IT Hardwares


eg: UK or Brides UK or Classical Art or Buy Music or Spirituality

Business & Money

Technology

Women

Health

Education

Family

Travel

Cars

Entertainment

Project Management Certification Exam

View:

When it comes to assigning an IP address to all the PCs on our network, along with their network mask, DNS server location, and more, we've got two choices on how to do it:

• Go to every workstation and configure the workstation statically

• Go to every workstation and configure them all to use DHCP

What we have here is the classic "static vs. dynamic" argument. I don't want you to think I'm lazy, but I'll take the dynamic way of doing things almost every single time.

You may wonder why it matters, since both methods involve visiting each workstation. You're right about that, and even though it's a lot quicker to configure a workstation to get its IP address and mask from a DHCP server than it is to configure the entire IP address and mask statically, the real benefits come in when the network changes.

And take it from me - your network will change. You'll remove hosts, you'll add hosts, and if the previous network manager didn't plan for future growth, the day may come when you've got to change the IP numbering scheme for your network. The choice that was made originally between static configuration and DHCP will then determine how easy the change will be.

• If the network was statically configured, you will now have to go to every single workstation and change their IP addressing to the new scheme.

• If the network is using DHCP, you simply change the networking scheme on the DHCP server and allow the workstations to get their new addresses dynamically.

Believe me, I've performed IP address changes in both fashions, and I'll take DHCP every time! Avoiding static IP address assignments also cuts down on the chances of two hosts in your network being assigned the same IP address.

When hosts receive an IP address from DHCP, it does not belong to that host permanently. The address is actually leased from a DHCP server. Let's walk through the DHCP process from the host's point of view.

First, the DHCP Client boots up, and sends a DHCP Discover packet onto the network. The host does this in order to "Discover" a DHCP server or servers. This Discover packet is a Level 3 broadcast, which has a destination IP address of 255.255.255.255.

Every DHCP Server that hears this broadcast will respond with a DHCP Offer, and it's an IP address and mask that is being offered. The DHCP Offer also includes a subnet mask, the IP address of the DHCP Server sending the response, and how long the host can keep this address (the DHCP lease duration).

If multiple DHCP Servers happen to hear this broadcast, they will each offer an IP address.

The DHCP Client will accept the first offer it receives. It does so by sending another broadcast, a DHCP Request packet.

The DHCP Server that made the offer that's being accepted will now send a DHCP Acknowledgement, which contains the rest of the information the host needs to function, including the location of a DNS server. The DHCP Server that made the offer that was not accepted will return the offered IP address to its range of assignable addresses, its address pool.

I made several mentions in this section about these DHCP packets being broadcasts. Do you remember which network connectivity device does not forward broadcasts? That's right, it's our old friend, the router!

If a PC is on one side of the router and the DHCP Server is on another side, we've got a problem. The initial DHCP Discover packet is a broadcast, and the router will not forward that broadcast to the DHCP Server. Luckily, this doesn't mean that we need a DHCP server on every single subnet on the network, because we can configure the router as a DHCP Relay Agent.

A DHCP Relay Agent will forward DHCP Requests to the DHCP Server. You can also configure a Windows server as a DHCP Relay Agent. Naturally, the Relay Agent must be on the same physical segment as the hosts that cannot reach the DHCP Server - never put it on the same segment as the DHCP Server itself.

Configuring a router as a DHCP Relay Agent is a lot different than configuring a Windows server, and your Network+ exam will not require you to configure either. You should know why the need for a Relay Agent exists, though, and should you need to configure one on your network, always check the vendor's documentation.

Project Management Certification Exam

Hubs and repeaters both work at the Physical layer of the OSI model. They do not help to direct traffic toward the proper destination, as routers and switches do. Their purpose in the network is simply to make the signal strong enough to reach its destination, by regenerating the signal. Hubs and repeaters both take an incoming data signal and make a clean, strong copy of the signal. Otherwise, the signal would suffer from attenuation, the gradual weakening of a signal as it gets farther and farther away from the source.

Repeaters are just about extinct in today's networks, because they only have one input port and one output port. Naturally, this means we could only regenerate a single data signal. Hubs are really just multiport repeaters, but their multiple ports give them some value in today's networks.

By using a hub to connect hosts, we're left with one collision domain. Any data sent by these hosts is subject to colliding with data sent by another host. We'll talk more about collision domains in another CompTIA tutorial, but keep in mind that hubs do not break them up.

Using a hub also results in one broadcast domain, since a broadcast sent by any of these hosts will be heard by all other hosts.

Bridges and switches work at the Data Link layer of the OSI model, making them "Layer Two" devices. Both read MAC addresses to create a MAC address table, which allows the switches to help send frames to their proper destination. You see very few bridges in today's networks, especially with the advent of Layer Three switches.

It's not that bridges are bad at their job, it's just that switches are better. They can operate at higher speeds and offer more options, especially now that many new switches can perform routing as well as switching. (Layer Three switches used to be called brouters, but as the usage of bridges has faded, so has the use of this term.)

There are two rather confusing facts regarding switches that confuse many newcomers, and I want to mention them again here. First, MAC addresses are also called physical addresses because they physically exist on the network interface card (NIC). They have nothing to do with the Physical layer of the OSI model.

Second, you're going to hear several different names for the MAC address table throughout your career and your networking certification studies.

MAC address table

bridging table

switching table

physical address table

Layer Two address table

They're all the same thing! But no matter what you call it, you should know how switches build this table and use it to determine the appropriate action to take with incoming frames. A switch is either going to forward, flood, or filter a frame, depending on whether it knows how to reach the destination MAC address of the frame. We'll look at that process in the next CompTIA Network+ certification exam tutorial! See you then!

Related Articles
Author
Most Popular

•	It Project Management Certification, by John-reiling6
•	Professional Project Management Certification, by Nitesh Ranjan Bharti
•	Project Management Certification Exam, by Chris Bryant
•	Project Management For It, by Michael Russell
•	Project Management Pmp Certification, by Nitesh Ranjan Bharti

Chris Bryant has sinced written about articles on various topics from CISCO CCNA, Personal Desktop and Cisco CCNP. . Chris Bryant's top article generates over 27100 views. to your Favourites.

Advantages And Disadvantages Of Video Conferencing
As a movie player, the iPod video is below the expectations, but it may be considered a great step in the video technology