C480 Chapter 13

verickle's version from 2016-11-01 19:43

Common Layer 1 Troubleshooting Issues

Question Answer
Bad cables or connectorsFaulty cables (with electrical characteristics preventing a successful transmission) or faulty connectors (which do not properly make a connection) can prevent successful data transmission at Layer 1. A bad cable could simply be an incorrect category of cable being used for a specific purpose. For example, perhaps you interconnected two 1000BASE-TX devices using a Cat 5 cable (instead of a Cat 6 or higher cable), resulting in corrupted data. See Lesson 4, "Ethernet Technology," for a listing of Ethernet types and their corresponding supported cable types.
Opens and shortsAn open is a broken strand of copper, preventing current from flowing through a circuit. However, a short occurs when two copper connectors touch each other, resulting in current flowing through that short rather than the attached electrical circuit because the short has lower resistance.
Splitting pairs in a cableAn unshielded twisted-pair (UTP) cable consists of eight separate copper leads. However, only four of those eight leads are used for data (two transmit leads and two receive leads). This results in four unused leads. Some installers use those four extra leads to support a second Ethernet connection on a single UTP cable. Although such an approach can function, nonstandard wires are being used for connecting the second Ethernet connection. Therefore, you should be aware of any nonstandard pinouts used in the network that you are troubleshooting.
dB lossThe signal power of a data transmission might be degraded to the point where the transmission is not correctly interpreted by a receiving device. This loss of signal power, called a decibel loss (dB loss), could result from exceeding the distance limitation of a copper or fiber cable.
Transposed Tx/Rx leadsSome Ethernet switches support medium dependent interface crossover (MDIX), which allows a switch port to properly configure its leads as transmit (Tx) or receive (Rx) leads. You can interconnect such switches with a straight-through cable (as opposed to a crossover cable). However, if a network device does not support MDIX, it needs an appropriate cable (that is, a crossover cable) to allow its Tx leads to connect to the Rx leads on a connected device, and vice versa. Therefore, care must be taken when selecting cable types interconnecting network components. More information on crossover cables and MDIX can be found in Lesson 3, "Network Components."
Cable placementBecause copper cables are subject to electromagnetic interference (EMI), you should arrange cables to minimize interference. Ideally, Ethernet cables should not be placed in close proximity with high voltage cables, generators, motors, or radio transmitters. For example, when running cables between buildings via underground conduit, network cabling is ideally placed in a separate conduit than electrical cables.
Distance limitations exceededIf Ethernet devices are interconnected using a cable that exceeds the Ethernet distance limitations for the cable type, a digital transmission between those devices can be degraded to the point where the receiving equipment is unable to correctly interpret the transmission. Therefore, network designs should consider distances between devices.
CrosstalkCrosstalk can occur when an analog connection creates an electromagnetic field around its conductors, inducing its waveforms on a nearby analog connection. This phenomenon is most commonly experienced in an analog phone call. Crosstalk can be minimized by using a higher category of cabling because higher categories of cables better limit the radiation of electromagnetic waves.
Near-end crosstalkOccurs close to the transmitting end of the cable, most commonly near the terminating connector and causes loss of signal.
Far-end crosstalkOccurs at the other end of the cable with respect to the transmitter causing the interference and causes loss of signal.

Common Layer 2 Troubleshooting Issues

Question Answer
Power failureEthernet switches are often not connected to a redundant power source (for example, an electrical outlet with a generator backup), in part due to the widely dispersed installation locations throughout a building (for example, in wiring closets or in a mechanical room). As a result, you might want to equip your Ethernet switches with an uninterruptable power supply (UPS). See Lesson 12, "Network Security," for a comparison of a UPS code standby power supply (SPS).
Bad moduleA modular switch gives you the flexibility to connect a variety of media types and speeds to the switch through the use of different modules. Examples of these modules include gigabit interface converter (GBIC) and small form-factor pluggable (SFP) modules. These modular interfaces can be swapped out during your troubleshooting, as opposed to swapping out an entire switch.
Layer 2 loopLesson 4 discussed issues resulting from Layer 2 loops, including MAC address table corruption and broadcast storms. You also read about how to mitigate these issues with Spanning Tree Protocol (STP). However, STP can fail (as illustrated by my personal example, which you read about at the beginning of this lesson). Or, STP might be functioning suboptimally, because a root bridge was automatically selected, rather than being specified, resulting in a suboptimal path. So, you should be able to examine your Ethernet switches, when troubleshooting, and determine the STP roles of your network's switches and switch ports.
Port configurationCommon settings for Ethernet switch ports include speed, duplex, and MDIX. Mismatched parameters between devices could result in slow communication (in the case of a duplex mismatch) or in no communication (in the case of a speed mismatch or incorrect MDIX settings).
VLANIn Lesson 4, you read about virtual LANs (VLANs), which were broadcast configuration domains and represented a single subnet Several troubleshooting issues could result from a VLAN misconfiguration on an Ethernet switch. Keep in mind that all devices belonging to the same VLAN should be assigned IP addresses in the same subnet. Also, if you want traffic to flow between VLANs, that traffic has to be routed.

