Welcome to Jeremy’s IT Lab. This is a free, complete course for the CCNA. If you like these videos, please subscribe to follow along with the series.
Also, please like and leave a comment, and share the video to help spread this free series of videos. Thanks for your help. Also, remember to download and use the Anki flashcards with the link in the description, which will help you very much in remembering what you've learned.
Let’s get started. In this video we’ll take a break from IP addresses and take a look at switch interfaces. When we configured IP addresses, I talked about router interfaces a little bit.
For example, we used the ‘show ip interface brief’ command to check the Layer 1 and Layer 2 status of the interfaces. I also talked about how router interfaces are disabled with the ‘shutdown’ command by default. This time we’ll take a look at switch interfaces and see how they are different from router interfaces, and also how they are the same.
Specifically, we’re going to talk about configuring Layer 1 characteristics of these interfaces, such as speed and duplex. So let me introduce what we’ll cover in this video. I’ll talk about interface speed and duplex.
Speed refers to the data rate in bits per second, such as 10, 100, or 1000 megabits per second. Duplex is something I mentioned briefly in a previous video, basically it refers to whether a device is able to both send AND receive data at the same time. I’ll talk about speed and duplex autonegotiation, which allows two devices to negotiate speed and duplex settings without us having to manually configure them.
I’ll talk about interface status. I already talked about this in the previous video on configuring router interfaces, but I’ll expand upon the topic. Finally, I’ll talk about interface counters and errors.
Cisco devices keep various counters regarding the traffic that passes through them, such as how many errors have occurred, etc. I’ll show you how to find and analyze these statistics, and what kinds of errors there are. Also let me review one point about switches that I mentioned in the very first video in this series.
Up top is a photo of a Cisco ASR 1000-X router, and under it is a Cisco Catalyst 9200 switch. What’s the big difference that stands out here? How about the number of interfaces?
The router has 8 SFP interfaces for fiber-optic cables, and then a few RJ45 interfaces for the console port and such. The switch, on the other hand, has 4 SFP interfaces, plus FORTY EIGHT RJ45 interfaces. This is because switches are used to connect end hosts to.
This catalyst switch might have 48 PCs connected to the RJ45 ports, and then connect to a router with the SFP fiber optic ports. Next let me introduce the network topology I’ll use for this video. This is a single LAN, 192.
168. 1. 0/24, with one router, R1, two switches, SW1 and SW2, and four PCs, PC1, 2, 3, and 4.
In a real network with two switches there would be many more end hosts, but just for this demonstration I’ve attached just these four PCs. SW1 is the device we’re going to focus on today, configuring its network interfaces, including F0/1, F0/2, F0/3, and F0/4 which are connected, as well as the remaining interfaces which are not connected at the moment. Let’s go right into the CLI of SW1.
First I use ‘enable’ to enter privileged exec mode. After entering privileged exec mode, I entered the ‘show ip interface brief’, using shortcuts. As you can see, the four interfaces which are connected to devices, fastethernet 0/1, 0/2, 0/3, and 0/4 have a status column, which is the Layer 1 status, and a protocol column, which is the Layer 2 status, of UP and UP.
We usually call this up/up. Keep in mind I haven’t done any configuration on SW1 yet except set the hostname. So already we can see a difference between Cisco routers and switches.
What is that? Well, I mentioned briefly in the Day 8 video, router interfaces are in an administratively disabled state by default, meaning they have the ‘shutdown’ command applied. Switch interfaces, however, are different.
They don’t have the shutdown command applied, so if you connect them to another device they’ll usually be in the up/up state with no configuration required. Now, the IP address is unassigned, and it will remain that way because these are layer 2 switchports, they don't need an IP address. The concept of multilayer switching, where you actually DO assign IP addresses to switches, will be for a future lesson.
For this lesson, don’t worry about the IP address column. Now take a look at the other interfaces, keep in mind I’m ignoring the ‘VLAN1’ virtual interface, that will be a topic for another video. These other interfaces aren’t connected to any other devices, so their status is down/down.
Keep in mind, down and down is different than administratively down and down. Here’s the show ip interface brief command on a router, notice administratively down and down, this is because of the ‘shutdown’ command’. ‘DOWN/DOWN’ doesn’t mean the interfaces are shutdown, it means they aren’t connected to another device.
