Cisco CCNA 200-301 – EtherChannel Part 2

4. EtherChannel Protocols and Configuration

In this lecture, you’ll learn about the different Ether Channel protocols and how to configure and verify them. So there’s three available protocols that’s LACP and PAGP and Static. So first one is LACP, the link aggregation control protocol. This is an open standard, so it’s supported on all vendors switches. With LACP, the switches on both sides negotiate the port channel creation and maintenance. And of the three available methods, this is the preferred one to use.

Next one we have is PAGP, which is Cisco’s proprietary port aggregation protocol. This works similarly to LACP, where the switches on both sides again negotiate the port channel creation and maintenance, but it’s not recommended to use PAGP because it’s proprietary. And the third option we have is to configure a static Ether channel. With static, the switches do not negotiate creation and maintenance, but the settings must still match on both sides for the port channel to come up.

So with all three, you’ve got two switches that have got links going between them, and you’re going to configure those into a port channel. You need to configure the same settings on both sides on both switches. And for static, this is the one that we’ll use. If LACP is not supported on both sides, LACP will be supported on all Cisco switches, but maybe you’re connecting to another vendor switch or a server or something like that. So those are the three different options. The configuration for them all is really, really similar. They all actually use the same command, which is the channel group command, but it’s on the keywords that we use along with channel group that decides which of the three protocols is going to be used for the parameters for Ether channels. I touched on this just a minute ago where I said that the configuration has to match on both sides of the link.

The interfaces need to have a matching configuration. So settings that have to be the same on both sides include the speed and duplex, whether the port is set to access or trunk mode, the native VLAN, the allowed VLANs on there, and the access VLAN if it’s an access part. So looking at the configuration, the first one we’ll look at is LACP. Your LACP interfaces can be set as either active or passive. So let’s say we’re going to configure a port channel between switch one and switch two. If switch ones interfaces are set as active and switch twos as passive, the port channel will come up. But if both sides are set as passive, the port channel will not come up. And if both sides are active, the port channel will come up. So it’s recommended to configure bulb sides as active. Your choices are either both sides is active or one side is active and one side is passive. It’s easiest just to configure Bob is active because then you don’t need to worry about which is the active side and which is the passive. Side. So here’s our configuration. It’s configured at the interface level. So specify the interfaces you want to group into the port channel. Here we’re seeing interface range fast zero slash 23 to 24.

And then our command is channel. Group One mode active. As soon as you enter that command it will create a new logical interface which is your port channel interface. Here. We’re Configuring Channel Group One. So that would create part channel one. The reason that there’s a number here is that you might have different port channels going to different neighbor switches. So say that we are switch one, we’ve got a port channel going to switch two. We could make that port channel one. If we’ve also got a port channel going to switch three we could make that port channel two. So that’s the command to create your port channel. After the port channel has been created, most of your interface settings are set at the port channel level.

So to do that we say interface port channel one and then switch port mode trunk. We would also set the native VLAN, the allowed VLANs, et cetera at this level. And you need to configure those matching settings on the switch on the other side as well. So you see that the configuration here is exactly the same. We did it first on switch one and then we do matching configuration on switch two. So that was LACP. So with LACP, it’s Channel Group command. And then you use either active or passive. The configuration for PAGP is the same but rather than using Active or Passive we use Desirable or Auto. And it’s similar rules that we had with LECP where if one side is desirable and the other is Auto then the port channel will come up. If both sides are Auto, it won’t, and if both sides are desirable it will. So again, if you are going to use PHP, which is not recommended, then set both sides as Desirable then you don’t need to worry about which is desirable and which is Auto.

So looking at the config exactly the same as it was for LACP. But here we say channel Group One mode desirable rather than Channel Group One mode active. So you can tell by the keyword at the end where whether it’s going to be LACP or PAGP or when we get to the next slide, static for static. Again. It’s exactly the same configuration. But we say channel group one mode on. So active for LECP, desirable for PHDP and on for static. Rest of the configuration exactly the same for all of them. So that’s how we configure it, how we verify it really. There is one Swiss Army knife command for checking Ether Channel. Which is show ether Channel Summary. And you see the flags just under there so it tells you what all the letters actually mean. Here, looking down at the bottom we can see that this is Group One. Our first port channel. The port channel interface is port channel one and in brackets I see a capital S and a capital U. The capital S means that it’s a layer two part channel.

