WEBVTT 00:00:00.000 --> 00:00:08.820 So now it's time to talk about GLBP, Gateway Load Balancing Protocol. 00:00:08.820 --> 00:00:15.640 So the difference between GLBP and VRP and HSRP is that GLBP allows you to use both gateways 00:00:15.640 --> 00:00:21.360 at the exact same time for a single VLAN, for a single subnet, right? 00:00:21.360 --> 00:00:24.799 And so some people have trouble wrapping their head around this concept. 00:00:24.799 --> 00:00:28.120 Let's just think about HSRP for a second. 00:00:28.120 --> 00:00:33.279 Could my routers both be used for HSRP? 00:00:33.279 --> 00:00:37.560 Yes, but for different groups, right? 00:00:37.560 --> 00:00:44.240 So for HSRP, I could have this as active for VLAN 10 and this would be standby for VLAN 00:00:44.240 --> 00:00:45.240 10. 00:00:45.240 --> 00:00:47.759 So they both can't be active for VLAN 10. 00:00:47.759 --> 00:00:53.000 But this could be active for VLAN 20 and this could be standby for VLAN 20. 00:00:53.000 --> 00:00:55.000 So that is perfectly acceptable. 00:00:55.000 --> 00:00:57.259 That is perfectly fine, right? 00:00:57.259 --> 00:00:59.299 This one's active for 20, that one's active for 10. 00:00:59.299 --> 00:01:00.299 No problem with that. 00:01:00.299 --> 00:01:05.980 Okay, what I'm talking about GLBP is that they're both essentially active. 00:01:05.980 --> 00:01:10.720 They're both forwarders, as we call it, for VLAN 10, okay? 00:01:10.720 --> 00:01:13.459 So the word is not even active now anymore. 00:01:13.459 --> 00:01:15.660 The word's going to be where I'm forwarding for VLAN 10. 00:01:15.660 --> 00:01:19.220 I'm forwarding for VLAN 10. 00:01:19.220 --> 00:01:20.940 So why is this so awesome? 00:01:20.940 --> 00:01:26.500 Because now, for example, if this is VLAN 10, this user's traffic could go this way and 00:01:26.500 --> 00:01:28.580 this user's traffic can go this way. 00:01:28.580 --> 00:01:33.900 But like I mentioned in our overview of first hop redundancy protocols discussion, you're 00:01:33.900 --> 00:01:38.620 only going to be able to do that if you have the right layer 2, layer 3 setup. 00:01:38.620 --> 00:01:41.339 This link here cannot be layer 2. 00:01:41.339 --> 00:01:44.860 It has to be layer 3 between DSW1 and DSW2. 00:01:44.860 --> 00:01:48.739 And then you're going to have to do some filtering here to make sure that you have a port here 00:01:48.739 --> 00:01:52.260 that's in the forwarding state and a port here that's in the forwarding state, again, 00:01:52.260 --> 00:01:53.260 or it's not going to work. 00:01:53.580 --> 00:01:57.139 So you're going to have to manipulate Spanishy protocol, convert that to routed ports. 00:01:57.139 --> 00:02:02.260 You're going to have to do some trunk filtering in order to get GLP to do what you really 00:02:02.260 --> 00:02:06.660 want it to do, and that is have traffic go this way and have traffic go this way. 00:02:06.660 --> 00:02:08.259 So Spanishy protocol gets in the way. 00:02:08.259 --> 00:02:10.779 You have to adjust Spanishy protocol to make it work. 00:02:10.779 --> 00:02:14.820 You won't have to do that on the exam, by the way, adjusting Spanishy protocol, but 00:02:14.820 --> 00:02:16.820 you might have to set up GLBP. 00:02:16.820 --> 00:02:28.580 All right, so what's going to happen now when you set up GLBP? 00:02:28.580 --> 00:02:36.220 You are going to have something called the active virtual gateway, the AVG. 00:02:36.220 --> 00:02:41.220 You'll notice here that only one device can be the active virtual gateway, right? 00:02:41.220 --> 00:02:44.679 The other device is not the active virtual gateway right now. 00:02:44.679 --> 00:02:50.399 So with the active virtual gateway, you're going to have one active active virtual gateway 00:02:50.399 --> 00:02:55.559 and one standby active virtual gateway. 00:02:55.559 --> 00:02:58.119 So what is the active virtual gateway? 00:02:58.119 --> 00:03:05.279 So think of the active virtual gateway as the one responsible for making all the decisions 00:03:05.279 --> 00:03:10.059 about what all the other routers should be doing that are part of the group. 00:03:10.059 --> 00:03:16.699 So it is the overseer of the actual group. 00:03:16.699 --> 00:03:24.500 It is the one that's also going to be responsible for responding and answering those ARP requests. 