Saturday, January 24, 2009

How ARP and Ethernet Work

So far we've talked about how IPv4 encodes data into a packet, and how routers learn which direction to forward those IPv4 packets based on the destination IP address and the route table.

But in the end, routers and hosts need to encode the IPv4 packet onto a physical medium. Examples of physical mediums include fiber, twisted pair, coax, radio waves, lasers, and microwaves. Each encoding rate and medium requires a specification or protocol definition.

Ethernet is a family of similar encoding specifications which is dominant on the Internet today. Ethernet uses a 48-bit address known as a MAC address or hardware address. There are ethernet specifications for speeds ranging from 10 megabit to 10 gigabit and for various copper and fiber physical media.

Using ethernet requires the ability to map an IPv4 address into a MAC address. That is accomplished using the address resolution protocol (ARP).

In this episode, we introduce the ethernet frame format, ethernet address, and how ARP works.

Labels: ,


Blogger Shivlu Jain said...

your blog is awesome.

shivlu jain

January 30, 2009 at 7:09 AM  
Blogger Kristian said...

Hello Darrell, thanks for your Cisco teaching and podcasts.
It is curious people don't leave more comments! So I will!: Actually a hint and question to it reg. ARP (not really Cisco, but general):
During my tech experience I found networking devices as Print Servers (HP JetDirect, Axis...) or net-web-cameras (Axis) and so on, with no initial IP address which should be setup-ed by connecting PC end host to same net segment (not routed), make static ARP entry and do a/the PING towards choisen IP:

{Win}>arp -s 01-02-03-04

where IP is correct what you choose and
MAC is real MAC address of network device

and check it
{Win}>arp -a

Interface: --- 0x2
Internet Address Physical Address Type 12-34-56-78-90-ab dynamic 01-02-03-04-05-06 static

and furthermore PING that


This way - you pinging device with till that time not existing IP address which you setup-ed with static ARP.
After this you could unconfigure the static arp entry by

{Win}>arp -d

But if everything went fine, the device will know and remember THIS is my IP address!
Reseting this learned IP is mostly done by push some hardware button on device and the process could be repeated...

I will add that sometimes the PING need to be "special" where you need to specify the size of packet or the TTL in the header of IP packet provided in PING:

{Win}>ping -l 66

{Win}>ping -i 127

This process need to be surely supported by device (I suposse),
but is universal for any system and on UniX is similar!...

I hope my explanation was clear,
if not please do not hesitate to contact me

March 10, 2009 at 11:55 AM  
Blogger Rod said...

Nothing better like the basics... the key to a strong foundation. Thanks for doing this Darrell.

December 9, 2009 at 10:58 PM  
Blogger Charles said...

I am very much enjoying your posts. I have one question/comnent have you thought of adapting you lecture for those of us using cisco's packet tracer?

Othe wise first rate. Thanks again

March 14, 2010 at 1:54 PM  
Blogger Darrell Root said...

I'm afraid I don't have access to the Cisco packet tracer since I'm not associated with a Cisco network academy. And even if I did become associated with an academy I'd have to be careful regarding confidentiality agreements. For example, I'm confident making videos of Cisco's training curricula slidedecks would be a violation. And rightly so. They worked hard on those high-quality presentations and get to control distribution as the authors.

March 14, 2010 at 4:02 PM  
Blogger Yellow Blade said...

This comment has been removed by a blog administrator.

January 18, 2016 at 1:40 AM  

Post a Comment

Subscribe to Post Comments [Atom]

Links to this post:

Create a Link

<< Home