. Like I 54 00:04:08,381 --> 00:04:12,819 said, the VPN address families are for later. 55 00:04:12,819 --> 00:04:15,122 They're just already there, already built, already 56 00:04:15,122 --> 00:04:17,824 established, and the neighborships are up. 57 00:04:17,824 --> 00:04:24,698 If we go to 4 and 5 they're going to look very similar. 58 00:04:24,698 --> 00:04:30,003 Okay, so on these guys we have the address families, the 59 00:04:30,003 --> 00:04:33,874 neighbors, and that's pretty much it just to send communities 60 00:04:33,874 --> 00:04:37,978 because they're not route deflectors. So, the multi 61 00:04:37,978 --> 00:04:42,315 protocol IBGP is already set up, and running and if 62 00:04:42,315 --> 00:04:47,854 I said do show IPv6 route OSPF we should already have-- 63 00:04:47,854 --> 00:04:53,426 in this case we have switch 2's routes here already, and 64 00:04:53,426 --> 00:05:00,834 if we were to go to router 1, do show IPv6 route OSPF we 65 00:05:00,834 --> 00:05:08,175 have the routes from switch 1. So for the most part, so far 66 00:05:08,175 --> 00:05:13,346 everything is just basic routing so we have an IPv6 BGP 67 00:05:13,346 --> 00:05:16,616 connection and again notice that with my IPV6 address 68 00:05:16,616 --> 00:05:21,321 family I'm doing IPv4 peers so we talked about this in the 69 00:05:21,321 --> 00:05:26,459 BGP section about running IPv6 routes over 70 00:05:26,459 --> 00:05:30,931 an IPv4 neighbor ship. Okay? The other thing we've already 71 00:05:30,931 --> 00:05:35,669 done is do show MPLS interface and you can 72 00:05:35,669 --> 00:05:38,471 see and we can go to every device in the middle. 73 00:05:38,471 --> 00:05:40,607 I'm not going to show you every single one. We're going to 74 00:05:40,607 --> 00:05:48,081 say there's one do show MPLS interface on router 3 75 00:05:48,081 --> 00:05:54,020 is running on three interfaces. Do show MPLS LEP neighbor. 76 00:05:54,020 --> 00:05:58,558 You can see the neighbor ships are up so two ways you can do 77 00:05:58,558 --> 00:06:01,127 this part. Of course, you can go to the interface 78 00:06:01,127 --> 00:06:07,867 and you can say, mpls ip, or the way I did it if you say, 79 00:06:07,867 --> 00:06:15,008 do show run section ospf. You can actually just turn it on 80 00:06:15,008 --> 00:06:19,679 with the MPLS LDP autoconfig command, which just turns 81 00:06:19,679 --> 00:06:24,050 on MPLS on all interfaces running that instance of OSPF. 82 00:06:24,050 --> 00:06:27,354 And since I'm using that instance of OSPF for the 83 00:06:27,354 --> 00:06:31,658 core, then that's all the interfaces in the core. 84 00:06:31,658 --> 00:06:35,161 So just for a little bit of a set up here, just turned on 85 00:06:35,161 --> 00:06:39,966 MPLS on all the interfaces inside the cloud in our 86 00:06:39,966 --> 00:06:43,403 diagram. So just for the sake of letting you know here, 87 00:06:43,403 --> 00:06:48,775 where we turned it on is basically we're running MPLS 88 00:06:48,775 --> 00:06:58,985 here, and here, and here, here. This is actually 89 00:06:58,985 --> 00:07:01,655 two interfaces on Router 3 on the drawing, 90 00:07:01,655 --> 00:07:06,860 I just have them coming together there and of course on router 4 91 00:07:06,860 --> 00:07:12,065 and router 5. So those are all running MPLS, nothing more 92 00:07:12,065 --> 00:07:16,536 than MPLS IP. Well, I actually did autoconfig 93 00:07:16,536 --> 00:07:20,073 but the same thing, just enabling MPLS and 94 00:07:20,073 --> 00:07:23,076 then our BGP session is set up. That's it, 95 00:07:23,076 --> 00:07:28,915 and of course running OSPF out at the edges for IPv6. 