NET 102 - Networking Essentials II

Chapter 3, Cabling and Topology; Chapter 6, Installing a Physical Network

Objectives:

This lesson goes into more detail about some concepts we have touched on already. Objectives important to this lesson:

  1. Differentiate among physical network topologies and their relationship to logical network topologies
  2. Distinguish access protocols and topologies among LANs, WANs and WLANs
  3. Demonstrate proficiency with common utilities in a command environment for diagnostics and troubleshooting
  4. Utilize the appropriate hardware tools in testing and troubleshooting the network infrastructure
  5. Demonstrate troubleshooting skills of common network connectivity issues using a systematic approach
Concepts:
Chapter 3

The author begins chapter 3 with a discussion of topology, which we talked about last week. Topology is the study of shapes and configuration. Physical topology is the way a network is wired. Logical topology is the way it works, regardless of the wiring. Physical network configurations typically fall into one of these types:

http://en.wikipedia.org/wiki/File:NetworkTopologies.svg

A few more facts are added to what we discussed last week. We are told that the first network topologies were bus and ring.

  • Bus - A bus works in a baseband mode, which means that every transmission can be received by all devices at about the same time. This is not the same thing as a broadcast.
    • Most devices ignore a transmission unless it is addressed to them
  • Ring - like a daisy chain, going from one station to the next and all the way back to the server or the first node in the ring;
    • transmissions are passed from one device to the next, adding a delay to the time it takes to get a transmission
  • Star - starting at hubs or switches with individual cables radiating away from them for each node, like the picture at the bottom of page 46
    • Units affected by media failure: One, unless it is the hub/switch/router, in which case all nodes are affected. This is what the text means by fault tolerance.
  • Mesh - redundant connections, to survive in case one cable is broken;
    • in the image above, the Mesh example is a partial mesh
    • in the image above, the Fully Connected example is a fully redundant mesh
    • The text gives us a formula to calculate the number of connections needed to fully mesh any number of nodes:

      Let
      n represent the number of nodes. Let c represent the number of connections to connect those nodes in a full mesh.
      c = n * (n - 1) / 2
      If n is equal to 10 nodes, we multiply 10 times 9, and divide by 2, which gives us 45 connections.

      You will want to know this formula.

    • Units affected by media failure: Few or none
    • commonly used in wireless networks; think of cell towers and cell phones
  • Hybrid - combination of more than one type, such as a series of bus networks, connected together in a ring, or a combination of networks running different Network Operating Systems.
    • Note the pictures on page 47, showing how a bus or a ring can be contained in one box, and the hosts on that network can be star-wired to that box
    • The text states that "any form of network technology that combines a physical topology with a signaling topology is called a hybrid technology:" This makes no sense to me. I do not know of any technology that cannot be described both ways. Let's ignore this statement.
    • Can be Easy or Difficult to install
    • Easy to reconfigure
    • Easy to troubleshoot
    • Units affected by media failure: All, if the central hub fails. One, if a workstation fails.
  • Point-to-Multipoint  - See the illustrations on page 49, showing how a point-to-multipoint network can resemble a star
    • like a mesh, this fits better if you think about cell phones; your smart device and the hubs around it (cell towers) determine which nodes should connect to the network through which hubs; the answer to this question changes as you move around with your device
  • Point-to-Point - not really a network, but this kind of connection can use network protocols; think of two smart phones that are sharing a play list, or a tablet that is printing by a Bluetooth connection to a printer

On page 50, the author begins several pages about network cable. Let's start with a quick list:

  • coaxial - several kinds
  • STP - a few versions
  • UTP -  several versions
  • fiber optic - several connector types
  • historic cable connectors - some old style connectors that were once used on networks

Coaxial cable is called that because it has two conductors, one wire in the center that carries signals and a conductive sheath around it that is used for a ground and an EMI shield (although not a great one). They share a common axis, hence coaxial or coax. Most people have seen this style of cable used with cable television. There are several varieties, and some look a lot alike.