So, to summarize, router interfaces have the shutdown command applied by default, so they will be in the administratively down/down state by default. Switch interfaces do not have the shutdown command applied by default, so they will either be in the up/up state, if they are connected to another device, or the down/down state, if not connected. Now let’s look at another useful command to check on switch interfaces.
That is ‘show interfaces status’. Let’s look at each field of the output. First, the port field simply lists each interface.
Next, you might be wondering what this ‘name’ field is. Well, its the description of the interface. I’m not sure why they called it ‘name’ and not ‘description’, but we’ll configure descriptions on these interfaces and you’ll see that it appears here.
Next is the status field. As you can see this is different than the status field of show ip interface brief. The four connected interfaces show a status of ‘connected, and the unconnected interfaces show a status of ‘notconnect’.
There are other possible statuses, we’ll cover them as they come up in the course. Next is the VLAN field. VLANs will get their own video, basically they can be used to divide LANs into smaller LANs.
As you can see, the default VLAN is 1. The only one that does not show a VLAN of 1 is F0/2, which shows trunk. I won’t explain trunks now, they’ll be covered in the VLAN video.
Just take note of the fact that the interface connected to the other switch, SW2, is a trunk interface. Next is the duplex field, which as I mentioned before indicates whether the device is capable of both sending and receiving data at the same time, which is known as full-duplex, or if its not, which is called half-duplex. Duplex is auto by default on Cisco switches, meaning it will negotiate with the neighboring device and use full-duplex if possible.
Notice that all of the unconnected interfaces have a duplex of auto, and the connected interfaces have a duplex of ‘a dash full’ Well, the ‘a’ stands for auto, and it means that it automatically negotiated a duplex of auto with the neighboring device. Next is the speed field, which is also auto by default. These are fastethernet interfaces, so they are capable of speeds up to 100 megabits per second.
However, they are also capable of operating at 10 megabits per second. Auto means they are able to negotiate with the device they are connected to and use the fastest speed both devices are capable of. This time, we see ‘a DASH 100’, meaning a speed of 100 megabits per second was auto-negotiated with the neighboring device.
I’ll talk in more detail about auto negotiation soon. Okay, finally is the Type field. These are all RJ45 interfaces for copper UTP cables, but if they were small form-factor pluggable, or SFP, modules, you’d see that here instead.
In this case, we see 10/100BASE-TX, the 10/100 of course referring to the speeds at which these interfaces can operate. Autonegotiation works well so usually you’ll leave it be, but lets go and manually configure the speed and duplex of an interface, F0/1 which is connected to R1. As you can see, I enter interface config mode, and then use the command speed.
I use the context-sensitive help to display the next option, and I can choose between 10 or 100 megabits per second, or auto. So, I set the speed to 100, since R1’s interface is a fastethernet interface also. Then, I use the command ‘duplex’ to set the duplex of the interface.
As the context-sensitive help shows, I can choose between auto, which is the default, or full and half. Both SW1 and R1 support full duplex, so I choose that. Finally, I configure a description on the interface, indicating that it is connected to R1.
Okay, so when I use the ‘show interfaces status’ command once again, you can see both the configured description and the configured speed and duplex. Notice the duplex is full, rather than a dash full, and the speed is 100, rather than a dash 100, because they are not being autonegotiated any more. Once again, normally you will keep autonegotiation on, but if perhaps there is some problem and its not working, you should know how to manually configure the speed and duplex of an interface.
Now, let me quickly do some configurations for the other connected interfaces off-camera… Okay, so I just configured descriptions on F0/2, 0/3, and 0/4. Now, how about the unused interfaces? Although the fact that switch interfaces are enabled by default is convenient, as you can just plug a device in and use it straight away, it can be a security concern.
Really, you should disable the interfaces. Fortunately, instead of having to configure each of the 8 interfaces one by one, there is a way to configure all 8 interfaces at once. Here’s the command that can save you a lot of time.
From global config mode, type ‘interface range’, and then enter the range like this, f0/5 to 12. This time, instead of interface config mode, I am brought to interface range config mode, and I enter a description, and then shut down the interfaces. All at once, I get messages saying the interfaces have changed to administratively down.
Now, before we check show interfaces status once more, let me show you a cool thing about this command. The interfaces in the range don’t all have to be consecutive. For example, if I later want to enter fastethernet 0/5, 6, 9, 10, 11, and 12, but leave 7 and 8 shutdown, I can do this: interface range f0/5 to 6, COMMA, f0/9 to 12.