96 00:07:28,915 --> 00:07:34,754 And the cool thing is, with all of that set up, you're actually 97 00:07:34,754 --> 00:07:42,762 95% of the way to 6PE, we're almost there, all we have to 98 00:07:42,762 --> 00:07:49,069 do is switch back over here to the command line and 99 00:07:49,069 --> 00:07:54,641 let's go to router 1 and there's really two things we have to do. 100 00:07:54,641 --> 00:08:01,514 We have to say IPv6 router OSPF1, and we have to say to 101 00:08:01,514 --> 00:08:07,187 redistribute BGP 100. I don't really have to do anything 102 00:08:07,187 --> 00:08:20,200 else. We have to do the same thing on router 4. 103 00:08:20,200 --> 00:08:24,938 There we go. And then, of course, under router BGP 100, 104 00:08:24,938 --> 00:08:35,148 you say address-family ipv6 unicast redistribute ospf 1. 105 00:08:35,148 --> 00:08:40,387 Now you don't have to do this part but I usually do to say, 106 00:08:40,387 --> 00:08:49,095 match internal external NSSA. The default behavior - and this 107 00:08:49,095 --> 00:08:53,366 applies to IPv4 and IPv6 - the default behavior 108 00:08:53,366 --> 00:09:00,507 when you redistribute OSPF into BGP is it only gets OSPF 109 00:09:00,507 --> 00:09:05,812 internal routes Well, you know, if this is our enterprise, 110 00:09:05,812 --> 00:09:10,450 and we have control over all of this, and we're controlling that 111 00:09:10,450 --> 00:09:13,787 they're all internal, then that's fine. But in 112 00:09:13,787 --> 00:09:16,489 the case of maybe this is a service provider, 113 00:09:16,489 --> 00:09:19,659 which I don't think they would be doing 6PE. They would 114 00:09:19,659 --> 00:09:22,762 probably be doing 6VPE. But we'll do the same thing 115 00:09:22,762 --> 00:09:25,432 when we get there. When I do the redistribution, I just 116 00:09:25,432 --> 00:09:28,501 include everything. If it's coming in, I should pass it 117 00:09:28,501 --> 00:09:30,804 through BGP. That's sort of what I'm-- sort of 118 00:09:30,804 --> 00:09:35,275 the way I'm looking at it here. So we hit enter there. 119 00:09:35,275 --> 00:09:41,047 We go over to router 1. We say router BGP 100. 120 00:09:41,047 --> 00:09:50,056 Address family IPv6 unicast. We say redistribute OSPF 1. 121 00:09:50,056 --> 00:09:55,728 Match internal external NSSA. Now if we've done this 122 00:09:55,728 --> 00:10:00,467 correctly, we should be able to say do show BGP IPv6 unicast. 123 00:10:00,467 --> 00:10:03,169 We should have some routes coming from the other side. 124 00:10:03,169 --> 00:10:12,712 And we do. Here they are. There's the seven that we're 125 00:10:12,712 --> 00:10:16,282 getting from our local connection. Now we have 126 00:10:16,282 --> 00:10:18,751 not selected these as best yet, it should go. Give it 127 00:10:18,751 --> 00:10:20,753 a second here. Yeah, it's not going to yet. 128 00:10:20,753 --> 00:10:23,756 That's right. I was trying to show you that this isn't quite 129 00:10:23,756 --> 00:10:26,359 right yet and that I'm expecting him to show up as best. 130 00:10:26,359 --> 00:10:30,163 They're not going to because it's not quite right yet. 131 00:10:30,163 --> 00:10:40,940 Here's the problem is if I ask to see that prefix, 132 00:10:40,940 --> 00:10:48,147 what I don't have for this one available, no best path, 133 00:10:48,147 --> 00:10:53,019 inaccessible. This is the exact same thing that we saw 134 00:10:53,019 --> 00:10:57,090 when we were doing this before. At this point, 135 00:10:57,090 --> 00:11:04,030 there is no knowledge, no tie between BGP and MPLS. 136 00:11:04,030 --> 00:11:07,233 So, oh yes, wonderful, we're running MPLS under all of this. 137 00:11:07,233 --> 00:11:12,839 In fact, if I were to say do trace to 4.4.4.4, it's actually 138 00:11:12,839 --> 00:11:17,644 going to show labels being passed through our environment. 