The wiring standards used for network coax have been different from those used for cable TV. This is a list of cable used through networking history:

  • 50 ohm cable, available as RG-8 and RG-11. Used in Thick Ethernet, with drop cables and vampire taps.
  • 50 ohm cable, available as RG-58. Used in Thin Ethernet, with BNC connectors
  • 75 ohm cable, available as RG-59. Used in cable TV operations, with F connectors.
  • 75 ohm cable, available as RG-6. Preferred for networking and cable TV, due to lower signal loss than RG-59, typically with F connectors.

Facts about coax:

  • The coaxial line is essentially a single bus, going from one station to the next.
  • At each end of the line, the cable has to have a terminator on it.
  • At one or both ends, depending on the network type, it also has to be grounded.
  • If using thin Ethernet, T-connectors are used at NICs, and BNC connectors on the cables.
  • If using thick Ethernet, vampire taps are used. They are called vampire taps because little teeth bite into the cable (to contact the shield), and a big tooth bites deeper to contact the central conductor when you screw the clamp down.
  • If using modern cable connections, such as connecting to a cable modem, RG-6 cable should be used when possible. It is compatible with existing RG-59 cable, which has the same impedance (75 ohms), but RG-6 provides better throughput.

BNC T-connector The example on the right shows a typical T-connector with BNC fittings. The fitting on the bottom of the image might attach to a port on a NIC that looks like the barrel on either end of the top of the T. Attachment is achieved by pushing the connector onto the barrel of the port, then twisting the collar of the connector to lock onto the pin that is part of the port. In other words, it mounts like a bayonet. It does not stress the cable like the ST connectors used with fiber optic.

BNC connectorThe next (enlarged) picture shows a BNC connector attached to a thin Ethernet cable. Such a connector would be used to attach to one of the T-connector barrels in the photo above. The other end of the cable would run to the next node on the network.

Shielded Twisted Pair (STP) is not commonly used any more. One example is shown on page 53,which shows that it has a shield, like coax, but it also has copper or aluminum wires in it, like the illustration of UTP on the same page. On page 95, there is an example of twinaxial cable, used with 1000BaseCX networks. Twinaxial cable is considered a special case of STP.

Unshielded Twisted Pair (UTP) was discussed last week. The author remarks that we must be careful to avoid running UTP near sources of electrical interference, like fluorescent lights, generators, and motors. Look at the chart on page 54 and the marginal note on page 55 for information about UTP categories that have been used or proposed. Know characteristics and uses of CAT 3, 5, 5e, and 6. Some good examples and a good discussion about UTP can be found at the Fiber Optic Association web page about the subject. Note the information on that page about three ratings for UTP that determines where it can be used:

  • CMX - general use PVC cable, which should not be used in the following two cases
  • CMR - riser rated cable, which is fire retardant, and should be used in channels that pass from one floor of a building to another
  • CMP - plenum cable, which does not emit toxic fumes when burned (unlike PVC), should be used when running through any air space around people, such as between ceiling tiles and a real ceiling or through an air shaft

Fiber optic has one or more conductors that can be glass or plastic, and it conducts light instead of electricity. The conductor is called a waveguide, and is covered with cladding, a material to reflect the signal back into the center of the conductor. Two basic types exist. Multimode fiber (MMF) can carry more than one signal on the same core conductor, and usually uses LEDs as light sources. Single mode fiber (SMF) can carry only one signal at a time, and usually uses lasers as light sources, but it has a longer segment length. Network standards that run segments longer than 2000 meters use SMF. The illustration on page 56 shows the three connector types that the author tells us to know for the Network+ exam:

  • ST - a bayonet mount, single fiber connector (author's mnemonic: stick and twist)
  • SC - a snap on, single fiber connector (author's mnemonic: stick and click)
  • LC - a smaller, snap connector that connects two fibers to a NIC at once (author's mnemonic: little connector)

A section on other cables starts on page 56. These cables are identified by their connector types.