00:03:24.500 --> 00:03:28.179 It is responsible for handing out MAC addresses. 00:03:28.179 --> 00:03:31.380 Now in this case, there's going to be more than one MAC address. 00:03:31.380 --> 00:03:38.300 There's going to be one IP address for the group, but there's going to be one MAC address 00:03:38.500 --> 00:03:43.380 for each physical router. 00:03:43.380 --> 00:03:48.339 So for example, this physical router gets this MAC address. 00:03:48.339 --> 00:03:52.020 This physical router gets that MAC address. 00:03:52.020 --> 00:03:53.899 Why? 00:03:53.899 --> 00:03:59.600 Because when we build a frame, we need a destination MAC address in it. 00:03:59.600 --> 00:04:08.759 So when PC1 sends out an ARP request, the AVG, let's say, hands out 0101, this one right 00:04:08.759 --> 00:04:10.679 here. 00:04:10.679 --> 00:04:19.440 When PC2 sends out an ARP request, the AVG responds back with 0102, 0102. 00:04:19.440 --> 00:04:21.239 So you see what the AVG is doing here. 00:04:21.239 --> 00:04:27.660 It's doing a load balancing of those MAC addresses. 00:04:27.720 --> 00:04:32.579 So what that means now is PC1, when it builds its frame, is going to be destined to this 00:04:32.579 --> 00:04:40.019 MAC address, and this switch is going to be pointing this way for that MAC address. 00:04:40.019 --> 00:04:45.220 But this PC here gets this destination MAC address for the default gateway, and this 00:04:45.220 --> 00:04:50.179 switch is going to be pointing this way to get to that MAC address. 00:04:50.179 --> 00:04:53.100 And that's why it's important that Spanning Sheet Protocol is not blocking any of those 00:04:53.100 --> 00:04:54.100 ports. 00:04:55.019 --> 00:04:57.899 So that way, we can have this occur. 00:04:57.899 --> 00:05:02.579 So now, because of that, all frames built from PC2 go this way. 00:05:02.579 --> 00:05:04.380 All frames built from PC1 goes this way. 00:05:04.380 --> 00:05:09.739 So let's say you had 250 users in VLAN10. 00:05:09.739 --> 00:05:12.459 That means half of them could have traffic going this way. 00:05:12.459 --> 00:05:16.260 The other half could have traffic going this way, because the AVG is going to hand out 00:05:16.260 --> 00:05:18.899 the MAC addresses round-robin, half of them. 00:05:18.899 --> 00:05:21.140 So I go this and this, this and this, this and this. 00:05:21.140 --> 00:05:23.160 I'll just keep flip-flopping through them. 00:05:23.220 --> 00:05:24.720 So if you ask for the MAC address, I'll give it to you. 00:05:24.720 --> 00:05:25.720 The next one, I'll give the next one. 00:05:25.720 --> 00:05:28.920 And then back to the first, second, first, second, first, second, first, and so on and 00:05:28.920 --> 00:05:29.920 so forth. 00:05:29.920 --> 00:05:30.920 That's all we do. 00:05:30.920 --> 00:05:33.760 Now, there's other ways we can manipulate that, but that is by default essentially what's 00:05:33.760 --> 00:05:36.200 going to happen. 00:05:36.200 --> 00:05:37.200 So what does that mean? 00:05:37.200 --> 00:05:39.720 It means that every other router is an active virtual forwarder. 00:05:39.720 --> 00:05:44.880 An active virtual forwarder, you can have a MACs of four only, four routers, four physical 00:05:44.880 --> 00:05:48.600 routers forwarding, but each of those active virtual forwarders get assigned one of those 00:05:48.600 --> 00:05:51.799 MAC addresses from the actual router. 00:05:51.799 --> 00:05:55.079 So that way, whenever they receive frames destined to that MAC address, they're going 00:05:55.079 --> 00:05:56.679 to be able to forward them and process them. 00:06:03.600 --> 00:06:05.200 So the configuration looks something like this. 00:06:05.200 --> 00:06:09.440 Now, I can't do a demo because my switches, my multilayer switches don't actually support 00:06:09.440 --> 00:06:11.839 GLBP in this case. 00:06:11.839 --> 00:06:16.559 But there are multilayer switches out there that support GLBP and routers support GLBP. 00:06:16.559 --> 00:06:22.079 What am I doing here? 00:06:22.079 --> 00:06:27.160 So let's go to interface VLAN 30 in this example on both devices. 00:06:27.160 --> 00:06:34.880 Notice we typed in GLBP, gave it a group number, and same IP address. 00:06:34.880 --> 00:06:39.399 So this is the IP address that's going to be given to the clients. 