139 00:11:17,644 --> 00:11:21,948 So it's using label 18, which is a little hard to tell here 140 00:11:21,948 --> 00:11:25,018 because I didn't change label space or label range or 141 00:11:25,018 --> 00:11:29,822 anything, so it's actually swapping label 18 for label 18, 142 00:11:29,822 --> 00:11:33,226 that's technically a label swap, but the point is it is label 143 00:11:33,226 --> 00:11:37,597 switching through the core. The problem that we have right 144 00:11:37,597 --> 00:11:42,669 now is he's using that mapped IPv4 address that we 145 00:11:42,669 --> 00:11:45,271 saw in the previous lesson but he's still 146 00:11:45,271 --> 00:11:49,776 saying it's inaccessible just like he did before and 147 00:11:49,776 --> 00:11:56,115 the reason for this is very simple: there's no label. 148 00:11:56,115 --> 00:12:01,888 We're not doing 6PE yet. Right now all we're doing is 149 00:12:01,888 --> 00:12:07,927 literally sending IPv6 routes across an IPv4 BGP connection. 150 00:12:07,927 --> 00:12:11,164 Okay, we've done that before, and we saw before that this said 151 00:12:11,164 --> 00:12:17,070 inaccessible, what makes this 6PE. So if you want to 152 00:12:17,070 --> 00:12:20,073 call this the magic, the super command, the part 153 00:12:20,073 --> 00:12:24,010 that makes it all work, is we have to say, 154 00:12:24,010 --> 00:12:35,254 neighbor 10-- not 10. Sorry. I'm on Router 1. Neighbor 4444 send 155 00:12:35,254 --> 00:12:39,959 label, and that's because it's a peer group. So actually 156 00:12:39,959 --> 00:12:43,696 you just say, neighbor-- I forget what I called it. 157 00:12:43,696 --> 00:12:56,909 Peers, I think, send label. And it's saying the update 158 00:12:56,909 --> 00:12:59,712 source is not loopback0, which I thought it was. 159 00:12:59,712 --> 00:13:06,286 So we'll go ahead and change that as well, up loopback 0 160 00:13:06,286 --> 00:13:19,932 and then hop over to Router 4 neighbor 1.1.1 send label. Okay, 161 00:13:19,932 --> 00:13:24,137 capability changes, neighbor comes back up and now let's 162 00:13:24,137 --> 00:13:31,244 go back and look at that route on Router 1 again. 163 00:13:31,244 --> 00:13:38,284 Notice now, it has an outgoing label of 23. So now 164 00:13:38,284 --> 00:13:42,288 it understands that when you're sending to this route, 165 00:13:42,288 --> 00:13:45,892 you're putting a label on it. So now that he knows he's no 166 00:13:45,892 --> 00:13:50,029 longer actually forwarding to this guy, he's going to 167 00:13:50,029 --> 00:13:55,334 actually put a label on it of 23 as it goes out. 168 00:13:55,334 --> 00:13:58,571 It also means of course that when-- that the-- it's now got 169 00:13:58,571 --> 00:14:03,743 [best?] as 1. So now it's using the route. 170 00:14:03,743 --> 00:14:08,681 Do show IPv6 route from BGP, and you're going to see that we have 171 00:14:08,681 --> 00:14:13,920 two labels, two routes, to get to the other side which means 172 00:14:13,920 --> 00:14:19,692 just like that-- just by saying send label we're doing 6PE. 173 00:14:19,692 --> 00:14:21,527 There's labels for this to go the whole way through. 174 00:14:21,527 --> 00:14:29,268 If I said do show IPv6 seff for one of these prefixes - and I'll 175 00:14:29,268 --> 00:14:39,912 pick on the 88 here - you can see here that he's imposing two 176 00:14:39,912 --> 00:14:44,617 labels. 18, which is the forwarding label, and we 177 00:14:44,617 --> 00:14:46,953 already knew that 18 was the forwarding label from when 178 00:14:46,953 --> 00:14:51,124 we did our trace up here. There it is, label 18. So you 179 00:14:51,124 --> 00:14:53,626 can call that-- let's see, oh goodness, can I think of all 180 00:14:53,626 --> 00:14:57,430 the names I've heard for it? Left label, top label, 181 00:14:57,430 --> 00:15:02,034 forwarding label, LDP label, that's all I'm getting off 182 00:15:02,034 --> 00:15:06,105 the top of my head, but that's all label 18. 