  • serial - the text shows a picture at the top of page 57 of a port that can be called a DB-9 port, or a D-9, or an RS-232 serial port (this one is confusing); signals are sent on one wire, one bit at a time (serially); to some people, a serial port looks like a VGA port, but VGA ports are blue, are female, and have three rows of sockets for pins
  • parallel - calling a serial port an RS-232 port is confusing because a parallel port can be called an RS-232C port, or a DB-25, or just a parallel port; signals are sent on (typically) eight wires at once (in parallel)
  • firewire - also called IEEE 1394, the text shows a poor picture of one kind of firewire connector on the bottom of page 57; more images can be seen on the Wikipedia page about IEEE 1394; technically, firewire is what Apple calls this kind of port, but other vendors use the term as well
Chapter 6

Chapter 6 promises to be more practical, and it begins with a poor example: what would a network be like if you just stuck a switch on a table in the middle of the room and ran cables across the floor to devices? Well, it would be like a lot of home networks. The text lists three problem areas with that approach:

  • exposed cables are safety hazards and they are often broken by staff
  • cables that run near other devices suffer degraded service
  • if you can't tell which cable goes to what machine, you have to diagnose that before you can change the layout of your network

The text proposes using structured cabling standards, which address the three problems above. It uses some specific vocabulary:

  • telecommunications room - also called a wiring closet, this is a place to hold your routers, switches, and other networking devices; typically, you will have one or more of these rooms on each floor of your buildings
  • horizontal cabling - this is the collective term for all the cables that connect from the telecommunications room to all your devices: the cables "mostly" run in horizontal directions
  • run - a single cable from the telecommunications room to a single device
  • work area - the general area that holds user devices that connect to the telecommunications room; typically, desks or cubicles hold PCs and phones in work areas

Horizontal cabling is usually UTP, because of costs and ease of deployment. It is possible to buy stranded core or solid core UTP, as illustrated on page 111. The TIA/EIA standard is to always use solid core UTP. In most cases, use standard 4 pair UTP, of the highest CAT rating you can buy. The text shows three kinds of UTP on page 111, one of them being 25 pair UTP. More wires does not mean a higher CAT rating. 25 pair UTP is more commonly used in setting up telephone lines than Ethernet lines. In a large facility, you may see 100 pair cable, but the text does not show an example of it. The text refers us back to chapter 3 to make the correct choice between PVC, riser, or plenum grade UTP, which completes its discussion of horizontal cabling characteristics.

Note: some exam questions may ask which CAT rating to choose to meet a bandwidth requirement and to save money. In that case, choose the cheapest cable that meets the performance requirement.

Twice on page 111, the text refers to the telecommunications room as an intermediate distribution frame, which simply sounds wrong after all the talk about OSI Data-Link layer frames. It may be easier to turn the page, look at the equipment racks that may be found in the telecommunications room, and imagine them as intermediate distribution frames. This idea is closer to the general definition of a distribution frame found at Wikipedia.

As I just mentioned, the text shows us some equipment racks on page112, which are commonly found in telecommunications rooms (and in server rooms). Let's assume that the author tells us that racks are typically 19 inches wide because he has seen a test question about it.

The author tells us to avoid moving horizontal wiring very often, because doing so can lead to broken wires. He uses this as a reason for the use of patch panels, as shown on page 113. Not sure I buy his reason, since a patch panel does not make moving cable any easier, but they are often used in telecommunications rooms, so you need to know about them as a practical matter and for the test.

This video on YouTube is done in high resolution with good camera work. It shows a technician preparing a CAT 5e cable, and attaching it to a 110 block patch panel, as discussed on page 113. In the video, his panel has convenient color codes for 568A and B wiring, which demonstrates the need to know what kind of wiring was used on the other end of the same cable run. The video also demonstrates the proper way to use a punchdown tool, using it to press each wire into place, and cutting off the excess wire with the blade side of the tool at the same time.

For the Network+ test, you will also want to know about 66 block patch panels, typically used for phone networks.

The text continues to discuss patch panels on the next page, advising that you should buy patch panels that match the grade of cable you are using. The discussion moves on to patch cables. You have probably heard this phrase before but may not have seen this proper use of it. A patch cable is meant to be used across a short distance, maybe from one patch panel to another, but probably to a switch or a router. It may contain stranded copper instead of solid copper conductors.