00:06:39.399 --> 00:06:43.040 That is a virtual IP address, so one virtual IP. 00:06:43.040 --> 00:06:44.040 So that's what they're going to be getting. 00:06:44.040 --> 00:06:48.119 Your default gateway is 10.0.3.254. 00:06:48.119 --> 00:06:51.799 Your default gateway is 10.0.3.254. 00:06:51.799 --> 00:06:55.839 So now, when they send out, and these MAC addresses are going to be created based on 00:06:55.839 --> 00:06:59.119 the fact that there's two devices that we've configured this on. 00:06:59.119 --> 00:07:02.519 If we also had DSW3 as an example, let's say there was another multilayer switch here that 00:07:02.519 --> 00:07:06.119 was part of this and we had three default gateways, well then, when we configured on 00:07:06.119 --> 00:07:11.239 that one, well then, the AUVG would know that and the AUVG would create another MAC address 00:07:11.239 --> 00:07:12.239 for it. 00:07:12.239 --> 00:07:21.760 So basically, the MAC addresses are going to be 0.0.0.7.B4.0.0.0.1.0.1.1.0.2.1.0.3.1.0.4. 00:07:21.760 --> 00:07:22.760 That's it. 00:07:22.760 --> 00:07:24.140 That's how it works for GLBP. 00:07:24.140 --> 00:07:39.320 That's the four different MAC addresses. 00:07:39.320 --> 00:07:46.839 So now, we are going to have – notice the priority here? 00:07:46.839 --> 00:07:51.880 So the priority is automatically 100, but I included in the config down here. 00:07:51.880 --> 00:07:53.600 The priority is used for what? 00:07:53.600 --> 00:07:58.399 The priority is used for the AUVG election process. 00:07:58.399 --> 00:08:00.320 That's what the priority is used for. 00:08:00.320 --> 00:08:02.200 It's not for the forwarders. 00:08:02.200 --> 00:08:05.000 It's for the AUVG election process. 00:08:05.000 --> 00:08:09.799 Likewise, the timers are also for what? 00:08:09.799 --> 00:08:10.799 AUVG. 00:08:10.799 --> 00:08:20.079 So basically, I am the standby AUVG and you are the active AUVG, and so hello, PAC. 00:08:20.079 --> 00:08:22.160 It's back and forth between us. 00:08:22.160 --> 00:08:27.640 So that way there, I know you're the active AUVG and I'm the standby AUVG. 00:08:27.640 --> 00:08:31.839 So essentially, right, the commands we see here and preemption is disabled by default, 00:08:31.839 --> 00:08:34.280 so let's turn on preemption. 00:08:34.280 --> 00:08:43.000 So now, those first four commands are essentially the initial setup for the whole AUVG process, 00:08:43.000 --> 00:08:47.799 and so once the active AUVG is determined, then everybody else are just active virtual 00:08:47.799 --> 00:08:53.239 forwarders based on the number of distribution gateways you have, then that's the number 00:08:53.239 --> 00:08:59.719 of MAC addresses the AUVG is going to create, and then as users start asking, hey, can I 00:08:59.719 --> 00:09:04.280 have the MAC address for 10.0.3.254, well, then the AUVG hands out the first one to 00:09:04.280 --> 00:09:07.159 this one, the second one to this one, the third one to this one, or sorry, the first 00:09:07.159 --> 00:09:12.520 one to the third, and the second to the fourth, the first to the fifth, the second to the 00:09:12.520 --> 00:09:15.520 sixth, and so on and so forth. 00:09:15.520 --> 00:09:19.960 And then you'll see at the bottom here that we have the opportunity to control. 00:09:19.960 --> 00:09:25.640 Round Robin is typically the default, which is what I just described to you, but if we 00:09:25.640 --> 00:09:35.960 want to turn around and we want to force, for example, that the same MAC address is 00:09:35.960 --> 00:09:44.239 always given to a certain host, then we can use host-dependent instead, meaning that once 00:09:44.239 --> 00:09:48.840 I give PC1 that MAC address, they're always going to use that MAC address going forward. 00:09:48.840 --> 00:09:54.599 Once I give PC2, so if they ask for it again, then I'll give it to them again. 00:09:54.599 --> 00:09:57.880 And that way, they're the same MAC is always used, and then those users always use the 00:09:57.880 --> 00:10:01.960 same default gateway over and over and over and over again. 00:10:01.960 --> 00:10:03.679 And then we have weighted as well. 00:10:03.679 --> 00:10:09.799 What if this default gateway, we had greater bandwidth than this default gateway? 