You can also configure a layer three part channel as well. That’s where on the interface you would say no switch port and put an IP address on there. But more commonly it’s going to be a layer two port channel. The capital U means it’s in use which basically means that it is up. I can see that the protocol is LACP and over on the right I can see that my ports are fast 00:23 and faster 00:24. A capital P means that they are in the port channel for a layer two port channel. If you see any letters other than exactly what you see here, there’s a problem with the port channel and it’s not going to come up. If there is a problem with your port channel, by far the most common issue is that your settings do not match on both sides. So look at the interface both at the physical interface level and also at the port channel interface level as well. And just make sure that the settings are exactly the same on both sides. Also make sure that you’ve selected the correct interfaces that are tabled to each other as well. Okay, so that was our show. Ether Channel summary command.

You can also do a show Ether Channel enter that will give you more verbose output but Summary really tells you everything that you need to know. The last command to look at is shows spanning 300 VLAN. Because as we explained before, the reason for using Ether Channel is to avoid spanning tree shutting down some of your links. So after you’ve configured this you want to check that spanning three is working as you would like. So the example here is before we’ve configured Ether Channel, if you look at the picture over and right we’ve got a switch access three with uplinks going to our switch CD one, the core distribution one. CD one is the root bridge for spanning three. And if we look at axis three it’s got two links going up to CD one. If we don’t put those in a port channel then spanning three sees them as a potential loop and it’s going to block on one of the links.

Looking at the output of our show spanning three VLAN one command here, if I look down on the bottom I can see that Fast 00:23 is forwarding, it’s the root port on axis three and Fast 00:24 is blocking to prevent the loop. So these are fast ethernet interfaces. In our example I’m only getting 100 meg of uplink bandwidth rather than 200 meg with my two ports. So to fix that we configure Ether Channel. After you have configured Ether Channel we can put the same command in again show spanning three VLAN one. And now if I go back a slide before the port channel, you see that spanning three. Saw the physical interfaces fast zero slash 23 and fast 00:24.

When we put those into a port channel spanning three just sees that one logical interface. So it sees it as being one link, which is not a potential loop as far as spanning three is concerned. So now on the port channel, it is forwarding. I don’t have any ports that are blocking I get the full bandwidth of both physical interfaces now. Okay, so that was our configuration and verification. See you in the next lecture.

5. EtherChannel Lab Demo

In this section you’ll see how to configure port channels ether channel with a lab demo. So I’m using the same lab topology that I used in the spanning three section. So we’ve got our two core distribution layer switches, CD One and CD two, and we’ve got a couple of access layer switches, access three and Access S four. I’ve already set up the VLANs and the trunks. I’ve configured CD One as the spanning three root bridge primary and CD Two as the spanning three root bridge secondary. So if you have a look at the diagram here, you can see from both AC three and access layer switch four that they’ve got four up links, but there’s only actually one of them which is active and forwarding traffic on both of them. It’s an interface which is facing CD one because CD One is the root bridge. So I’ve got four up links there, but I’m only able to use the bandwidth from one, so I want to improve on that.

So I’m going to create a port channel from Access Free going up to CD one. I’m also going to create a different port channel from Access three going up to CD two. It’s going a different switch, so it’s going to be a different port channel. And then from axis four I’ll configure its port channels. One going to CD one and one going to CD two. The first port channel I’ll do from axis three to CD one and I’ll use LECP for that. So it’s going to be on interfaces on axis three, Fast 00:23 and 24, and it’s the same on the CD one side as well. So I’m going to be configuring them with the same matching configuration. I’ll then do the port channel from axis three to CD two and I’ll use PAGP for this one. Now, real world, hopefully it’s obvious that would be an insane thing to do. You would always standardize on one protocol for your port channels and it’s recommended to use LACP.

You can use static if one side doesn’t support LACP. But for the lab demo, I want to show you all the different protocols. So I’ll do the first one with LACP, I’ll do the second one with PAGP. The third port channel will go from axis four up to CD two on the same side. On the right hand side of a topology, I’ll use static for that. So you’ll see all three and we’ve got one left over. So I’ll use LACP for that again. Okay, so that’s what I’m going to configure. So let’s start off on our switch, axis three. I’ll do the first port channel, so that’s going to be LACP up to CD one on interfaces Fast 00:23 and 24. So I’ll go on to the Access three switch if I can find it in here. There it is. And actually, let’s do a show spanning three first as well so we can check this. So I’ll just show spanning three and my PCs are in V 110. So let’s do a show spanning three V 110 and I can see that I am forwarding on fast zero one actually facing down towards the PC. Let’s do a show run and check the configured interface fast zero one.

Yeah, that all looks good. So it’s in view ant switchboard mode axis facing down to the PC. So before up, links are fast 00:21 to 24 and only one of them is forwarding. Now that’s a part that is connected to CD one, which is my root bridge. So let’s put this into an ether channel. So I’ll go to global config interface range fast zero slash 23 to 24 is what’s connected to CD one. And to put these in a port channel, it’s channel group one mode and then I want it to be LACP so I’ll make the mode active and as soon as I do that it creates the port channel interface for me. So if I now do a do show IP interface brief, I can see up at the top there I’ve now got this new port channel interface and I need to configure my interface level settings on the port channel. So to do that I’ll go to interface port.