The work area is discussed next. The book points out that if you are a cable installer, your only concern about the work area is that the horizontal cable runs reach it, and that you have connected the right cables to the right sockets, matching the wiring standards at the patch panel. The sockets in the wall (or cubicle) plates are called jacks, and the connectors you install for the jacks should, again, match the CAT level you are using and the wiring standard you choose. The text tells you to install a patch cable from the PC to the wall jack. In practice, I have see techs simply make shorter cables from their reels of wire that are used for the runs.

This video shows the technician from the video above, Mercy Salinas, connecting a CAT 6a cable to a keystone (modular) jack. Note the flexible spline in the center of the cable and the greater number of twists in each conductor pair. Our hero, Mr. Salinas, makes a mistake when he attaches two of the wires. See if you can catch it. He corrects the mistake before he cuts the excess cable, but he does not mention it in the video. Possible test question: why might it not matter that he made that particular error?

In the scenario above, we would have a patch cable from the PC to the jack, a horizontal run from the jack to the patch panel, and a patch cable from the patch panel to a switch. The text explains that this entire length of cable counts toward the hundred meter limit on cabling from a host to a concentrator. Because of the patch cables, whatever they are made of, the TIA/EIA standard for a horizontal run sets its limit at 90 meters.

If your network is self contained, and is never meant to connect to any other network, wow, what's it like back in the sixties? If you are in the 21st century with the rest of us, you need to connect to the Internet, which means you need an Internet Service Provider (ISP). You may connect to your ISP with a DSL (if they are a phone company) or a cable modem (if they are a cable TV company). Whichever it is, the piece or equipment that connects you to their system is the demarc, the point of demarcation between your network and theirs. The text mentions in a marginal note on page 117 that this actual device can be called by three similar names that all mean the same thing:

  • network interface unit (NIU)
  • network interface box (NIB)
  • network interface device (NID)

Sometimes, the provider's authority extends beyond the NIU (we have to call it something), which leads to a demarc extension. The provider would have administrative control, for example, over wiring that leads to another patch panel that leads finally to equipment managed by the customer.

The text uses the term smart jack on page 117. A smart jack is not a jack. It may refer to the demarc device from your ISP (or phone company, if it is for phone service only). It is smart in the sense that it can be used to diagnose problems with your connection to their system.

We talked about the telecommunications room being the intermediate distribution frame. Page 118 brings us to the main distribution frame. This would be the location of the demarc, as well as the location where the runs from various telecommunications rooms come together. A panel connecting various telecommunications rooms is referred to as a vertical cross connect. Imagine an IDF on each floor of a building, and an MDF on the floor where the demarc is located. Vertical runs would come together in this room on a vertical cross connect.

The text continues with a discussion of running actual cable in a building. It starts with a plan of each floor, and the constraints the installer must include in the plan. The installer works with the site owner to decide where to drill, where to place wiring closets, where and how to secure the cable in the structure. Page 120 shows a picture of a raceway, a cable conduit that is mounted on a wall. Local building codes may determine that you either must or must not use this kind of conduit.

The text lists several factors that affect the placement of each telecommunications room.

  • distance from the room to the jacks it services: no more than 90 meters
  • adequate power for all devices in the room
  • humidity - the author recommends avoiding high humidity, although higher humidity levels decrease the risk of electrostatic shock
  • cooling is essential for an enclosed room holding computer equipment; I sometimes turn on all my devices to warm up my work room
  • control access to the room; only authorized staff should ever enter it

The author continues with a discussion of pulling cable from the telecommunications room to its next destination, probably a work area. He discusses the use of cable trays, which are meant to control the placement of network cable runs above ceiling tiles. Once the cables are run, they need to be brought down to the mounting hardware you will use, whether that means wall plates or outlets made for cubicles.

The text discusses crimping your cables to jacks the fit the mounting brackets you are using. We have already seen a jack mounted in the video above. (If you did not watch it, watch it now.)