Then if I issue ‘no shutdown’, you can see only those interfaces and enabled. F0/7 and 8 remain down. Okay, so I shutdown interfaces f0/5,6,9,10,11, and 12 again, and here’s the output of show interfaces status again.
Now you can see the description of each interface, and the status is disabled. Once again, this status is different than the status shown in ‘show ip interface brief’, which will now be ‘administratively down’, although really they mean the same thing. Before talking about autonegotiation, I want to explain full and half duplex.
I have briefly mentioned it a few times, but haven’t explained it in depth yet. Half duplex means that the device cannot send and receive data at the same time. If it is receiving a frame, it must wait before sending a frame.
Full duplex means that the device CAN send and receive data at the same time. It does not have to wait. Clearly, full duplex is the preferred way to go.
In modern networks that use switches, all devices can use full duplex on their interfaces. So, you may be wondering where is half-duplex used? Well, in modern day networks, almost nowhere.
But to understand half-duplex let me introduce you to an old type of network device which was around before the network switch. The hub is much simpler than a switch, in fact it is simply a repeater. Any frame it receives, it floods like a switch does with a broadcast or unknown unicast frame.
For example, if PC1 wants to send a frame to PC2, it will send the frame out of its network interface, and after the hub receives it, it will repeat the frame out of its other interfaces, to PC2 and PC3. PC3 will recognize that the destination MAC Address is not its own and ignore the frame, and PC2 will receive it normally. Now, what if two PCs try to send frames at the same time?
In this case, PC1 is trying to send a frame to PC2, and PC3 is trying to send a frame to PC1. They both send the frames out of their network interfaces, and this is where a problem occurs. The hub won’t send one first and then send the other after, it simply tries to flood both at the same time, and this will result in a collision on the interface, and PC2 won’t receive either frame intact.
All devices connected to a hub are part of what’s called a collision domain. The frames they send could collide with frames any of the other devices connected to the hub send. To deal with collisions in a half-duplex situation like this, Ethernet devices use a mechanism called ‘CSMA/CD’.
Let’s check it out. CSMA/CD stands for ‘carrier sense multiple access with collision detection’. It describes how devices avoid collisions in a half-duplex situation, and how they react if collisions do occur.
It works like this. Before sending frames, devices listen to the collision domain until they detect that other devices are not sending frames. If a collision does occur, which can still happen because of bad timing and such, the device sends a jamming signal to inform the other devices that a collision happened.
Each device then waits a random period of time before sending frames agan. The process then repeats, with each device listening to check if other devices are sending frames before sending their own frames. Now, that process works, and it was how networks operated for a long time.
But switches are more sophisticated than hubs. Hubs are simple repeaters which operate at layer 1, repeating whatever signals they receive. Switches operate at layer 2, using layer 2 addressing, MAC addresses, to send frames to specific hosts.
They also won’t try to send two frames to the same host at once. So, this network which was one collision domain when connected to a hub, is now 1, 2, 3 collision domains. Because of the improved functionality of switches over hubs, these devices can now operate in full duplex, meaning they don’t have to worry about whether or not other devices are sending data at the same time, they can send data freely.
Although problems like collisions still do occur, they are rare and usually are a sign of a problem, like a misconfiguration, rather than a regular occurrence like in a half-duplex network. So let’s review half and full duplex once more. In half duplex, the device cannot send and receive data at the same time.
If it is receiving a frame, it must wait before sending a frame. Devices attached to a hub must operate in half duplex. In modern networks, you’re probably not going to use a hub, but it’s a possibility.
In full duplex, the device can send and receive data at the same time, it does not have to wait. Devices attached to a switch can operate in full duplex. Okay, now that you know what duplex is let’s talk about speed and duplex autonegotiation on interfaces, and this applies to both routers and switches by the way.
Interfaces that can run at different speeds, for example 10/100 or 10/100/1000, have default settings of speed auto and duplex auto. Interfaces advertise their capabilities to their neighbors, and they negotiate the best speed and duplex settings they are both capable of. So, let’s look at an example.
Here’s a small network, a switch with three PCs connected. Connected to G0/1 is a PC with a regular ethernet interface. Connected to G0/2 is a PC with a fastethernet interface.
and connected to G0/3 is a PC with a gigabit ethernet interface. As you probably know by now, an Ethernet interface can run at 10 megabits per second, a fastethernet interface can run at 10 or 100 megabits per second, and a gigabit ethernet interface can run at 10, 100, or 1000 gigabits per second. So, G0/1 and the PC will negotiate to a speed of 10 megabits per second, and full duplex.