00:10:09.799 --> 00:10:14.799 Then I could weight that so that 75% of the traffic goes through here, and maybe 25% of 00:10:14.799 --> 00:10:17.159 the traffic goes through here. 00:10:17.159 --> 00:10:20.039 We could do that. 00:10:20.039 --> 00:10:21.039 And how would you do that? 00:10:21.039 --> 00:10:26.320 Well, if you set the weighting now, the weighted option instead of doing Round Robin or host-dependent, 00:10:26.320 --> 00:10:31.119 then that means I'll hand out one MAC address, so the MAC address that is associated with 00:10:31.119 --> 00:10:39.039 this active virtual forward, I'll send it out more often than the one down here. 00:10:39.039 --> 00:10:43.039 So let's say I wanted 75% of the traffic to go through here, 25% to go through here, well 00:10:43.039 --> 00:10:53.919 then for every 75 I hand out of this, I'm only going to hand this one out 25 times. 00:10:53.919 --> 00:10:57.000 Approximately. 00:10:57.000 --> 00:11:01.400 Now also something to keep in mind is what happens if we have a failure? 00:11:01.400 --> 00:11:07.000 So if we have a failure here of the actual device, well then this device is going to 00:11:07.000 --> 00:11:10.520 know about it because hello packets cease to exist. 00:11:10.520 --> 00:11:11.520 So what happens? 00:11:11.520 --> 00:11:16.280 This one remains the AVG, but now it's also the AVF for that other MAC address. 00:11:16.280 --> 00:11:18.840 So it would take over that MAC address. 00:11:18.840 --> 00:11:23.679 So now it's two AVFs and one AVG. 00:11:23.679 --> 00:11:29.359 And then when this comes back online, then we've got, again, normalcy here. 00:11:29.359 --> 00:11:34.760 So what happens if this device fails, then I become the AVG and I become the active forwarder 00:11:34.760 --> 00:11:36.799 for all of this. 00:11:36.799 --> 00:11:39.580 And I'm also the active forwarder for myself. 00:11:39.640 --> 00:11:44.460 So whenever a failure happens to the device, depending on who fails, if the AVG doesn't 00:11:44.460 --> 00:11:45.460 fail, no big deal. 00:11:45.460 --> 00:11:49.780 It stays the AVG. 00:11:49.780 --> 00:11:55.880 But anytime an AVF fails, then that AVF, everything it was doing has to be taken on by some other 00:11:55.880 --> 00:12:02.099 device in that environment, in that group, I should say. 00:12:02.099 --> 00:12:04.820 And again, just like with HSRP or VRP, you would have to create a different group for 00:12:04.820 --> 00:12:18.619 each and every single subnet size, VLAN size, broadcast domain you have. 00:12:18.619 --> 00:12:21.299 Does the AVG always have to be an AVF too? 00:12:21.299 --> 00:12:25.940 Yeah, it automatically makes itself an AVF, automatically. 00:12:25.940 --> 00:12:28.700 And you would want that, because you don't want to waste it, right? 00:12:28.700 --> 00:12:34.059 You don't want to waste it, because your whole point is to use all of the routers in the 00:12:34.299 --> 00:12:36.700 group at the same time for that same subnet. 00:12:36.700 --> 00:12:40.380 So yeah, the AVG will be an AVF too. 00:12:40.380 --> 00:12:42.380 And that's good. 00:12:42.380 --> 00:12:47.140 So folks, we are done. 00:12:47.140 --> 00:12:53.419 Our coverage of first operandency protocols. 00:12:53.419 --> 00:13:02.580 What can I tell you to kind of let you know how important it is to ensure that you are 00:13:02.580 --> 00:13:06.419 very familiar with all three of these? 00:13:06.419 --> 00:13:15.580 Not just from a theoretical standpoint, but from a configuration standpoint as well. 00:13:15.580 --> 00:13:24.460 I have taken many Cisco exams over the years. 00:13:24.460 --> 00:13:36.099 At the CCNP level, every time I have taken a Cisco exam that had in the exam objectives, 00:13:36.099 --> 00:13:45.460 something like HSRP, VRRP, or GLBP, you've always had to know more than the theory of 00:13:45.460 --> 00:13:50.979 those options, meaning can you configure it? 00:13:50.979 --> 00:13:52.780 Can you verify? 00:13:52.780 --> 00:13:55.500 Can you troubleshoot? 00:13:55.500 --> 00:14:10.059 So it is exceptionally important that you don't under-prepare for these first operandency 00:14:10.059 --> 00:14:12.299 protocols for the exam. 00:14:12.299 --> 00:14:13.299 Let's just put it this way. 00:14:13.299 --> 00:14:26.979 The probability of you maybe having to hype stuff in regards to first operandency protocols 00:14:26.979 --> 00:14:31.900 is extremely high in your future. 00:14:31.900 --> 00:14:34.979 As a result of that, make sure you are ready for that.