183 00:15:06,105 --> 00:15:12,745 Then then VPN label is label 23. So, even though this 184 00:15:12,745 --> 00:15:17,016 is not using VRFs it's not separating the routing 185 00:15:17,016 --> 00:15:21,721 tables on the PE routers. It's still using labels because this 186 00:15:21,721 --> 00:15:27,393 is still a VPN so-- or again, a form of a tunnel because 187 00:15:27,393 --> 00:15:35,334 we're encapsulating our IPv6 in the labels. You can see 188 00:15:35,334 --> 00:15:38,371 right there it's going to put those two labels on. 189 00:15:38,371 --> 00:15:40,039 And you know what? I said I wasn't going to bring 190 00:15:40,039 --> 00:15:43,776 router 6 into this but I think I am. Let's go to router 191 00:15:43,776 --> 00:15:46,646 5 because it'll really only take a minute anyway. 192 00:15:46,646 --> 00:15:55,888 And on router 5 we will do IPv6 router, OSPF 1, redistribute BGP 193 00:15:55,888 --> 00:16:10,803 100, router BGP 100 address family IPv6 unicast neighbor 194 00:16:10,803 --> 00:16:23,149 1.1.1.1 send label and redistribute OSPF 1 match 195 00:16:23,149 --> 00:16:36,195 internal/external NSSA, and if we've done everything right 6 196 00:16:36,195 --> 00:16:43,970 should now have the routes to say 1. So there we go. 197 00:16:43,970 --> 00:16:49,642 The reason I did this is I actually want to show you a 198 00:16:49,642 --> 00:17:00,286 trace from a router. Took it a couple of seconds to 199 00:17:00,286 --> 00:17:02,955 stabilize there, let's do it again just so we get a nice 200 00:17:02,955 --> 00:17:07,693 clean trace, so the main thing I want you to see is as 201 00:17:07,693 --> 00:17:12,231 this passes through it goes to router 5, that's the first 202 00:17:12,231 --> 00:17:16,502 one here of course, router 5 sends it to 3 and it's showing 203 00:17:16,502 --> 00:17:20,072 you those mapped IPv4 addresses, that's fine, but 204 00:17:20,072 --> 00:17:23,276 look what it's showing you for the labels. It gets 205 00:17:23,276 --> 00:17:28,281 a 16 and a 23 then that gets to 23 where it gets a 16 206 00:17:28,281 --> 00:17:34,720 and a 23 again. For those of you familiar with MPLS, if you're 207 00:17:34,720 --> 00:17:36,689 not then this is going to be a little bit beyond the scope 208 00:17:36,689 --> 00:17:39,992 of this course, but for those of you who are familiar with MPLS 209 00:17:39,992 --> 00:17:45,031 already, router 1 is doing PHP or Penultimate 210 00:17:45,031 --> 00:17:47,767 Hop Popping for his directory-connected 211 00:17:47,767 --> 00:17:52,638 loopback which is the next hop, therefore you lose the label 16, 212 00:17:52,638 --> 00:17:56,943 you deliver it to router 1 with just label 23 on it, 213 00:17:56,943 --> 00:18:01,380 and that's how he knows which IPv6 route this goes to and then 214 00:18:01,380 --> 00:18:06,419 he delivers it to the IPv6 destination. So it's pretty 215 00:18:06,419 --> 00:18:08,387 cool it's using label switching through the core. 216 00:18:08,387 --> 00:18:13,059 Now again it's not really exciting in here because of my 217 00:18:13,059 --> 00:18:19,198 labels are 16 and 16, and 18 and 18. You'll see this 218 00:18:19,198 --> 00:18:23,769 a lot if you turn on MPLS after your core routing 219 00:18:23,769 --> 00:18:29,575 protocol is stabilized, because every router has the same routes 220 00:18:29,575 --> 00:18:33,646 in the routing table and MPLS just starts at label 16 and goes 221 00:18:33,646 --> 00:18:39,719 up. So in this case, if I were to say do show ip route, 222 00:18:39,719 --> 00:18:43,889 it's pretty clear-- and I'm on router 6, so I didn't 223 00:18:43,889 --> 00:18:46,592 want that because 6 is barely running IP. 224 00:18:46,592 --> 00:18:58,337 I wanted to look at it on router 5 PE router. Do show ip route. 225 00:18:58,337 --> 00:19:01,440 If you look at this routing table, router 1's loopback is 226 00:19:01,440 --> 00:19:05,511 the first route that it hits. So it gets label 16 and you can 227 00:19:05,511 --> 00:19:11,150 see this. Do show mpls ldp bindings. 