Ethernet Introduction for CCNA R&S

This topic is to discuss the following lesson:

https://networklessons.com/cisco/ccna-routing-switching-icnd1-100-105/introduction-to-ethernet/

Hello Florian

There may be a slight change in what Rene has in this post. I’ll post a clarification for you here and we’ll let Rene know about the possible changes needed.

As you know, a MAC address is composed of 48 bits, or 6 sets of 8 bits. The bit that determines if a transmission is unicast or not is the least significant bit of the first octet, or bit 8. So, if a device sends a frame to a destination MAC address with bit 8 set to 0, then that transmission is considered unicast. In other words, it goes to one host on the network. For this reason, all devices in the world have MAC addresses that have 0 as the value of bit 8. (I suggest you check this out by looking at the MAC address of your PC for example).

When a device sends a framne to a destination MAC address whose bit 8 is set to 1, then the transmission is NOT unicast. That is, it is either multicast or broadcast. If it is multicast, it will be in the range of 01-00-5E-00-00-00 to 01-00-5E-7F-FF-FF. If it is broadcast, then it is FF-FF-FF-FF-FF-FF. Note that for all of this range, bit 8 is set to 1.

Now, bit 7, or the second least significant bit of the first octet, is the bit that indicates whether or not the MAC address is universally administered or locally administered. A value of 0 indicates a universally addministerd MAC while a value of 1 indicates a locally administered MAC. A universal MAC address is one that belongs to a physical device. Similar to the case with bit 8, all physical devices in the whole world have MAC addresses with bit 7 set to 0.

Now, you may ask, what if bit 7 is set to 1? When would that occur? There are cases where “virtual” MAC addresses are needed, that don’t belong to a specific piece of hardware. One such case is when using HSRP between two routers, where a virtual interfaces is created with a virtual IP address and a virtual MAC. It is even possible to manually enter the MAC address of a device to replace the burned in address. How is this done? Well, MAC addresses are never read directly from the hardware itself. They are copied into RAM and referenced from there. If you can go in and change the value in RAM, then you essentially change the MAC address.

I hope this has been helpful!

Laz

hi rene,

above you write:

BC which stands for broadcast; If your Ethernet frame is a broadcast than you have to set this bit to 1.

and

Local: this bit has to be set when you change your MAC address. Normally a MAC address is unique on the planet; if you change it it’s only locally unique within your network.

But if your ethernet frame is a broadcast than it would be FF:FF:FF:FF:FF:FF, right? you mean that in order to get all F´s the first bit would need to be set to 1?
And if you change your mac address you can change it to whatever you like too, or not? why would i need to set this bit to 1?

thanks

florian

Hi Rakesh,

It’s unlikely that we ever will have issues with MAC address exhaustion. 2^24 = 16777216, that’s close to 17 million unique MAC addresses for 1 OUI.

Within a broadcast domain, you need to have unique MAC addresses. Broadcast domains are typically very small so it’s very rare to see duplicate MAC addresses within a broadcast domain. If a vendor runs out of MAC addresses for their OUI, they probably can get another OUI.

Perhaps someday we run out of MAC addresses but I doubt we will still be using MAC addresses by then :slight_smile:

Rene

Hi Romani,

H1 and H3 will send a short jam signal (long enough for everybody to hear it) and then stop transmitting and start their random clock. When the clock expires, then they will attempt transmission again.

Rene

hi ,

in old times Ethernet used to be known as RJ45 cable. is it not changed now ? I mean now if I have 2 routers on a distance of 500 meters , we use fiber right ? so fiber cable also supports enternet ? how long it can go then or we need some repeater sort of thing ?

  1. I know that we also have PoS but what it is ? is that only for wan ? what all other technologies exist like as PoS in today’s networks ?

Thanks

Abhishek,
Technically, RJ-45 refers to just the modular connector–just like RJ-11 refers to the smaller “phone” type connector. The ethernet cable itself would be referred to as CAT-3, CAT-5, or CAT-6 (most modern cable). The length limit of CAT-5 / CAT-6 cable is 100 meters. For longer than that, you would have to use fiber. As far as the length of fiber, that depends on what kind of speed you are trying to push over it. The higher the speed, the shorter the distance. For example, you can push 100 Mbit/s over 2 kilometers, while 10 Gb/s drops the distance all the way down to about 500 meters.

Ethernet is a layer 2 (data-link) standard, so it is independent of layer 1 (physical). Therefore, you can run Ethernet on Fiber or CAT-5 / CAT-6.

POS = Packet over SONET. These are popular (and very expensive) long haul circuits that are point to point in a WAN environment.

The type field indicates the payload of the frame, for example:

  • IPv4: 0x800
  • ARP: 0x806
  • IPv6: 0x86DD

Hi Lynkaran,

Multicast isn’t really used on the Internet. 99% of it is unicast, even if you watch a streaming video they just unicast it to you.

Rene

Hi Rene anyway you can explain the IEEE 802.3 Frame?

Hi Siva,

This probably has something to do with efficiency. When a switch receives a portion of the Ethernet frame, it can already make some decisions when it sees the destination MAC address. For example, cut-through switching is a technique where the switch will forward the frame just after reading the destination MAC address, even though it hasn’t received the entire frame yet.

Rene

Hi Rene,

I need a little clarification on Type and length field. In case when this field represent Length of the frame then how will it interpret the upper layer protocol in use?

Regards,
Himanshu

Hello Himanshu

Let me refer you to a previous answer in this thread:

Now concerning your specific question: How will an IEEE802.3 frame know what the upper layer protocol in use is if this field is used as a length field? Remember that the Data Link layer is composed of two sub-layers, the Media Access Control (MAC) layer and the Logical Link Contro (LLC) layer. The LLC layer can add an additional header extension to an Ethernet frame called a SubNetwork Access Protocol (SNAP) header. This SNAP header includes things like an organization code and an Ether Type code which tells the device what upper layer protocol is being used.

Keep in mind that this information is beyond the scope of a CCNA R&S exam.

I hope this has been helpful!

Laz

1 Like