As the text points out, the critical tool needed when crimping RJ-45 connectors onto cable is the crimping tool. You can use a variety of tools to cut wire, but you must have a crimper to put a plug on the cable. Many people have their own style, and most of them lead to a good product. In the case of our author, I think he did not write what he meant at the top of page 124.

Yes, you cut the cable end, then trim the insulation. In his step three, he says to "insert each individual wire into the correct location". I read that and imagined a technician trying to insert the wires one by one into the RJ-45 plug. That would not work: the wires might easily go into the wrong channels. That, and he just told us not to bend them. How's that going to work, putting in one wire at a time? I hope that is not what he meant. Try this instead for a straight through cable:

  1. Cut the cable.
  2. Cut off some outer insulation The text says a half inch. Cut more and you will have enough length on the wires to work with them. You will trim them to about a half inch in a couple of steps.

    Note: If you are concerned about maybe cutting the wires when you cut the insulating jacket, follow Mr. Salinas' technique in this video: use the nylon string to tear the jacket farther down, trim the wires at the point of the initial jacket cut, and trim the jacket again. A good pair of scissors is nice for trimming.
  3. Cut the nylon thread and/or the spline, depending on what you have. (CAT 5, CAT 5a, CAT 6, etc.)
  4. Separate the wires, and lay them out in 568A or 568B alignment, whichever matches (or will match) the other end of the cable.
    The chart below assumes you are holding the plug with the gold connectors up, and the retaining clip down.
  5. Look at the plug, and trim the wires with an even cut so they are the same length, just long enough to reach the end of the plug, and so that the clamp in the plug will grip the outer insulation.
  6. Press your aligned wires into the connector, all at the same time. Last chance to check the position of the outer insulation. If it is wrong, pull the wires out now and adjust or start over.
  7. Once the wires and the outer insulation are in the plug correctly, place the plug in the crimper and squeeze firmly. If it works, it will feel like you have punched through something. If it does not work, try another plug, checking your work carefully before squeezing.
Pin Assignments for TIA/EIA 568B

Cable

end

Pin Color Color Signal Circuit

NIC connect

end

1 Orange/White
Orange White
TX data + Orange Circuit
2 Orange Orange TX data - Orange Circuit
3 Green/White
Green White
RX data + Green Circuit
4 Blue Blue unused Blue Circuit
5 Blue/White
Blue White
unused Blue Circuit
6 Green Green RX data - Green Circuit
7 Brown/White
Brown White
unused Brown Circuit
8 Brown Brown unused Brown Circuit

In the steps on page 124, the author refers to the RJ-45 plug as a crimp. I have not heard this word used as a noun before, but you may want to be aware of it in case you see it on a test. You will see it again in the text.

On page 126, the text discusses testing cables, which requires you to have a test meter that is more expensive than the average home user would accept. The author presents a list of diagnostic questions about a malfunctioning cable, some of which are a tad late. How long is the cable? Didn't we cover that? Are any of the wires not connected in the crimp (plug)? That happens, but it's hard to test. The author shows us a $100 continuity tester on the bottom of page 126, which he admits is of no use unless you can insert both ends of a cable into it. For a horizontal run, or a vertical run, this is pointless.

We are shown a time domain reflectometer (TDR), that is meant to tell us how far away the end of a wire (or a break in the wire) is. This could be handy, if we have a way to access the wire and fix it. In many cases, such as cables that go through walls, we may have to run a new cable. We have talked about crosstalk before. The author discusses using an advanced meter to measure near end crosstalk (NEXT) and far end crosstalk (FEXT). Both measurements are about signal leakage from one circuit to another. NEXT takes place in the connector we have in our meter. FEXT takes place at the other end of the cable.

The text finally mentions the word attenuation, which means the fading of a signal over distance. Attenuation is one of the reasons that cable segments have limits on their length.