228 00:19:11,150 --> 00:19:13,552 This is what's referred to as the LIB or the label 229 00:19:13,552 --> 00:19:16,155 information base and this is all the labels that's 230 00:19:16,155 --> 00:19:18,824 going to be using for next hops. So you can 231 00:19:18,824 --> 00:19:22,628 see it literally. It just starts at one and gives a label of 232 00:19:22,628 --> 00:19:27,900 16, 17, 18, 19. It's really exciting. It just walks the 233 00:19:27,900 --> 00:19:33,205 table. Since everybody pretty much has the same routes, 234 00:19:33,205 --> 00:19:36,442 they're all pretty much going to generate the same labels for the 235 00:19:36,442 --> 00:19:40,413 same destination. So it's a little bit tricky to see 236 00:19:40,413 --> 00:19:44,183 label 16 is swapping with label 16, but label 23 237 00:19:44,183 --> 00:19:53,726 stays end to end, but that is in fact what's happening. 238 00:19:53,726 --> 00:19:55,928 By the way, if we go to say router 1 and see how this all 239 00:19:55,928 --> 00:20:04,737 ends, you can also look at the LFIB which is do show 240 00:20:04,737 --> 00:20:07,873 mpls forwarding table. And you can see he's 241 00:20:07,873 --> 00:20:09,909 got the IPv6 entries, and right there 242 00:20:09,909 --> 00:20:13,813 you can see where he ties that label 23 to the 243 00:20:13,813 --> 00:20:16,615 destination that we trace to. That means that if 244 00:20:16,615 --> 00:20:19,418 we went to the actual loopback, then we should 245 00:20:19,418 --> 00:20:23,022 see label 24. So if we go back to 6, and instead of 246 00:20:23,022 --> 00:20:28,894 tracing to that client network, if we trace to his loopback, 247 00:20:28,894 --> 00:20:31,664 it should be label 24. See. There you go, 24, 248 00:20:31,664 --> 00:20:35,468 24, 24 the whole way through. So as it comes 249 00:20:35,468 --> 00:20:40,639 in on Router 5, Router 5 is imposing or pushing two 250 00:20:40,639 --> 00:20:47,379 labels unto it, 16 and 24. The 16 is to get to the next hop 251 00:20:47,379 --> 00:20:52,384 or Router 1's loopback. 24, is to ultimately tell us 252 00:20:52,384 --> 00:20:56,522 once it gets to Router 1, to actually tell us which route to 253 00:20:56,522 --> 00:21:01,694 follow and out which egress interface and next hop. So 254 00:21:01,694 --> 00:21:06,065 all of this put together all you're really doing is 255 00:21:06,065 --> 00:21:09,869 enabling MPLS in the core, so that's a given, and then the 256 00:21:09,869 --> 00:21:16,375 real trick is two-fold. Well, really it's sort of just 257 00:21:16,375 --> 00:21:18,978 one-fold honestly, two-fold because you're redistributing 258 00:21:18,978 --> 00:21:26,819 two ways but really it's just do show run section router. 259 00:21:26,819 --> 00:21:29,188 The real magic to this whole thing, of course, the really 260 00:21:29,188 --> 00:21:34,994 really big piece, is this one right here, 261 00:21:34,994 --> 00:21:38,230 the send label. If you don't send that label 262 00:21:38,230 --> 00:21:42,134 then he doesn't generate the label, he doesn't send the 263 00:21:42,134 --> 00:21:44,336 label to the other side, and then it's going to 264 00:21:44,336 --> 00:21:47,640 be trying to forward IPv6 packets across IPv4 265 00:21:47,640 --> 00:21:52,545 and that's not going to work. So this is how we do it for an 266 00:21:52,545 --> 00:21:57,817 MPLS-enabled network if we don't want to use VRFs. This 267 00:21:57,817 --> 00:22:02,588 is going to be very very very similar to 6VPE, 268 00:22:02,588 --> 00:22:07,326 the only real difference is we're going to be doing VRFs 269 00:22:07,326 --> 00:22:11,530 rather than, in this case, it's just straight routing table, 270 00:22:11,530 --> 00:22:19,705 straight BGP, no VRFs. So 6VPE is simply bringing VRF to 271 00:22:19,705 --> 00:22:22,341 this equation. So that's what we're going to hit in 272 00:22:22,341 --> 00:22:25,511 our next lesson and I'm going to clean up a 273 00:22:25,511 --> 00:22:27,479 little bit of what we have here before we jump into 274 00:22:27,479 --> 00:22:33,319 that so we'll see 6VPE coming up next. 275 00:22:33,319 --> 00:22:35,321 [music]