Common Layer 3 Troubleshooting Issues

Question Answer
Mismatched MTURouter interfaces have a parameter called the maximum transmission unit (MTU) that defines the largest packet size the interface will forward. For example, a 1500-byte packet could not be forwarded via a router interface with an MTU of 1470 bytes. A router attempts to fragment a packet that is too big unless the packet has its don't fragment (DF) bit set. If a packet exceeds an interface's MTU and has its DF bit set, the router drops the packet. Normally, the router responds to the sender with an ICMP message indicating why the packet was dropped. However, if a router is configured to not respond to such a condition by sending an Internet Control Message Protocol (ICMP) message, the packet is dropped without the sender being notified. Such a router is called a black-hole router. You can use the traceroute utility (as described in Lesson 10, "Command-Line Tools") to help locate a black-hole router.
Incorrect subnet maskWhen one host attempts to communicate with another host on the same subnet, the sending host sends an ARP request in an attempt to determine the MAC address of the destination host, rather than forwarding traffic to the sending host's default gateway. Therefore, if a host has an incorrect subnet mask, it could incorrectly conclude that another host is on its local subnet, when in reality, the other host is on a remote subnet. As a result, the remote host is unreachable from the perspective of the sending host.
Incorrect default gatewayIf a host has an incorrect default gateway configuration, traffic from that host is not forwarded off that host's local subnet
Duplicate IP addressHosts on a subnet should have unique IP addresses. If two hosts are configured with the same IP address, unpredictable traffic patterns for those hosts can occur.
Incorrect DNS configurationBecause hosts often use Domain Name System (DNS) to resolve domain names to IP addresses, if a host has an incorrect DNS configuration, that host will be unable to, for example, browse the Internet using domain names (as opposed to IP addresses).

Common Wireless Troubleshooting Issues

Question Answer
RFIWireless communication can be interrupted because of radio frequency interference (RFI). Common RFI sources that impact wireless networks include 2.4-GHz cordless phones, microwave ovens, baby monitors, and game consoles.
Signal strengthThe received signal strength indicator (RSSI) value measures the power of a wireless signal. An RSSI value varies based on distance from a wireless antenna and physical objects interfering with line-of-sight communication with a wireless antenna (for example, drywall, metal file cabinets, and elevator shafts). Some wireless networks automatically drop their wireless transmission rate when an RSSI value drops below a certain value.
Misconfiguration of wireless parametersA variety of wireless parameters must match between a wireless client and a wireless access point (AP) for communication to occur. For example, the client needs to be using a wireless standard supported by the wireless AP (for example, IEEE 802.11a/b/g/n). Wireless channels must also match. However, wireless clients usually automatically set their channel based on the wireless AP's channel. Encryption standards must match. For example, a wireless client using WPA would not successfully communicate with a wireless AP using WPA2. In addition, the service set identifier (SSID) of a wireless AP must be selected by the wireless client. In many cases, a wireless AP broadcasts its SSID, and a wireless client can select that SSID from a listing of visible SSIDs. In other cases, a wireless AP does not broadcast its SSID, thus requiring a wireless client to have a matching SSID manually configured.
LatencyWireless networks can experience more delay than their wired counterparts. One reason for the increased delay is the use of carrier sense multiple access collision avoidance (CSMA/CA) in WLANs, which introduces a random delay before transmitting data, in an attempt to avoid collisions. Another, yet similar, reason for the increased delay is the fact that all wireless devices associated with a single wireless AP are in the same collision domain, introducing the possibility of collisions (retransmissions), which can increase delay.
Multiple paths of propagationAn electromagnetic waveform cannot pass through a perfect conductor. Admittedly, perfect conductors do not exist in most office environments. However, very good conductors, such as metal file cabinets, are commonplace in offices. As a result, if the waveform of a wireless transmission encounters one of these conductive objects, most of the signal bounces off the object creating multiple paths (modes) of propagation. These multiple modes of propagation can cause data (specifically, bits) to arrive at uneven intervals, possibly corrupting data. This problem is similar to multimode delay distortion, which is seen in multimode fiber-optic cabling.
Incorrect AP placementWireless APs should be strategically located in a building to provide sufficient coverage to all desired coverage areas. However, the coverage areas of wireless APs using overlapping channels should not overlap. To maintain coverage between coverage areas, you should have overlapping coverage areas among wireless APs using nonoverlapping channels (for example, channels 1, 6, and 11 for wireless networks using the 2.4-GHz band of frequencies). A common design recommendation is that overlapping coverage areas (using nonoverlapping channels) should have an overlap of approximately 10 percent to 15 percent.


Question Answer
Black-hole routerA router that drops packets that cannot be fragmented and are exceeding the MTU size of an interface without notifying the sender.
Decibel (dB) lossA loss of signal power. If a transmission’s dB loss is too great, the transmission cannot be properly interpreted by the intended recipient.
Maximum transmission unit (MTU)The largest packet size supported on an interface.
OpenA broken strand of copper that prevents current from flowing through a circuit.
ShortA short occurs when two copper connectors touch each other, resulting in current flowing through that short rather than the attached electrical circuit, because the short has lower resistance.
Trouble ticketA problem report explaining the details of an issue being experienced in a network.