You can just do PO is shorthand interface PO one for port channel one. And then on here I want to say switchport mode trunk because it’s connecting to another switch and switchport trunk. Native VLAN 199 is the native VLAN we’re using. I would also set the allowed VLANs on here normally as well in a production environment. Okay, so that’s it done on the axis three side. Let’s do it on the other side of the links, which is on CD one. I’m going to put in the exact same configuration here. So configt and it’s interface range fast 00:23 to 24.

Now the interfaces on both sides don’t have to be the same number interface. It could have been interface fast zero nine and ten on this side and 23 and 24 on the other side and that will work just fine. But it’s best and easiest if possible. You can use the same interface number on both sides. It just makes it easier for us as human administrators to understand what’s going on. Okay, so I’ve specified interfaces and then channel grew one mode active. I could say passive here, but it’s best practice to do active on both sides and then you can’t get them mixed up. I see the port channel comes up. I’m getting a native VLAN mismatch command because I haven’t configured it yet. So let’s do that now. So I’ll go to interface port channel one, switch port mode trunk and switchport trunk native VLAN 199 and I see it unblocks it because I’ve got the same native VLAN on both sides now. So that looks good. So now the next thing to do is check that the port channel came up okay, so I’ll do a show Ether Channel Summary and this all looks good. That’s what I wanted to see for port channel one, su which means it’s a layer two interface and the U is for in use which basically means it’s up and it’s portsfast 00:23 and 24 and they are both in the port channel. So that all looks good. So that was my LECP configuration.

Next let’s do PAGP from access three, going up to switch CD two. Get on my command line here and this is going to be interface range fast 00:21 to 22 is the interfaces that are connected over to the CD two switch on the right hand side and I’ll say channel group. Now I’ve already used one so I’ll use two for the next one and I’ll say Mode desirable to make it PHP. I see the port channel is created and then I go onto the interface port channel two. Now switchport mode trunk and switchport trunk native VLAN 199. I need to make sure I configure matching settings on the other side which is over on CD two. So config T and it’s interface fast 00:21 to 22. You can hopefully see the benefit of using the same interfaces on both sides. Now I’m not really having to think about it. I know that it is the same on both sides and I just forgot to put in the range keyword there.

So I specify a range and I’m using the same on both sides. So it’s nice and logical. It saves me getting confused and mixing things up. And channel group now I haven’t created a port channel one here yet, but because I use two on the other side, I’m going to use two on this side as well. Again to keep things logical. So channel Group two mode desirable. This is also a PAGP. Then I need to go to interface port channel two this time and switch port mode trunk switchport trunk native VLAN okay, so that is my first two done. Let’s check. It came up okay, so show Ether Channel Summary and this one is up as well. It’s all good too. So it’s first two done. Next up I’ll go on to access four and I’ll configure report channel going up the same right hand side of the topology.

So I go to Global Config and it’s interface range fast 00:23 to 24 on the right hand side and channel grew one mode. I’ll make this static this time so it’s going to be on and that is my static configured. I can now go interface port channel one, switchport mode trunk and switchport trunk native VLAN 199. I need to configure the other side of this link which is on CD two. So I’ll go to Global Config, it’s Interface range fast 00:23 to 24 and then channel Group one mode on and I’m seeing I’m getting an error message here that fast 23 is not compatible because of DTP mode.

Well, if I have a look at my interfaces. So I’ll do a show run and I can see on interface fast zero slash 23 and 24 that the settings are the same on there and they’re the same on the other side as well. So this is just a glitch with the packet tracer. The Ether channel should come up in a second anyway. So let’s just complete the configuration. So I’ll go configte and then interface port channel one, switchport mode trunk and switchport trunk native VLAN one nine nine. And I should get rid of that error message.

Yeah, that looks good. And then if I do a show Ether channel summary and yeah, I can see that it did come up there. Sometimes when you first configure this in packet tracer, when you do the static, it might throw that error message at you, but just complete the configuration anyway. You might have to give it a minute and then the port channel should come up. So there’s the port channel I created earlier, the PAGP one, which is going to axis three. And this is the static that I just configured going down to access four. You can see where it shows a dash of the protocol. That means that it is static. Okay, so that’s my first three port channel is configured. I’ve verified that they’re up and working. I’ll do the fourth one, which is going to be another LACP as well. You don’t need to see me doing that because you already saw me configure it. So I’ll see you back here for the next lecture where we’ll have a look at how this is going to affect spanning tree.