G0/2 and the PC will negotiate to a speed of 100 and full duplex. And g0/3 and the PC will negotiate to speed 1000 and full duplex. The PCs are all able to use the max speed of their network interfaces, and the switch adjusts the speeds of its interfaces to match.
In a network like this with all PCs and switches, there’s no reason to use half-duplex, so they all negotiate to use full-duplex. Now, let’s talk about another situation. What if autonegotiation is disabled on the device connected to the switch?
So, the switch is trying to autonegotiate, but the other devices don't respond. Well, this is how the switch will respond. For the speed, the switch will try to sense the speed that the other device is operating at.
If it fails to sense the speed, it will use the slowest supported speed, for example 10 Mbps on a 10/100/1000 interface. Now, as for duplex, if it ends up using a speed of 10 or 100 mbps, the switch will use half duplex. If the speed is 1000 mbps or greater, it will use full duplex.
So let’s see how this works. In this case, only the switch is using autonegotiation, and the three PCs have manual, fixed speed and duplex settings. We’ll also assume that the switch successfully detects the speed that the PCs are using.
The Green PC has a speed of 10 Mbps, so the switch detects that and sets its speed to the same. Because its speed is 10 Mbps, it sets its speed to half duplex. Now, how about G0/2, connected to the Red PC?
It senses that the PC is using a speed of 1000 mbps, so it uses the same. Because the speed is 1000 Mbps, it uses full duplex. Now how about G0/3, connected to the blue PC?
It senses the speed of 100 mbps, but then what about the duplex? The PC is using full duplex, but without autonegotiation the switch cant sense that. So, because the speed is 100 Mbps, the switch uses half duplex.
This results in a duplex mismatch, which will cause collisions to occur, resulting in poor network performance. So, really you should be using autonegotiation on all devices in the network. Now let’s take a look at some of the errors that can show up on interfaces that otherwise seem to be working.
The switch will take count of some of these things and you can view them with the ‘show interfaces’ command. We took a quick look at this command when we were configuring router interfaces, so these things aren’t specific for switch interfaces. This time, let’s focus on some of these statistics at the bottom.
There are lots of different kinds of counters shown here and you don’t have to know all of them at this point, so let’s just focus on some. First up, not errors, but you can see the total number of packets received on the interface, and the total number of bytes in those packets. Next up is runts.
These are frames that are smaller than the minimum Ethernet frame size of 64 bytes. Beside the runts counter is the giants counter, which counts received frames which are larger than the ethernet maximum frame size of 1518 bytes. Next up, CRC, which counts frames which failed their CRC check.
If you remember, the CRC is the cyclic redundancy check done via the FCS, or frame check sequence, in the trailer of an ethernet frame. It’s used to detect errors, and if an error is detected this counter goes up. Next is frame, which counts frames that have an incorrect or illegal format.
Input errors is a total of various counters, including the four I just mentioned. Finally, output errors counts frames the switch tried to send, but failed due to an error. Keep in mind, I’m showing you all of these counters on a switch, but they are the same on a router.
Before moving on to today’s quiz, let’s review what we covered. We talked about interface speed and duplex. I explained the concepts of full and half duplex.
Full duplex is preferred, and in most modern networks all devices will be able to operate in full duplex, so they can send data freely without waiting for other devices to stop sending. We talked about speed and duplex autonegotiation, which allows devices to find the appropriate speed and duplex settings without manual configuration. We looked a bit at interfaces status using the ‘show interface status’ command, such as the statuses connected, notconnect, and disabled.
We’ll look more at this throughout the course. Finally, we looked at some interface counters and errors such as runts, which are frames that don’t meet the minimum size requirements, and giants, which exceed the maximum frame size requirements. Okay, let’s get started on the quiz.
Remember, there will also be flashcards which you can download from the link in the description to help you remember what you learned in this video. Let’s go to question 1. There is a duplex mismatch between SW1’s F0/1 interface and SW2’s F0/1 interface, which are connected.
Autonegotiation is disabled. What will be the result? A, improved performance.
B, collisions will occur. Or C, SW1 will sense SW2’s duplex setting and adjust to match. Pause the video to think about your answer.