The discussion moves on to fiber optic, and mentions the difficulty that has existed and still exists in putting connectors on fiber optic cable. The text reminds us that we can expect to see ST, SC, and LC connectors. It also mentions that fiber optic cable is immune to crosstalk and to electromagnetic interference problems. Fiber has other problems:

  • Fiber can break. The fiber equivalent of the tool to find breaks in the line is the optic time domain reflectometer (OTDR).
  • Attenuation is still a problem. Signals fade over distance on all media. You can also see dispersion, spreading of the signal.
  • Light likes to travel in straight lines. Fiber optic cable loses signal to light leakage when it is bent beyond its rated bend radius. Note the picture of the optical cable on page 130. The cable is glowing where it is bent beyond its useful bend radius.
  • Modal distortion only happens with multimode fiber optic. With multimode cable, we send two or more beams of light of different frequencies down the cable. This can lead to the signals not arriving at the same time, putting them out of synchronization.

NICs

The next discussion is about NICs. The first point is to recognize the kind of medium that a card is made for.

  • You should be able to recognize a card that has an RJ-45 jack, which means it is meant for a system the uses UTP. That socket does not look like the connectors for a fiber optic NIC.
  • I would like to show you some images of NICs for fiber, but most available images have protective covers over the ports.
  • Wireless cards typically do not have jacks, and may have an antenna.
  • Cards made for multiple media are less common, but you may see them. As a concept, they demonstrate that the cable medium (OSI layer 1) does not determine the network type (OSI layer 2).

On page 132, the text shows you three different physical connections you might see on a NIC. The is about how the NIC connects to the motherboard of a computer.

  • PCI - goes into a PCI slot on a motherboard; these slots are typically white
  • PCIe - goes into a PCI Express slot; the text shows you a PCIe x1 card; look at the photo on Tom's Hardware web site, and note the two kinds of yellow PCIe slots shown, as well as the two white PCI slots.

    Note that both types of slot are broken into two sections. The separator in the PCIe slot is closer to the jack on the card.

    There are more kinds of PCIe slots, so you need to make sure that the PCIe card you buy (e.g. x1, x2, x4, x8, x16) will fit the actual slot you want to place it in.
  • USB - for portable devices, needing temporary connections

The text also mentions drivers for a NIC. The author's advice is to insert the CD that came with the NIC and install... what, no CD? Yes, actually. My dad recently lost the driver (no idea how) for the NIC that is part of the motherboard of his desktop PC. I had to go online with another computer, find a driver for that model, download it, copy it to a USB stick, copy that to his computer, and then install the driver. After all of that, yes, the card was recognized again by Windows.

The last topic for NICs is the idea of installing more than one NIC in a computer. The text calls this by two names: bonding and link aggregation. The author found that doing so for a computer that was monopolizing his bandwidth improved his network performance, essentially widening the bandwidth for that device. He advises using the same brand and model for both NICs in the computer in question, and using the same brand of switches if you are passing traffic through multiple switches on your network.

The next topic addresses NICs, switches, hubs, and more. Link lights are LED lights that are associated with RJ-45 ports, and they give us diagnostic and performance information about the equipment using those ports. Technically, one LED for each port is called a link light or a connection light, which I have told you proves an electrical connection is present. Confusingly, all such lights on a port are called link lights as well. Take a deep breath, hold it, let it out slowly. Repeat until calmer.

The next useful link light is properly called a traffic light or an activity light. This light should blink intermittently when frames are being sent or received by the NIC.

The third light, that you will not see on newer NICs, is a collision light, which only lights up when a collision is detected. The text warns us that the Network+ test may refer to this light.

The text shows us a picture on page 134 of link lights on a switch. In this case, the switch has one light per port, but some have two. That switch probably has one per port because it has one row of lights and two rows of ports. If there is only one light for a port, it may show a steady light for connection, and a flashing light for activity.

The text follows this section with a discussion of troubleshooting. It does not provide much guidance. We will discuss these concepts in class.

Assignment 1: Chapters 3 and 6

  1. Do LabSim exercises 2.4.6 and 2.4.7 to practice making connections with patch panels and setting a PC to use DHCP.
  2. Do LabSim exercises 2.3.3 and 2.3.4 to learn about connecting ST, SC, and LC fiber optic cables.
  3. Do the even numbered multiple choice questions for each chapter.