The answer is B, collisions will occur. The half duplex side is unable to send and receive data at the same time. However, the full duplex side is unaware of this, and will send data even if the half duplex side isn’t ready to receive it, causing collisions.
So, B is correct. It definitely won’t result in improved performance, in fact performance will probably be much worse, so A is incorrect. SW1 won’t be able to sense SW2’s duplex setting without autonegotiation enabled, so C is incorrect too.
Let’s go to question 2. What is used on half-duplex interfaces to detect and avoid collisions? A, CSMA/CD, B, CSMA/CA.
C, Autonegotiation, or D, Duplex Auto. Pause the video to think about your answer. The answer is a, CSMA/CD.
CSMA CD stands for carrier sense multiple access with collision detection. It describes how devices using half-duplex listen for activity on a network segment, and then send data only when other devices aren't sending. It also describes how a device will react when a collision does occur.
So A, CSMA/CD is correct. B, CSMA/CA, is a similar concept that we’ll cover later in the course that stands for carrier sense multiple access with collision avoidance. C, autonegotiation, and D, duplex auto, are unrelated to how a half-duplex interface detects and avoids collisions.
Let’s go to question 3. Which command shows various counters of errors detected on an interface? A, show interfaces.
B, show ip interface brief. C, show interfaces status. Or D, show interfaces errors.
Pause the video to think about your answer. The answer is A, show interfaces. Let’s check the output of each command.
As you can see, show ip interface brief doesn’t display any counters of errors detected on the interface. It’s a useful command for checking the IP addresses of interfaces, checking if they’re shutdown, etc. Show interfaces status also doesn’t show any error counters.
It gives some basic information about interfaces such as description, status, duplex and speed. show interfaces errors isn’t a real command, so an error message appears when I try to enter it. Show interfaces displays counters of errors detected on the interface.
In this case I used ‘show interface f0/1. Because so much information is given for each interface, you probably won’t use this command to view all interfaces at once, but rather specify an individual interface to check. The counters are at the bottom of the output.
So A, show interfaces, is correct. Let’s go to question 4. Which are examples of errors that might occur on a network interface?
A, runts, giants, broadcasts. B, shorts, longs, oversizes. C, packets, bytes, inputs, outputs.
D, runts, giants, CRC. Pause the video to think about your answer. The answer is D, runts, giants, CRC.
The names in B and C are not real errors that can occur on an interface. In option A, runts and giants are possible errors, but broadcasts are not errors, they are part of normal network operations. So D, runts, giants, CRC, is the correct answer.
Remember, Runts are frames that are too small, Giants are frames that are too big, and CRC counts the frames that failed their CRC check, which is done in the frame check sequence in the trailer of the ethernet frame. Let’s go to question 5. SW1 is trying to autonegotiate interface speed settings with SW2.
However, autonegotiation is disabled on SW2’s interface. SW2’s interface is configured with a speed of 100 Mbps and full duplex. What speed and duplex settings will SW1 use?
A, speed 100 megabits per second, duplex full. B, speed 100 megabits per second, duplex half. C, speed 10 megabits per second, duplex full.
Or D, Speed 10 megabits per second, duplex half. Pause the video to think about your answer. The answer is B, speed 100 megabits per second, duplex half.
If the remote device has autonegotiation disabled, but it's enabled on the local device, the local device will attempt to sense the speed the remote device is using, so SW1 will be able to correctly set its interface speed to 100 megabits per second to match SW2’s. However, it can't sense the duplex mode of SW2’s interface, so the rule is if the speed is 10 or 100 megabits per second, half duplex will be used, otherwise full duplex will be used. In this case, half duplex will be used, which will result in a duplex mismatch between the two devices, and probably poor performance on the link between them due to collisions.
Okay, that’s all for the quiz. Remember to make use of the supplemental materials for this video, I’ve made a set of flash cards to be used with the flashcard software Anki to help you remember what you learned in the video. Also, there will also be a packet tracer practice lab to give you hands-on practice configuring Cisco devices.
That will be the next video. Good luck with your studies. Thank you for watching.
Please subscribe to the channel, like the video, leave a comment, and share the video with anyone else studying for the CCNA. If you want to leave a tip, check the links in the description. I'm also a Brave verified publisher and accept BAT, or Basic Attention Token, tips via the Brave browser.
That's all for now.
