[CCNA,CCNP,CCIE,국제공인자격증,국제자격증]CCNP - Switching Exam Certification Guide8

[사쿠라]

 

추천싸이트: 즐겨찾기 해놓으면 도움이 되는 사이트

추천 까페 : 네이버 네트워크 무료 강좌 추천 까페

카페운영자에게 메신져로 물어보기

네이트온 : assaman@lycos.co.kr

MSN : superljk@hotmail.com

무료상담 : 080-384-3233

72 Chapter 3: Basic Switch and Port Configuration

Table 3-2 Cabling Specifications for Fast Ethernet

Technology Wiring Type Pairs Cable Length
100BaseTX EIA/TIA Category 5 UTP 2 100 m
100BaseT2 EIA/TIA Category 3,4,5 UTP 2 100 m
100BaseT4 EIA/TIA Category 3,4,5 UTP 4 100 m
100BaseFX Multimode fiber (MMF) 1 400 m half duplex62.5 micron core, 125 micron outer or
cladding (62.5/125) 2000 m full duplex
Single-mode fiber (SMF) 1 10 km

Full-Duplex Fast Ethernet

As with traditional Ethernet, the natural progression to improve performance is to use full-
duplex operation. Fast Ethernet can provide 100 Mbps in each direction on a switched
connection, for 200 Mbps total throughput. This throughput is only possible when a
workstation or server is directly connected to a switch port, or when two switches directly
connect to each other.

The Fast Ethernet specification also offers backward compatibility to support traditional 10
Mbps Ethernet. To provide this support, two devices at each end of a network connection can
automatically negotiate link capabilities so that they both can operate at a maximum common
level. This negotiation involves the detection and selection of the highest physical layer
technology (available bandwidth) and half-duplex or full-duplex operation. Even if one of the
devices uses a fixed configuration, the other device can detect this and match the capabilities.

Autonegotiation uses the priorities shown in Table 3-3 for each mode of Ethernet to determine
which technology to agree upon. If both devices can support more than one technology, then
the technology with the highest priority will be used. For example, if two devices can support
both 10BaseT and 100BaseTX, both devices will use the higher priority 100BaseTX mode.

Table 3-3 Autonegotiation Selection Priorities

Priority Ethernet Mode
7 100BaseT2 (full duplex)
6 100BaseT2 (half duplex)
5 100BaseTX (full duplex)
4 100BaseT4
3 100BaseTX
2 10BaseT (full duplex)
1 10BaseT

Desktop Connectivity with Ethernet 73

NOTE To assure proper configuration at both ends of a link, Cisco recommends that the appropriate
values for transmission speed and duplex mode be manually configured on switch ports.

Cisco provides one additional capability to Fast Ethernet, which allows several Fast Ethernet
links to be bundled together for increased throughput. Fast EtherChannel (FEC) allows two to
eight full-duplex Fast Ethernet links to act as a single physical link, for 400- to 1600-Mbps
bandwidth. This technology is described in greater detail in Chapter 5, “Redundant Switch
Links.”

For further reading about Fast Ethernet technology, refer to Cisco’s web site: www.cisco.com/
warp/public/cc/so/neso/lnso/lnmnso/feth_tc.htm

Gigabit Ethernet

Fast Ethernet can be scaled by an additional order of magnitude with the use of Gigabit Ethernet
(which supports 1,000 Mbps or 1 Gbps) using the same IEEE 802.3 Ethernet frame format as
before. This scalability allows network designers and managers to leverage existing knowledge
and technologies to install, migrate, manage, and maintain Gigabit Ethernet networks.

However, the physical layer has been modified to increase data transmission speeds. Two
technologies were merged together to gain the benefits of each: The IEEE 802.3 Ethernet
standard and the American National Standards Institute (ANSI) X3T11 FibreChannel. IEEE

802.3 provided the foundation of frame format, CSMA/CD, full duplex, and other
characteristics of Ethernet. FibreChannel provided a base of high-speed ASICs, optical
components, and encoding/decoding and serialization mechanisms. The resulting protocol is
termed IEEE 802.3z Gigabit Ethernet.
Gigabit Ethernet supports several cabling types, referred to as 1000BaseX. Table 3-4 lists the
cabling specifications for each type.

In a campus network, Gigabit Ethernet can be used in the switch block, the core block, and in
the server block. In the switch block, it is used to connect access layer switches to distribution
layer switches. In the core, it connects the distribution layer to the core switches, and also
interconnects the core devices. For a server block, a Gigabit Ethernet switch in the server block
can provide high-speed connections to individual servers.

74 Chapter 3: Basic Switch and Port Configuration

Table 3-4 Gigabit Ethernet Cabling and Distance Limitations

GE Type Wiring Type Pairs Cable Length
1000BaseCX Shielded Twisted Pair (STP) 1 25 m
1000BaseT EIA/TIA Category 5 UTP 4 100 m
1000BaseSX Multimode fiber (MMF) with 62.5 micron core; 1 275 m
850 nm laser
MMF with 50 micron core; 1300 nm laser 1 550 m
1000BaseLX/LH MMF with 62.5 micron core; 1300 nm laser 1 550 m
Single-mode fiber (SMF) with 50 micron core; 1 550 m
1300 nm laser
SMF with 9 micron core; 1300 nm laser 1 10 km
1000BaseZX SMF with 9 micron core; 1550 nm laser 1 70 km
SMF with 8 micron core; 1550 nm laser 1 100 km

Finally, Cisco has extended the concept of Fast EtherChannel to bundle several Gigabit
Ethernet links to act as a single physical connection. Gigabit EtherChannel (GEC) allows
two to eight full-duplex Gigabit Ethernet connections to be aggregated, for up to 16 Gbps
throughput. Port aggregation and the EtherChannel technology are described further in
Chapter 5.

NOTE The Gigabit Ethernet Alliance offers further reading about Gigabit Ethernet, and its operation,
migration, and standards. Refer to the web site: www.gigabit-ethernet.org

Desktop Connectivity with Token Ring

Token Ring is also a LAN technology that provides shared media access to many connected
stations. Rather than sharing a common bus or “wire” as Ethernet does, Token Ring stations are
arranged in a ring, in a daisy-chain fashion. A token is passed from station to station around the
ring, giving the current token holder permission to transmit a frame onto the ring. Once the
frame is sent, it is passed around the ring until it is received again by the source. The sending
station is responsible for removing the frame from the ring and for introducing a new token to
the next neighboring station.

Notice that only one station can transmit at a given time.the one with the token. This
restriction prevents a Token Ring network from ever becoming a collision domain. Stations can
expect to receive the token at regular intervals as it circulates the ring. This feature makes Token
Ring deterministic and useful for delay sensitive protocols. Frames can be sent to a broadcast

Desktop Connectivity with Token Ring 75

MAC address, like Ethernet, causing all stations on the ring to listen. Therefore, a token ring is
a broadcast domain.

A Token Ring network offers a bandwidth of 4 Mbps or 16 Mbps. At the higher rate, stations
are allowed to introduce a new token as soon as they finish transmitting a frame. This early
token release increases efficiency by letting more than one station transmit a frame during the
original token’s round trip. One station is elected to be the ring monitor, to provide recovery
from runaway frames or tokens. The ring monitor will remove frames that have circled the ring
once, if no other station removes them.

Traditional Token Ring networks use multistation access units (MSAUs) to provide connectivity
between end user stations. MSAUs have several ports that a station can connect to, with either
a B connector for Type 2 cabling or an RJ-45 connector for Category 5 UTP cabling. Internally,
the MSAU provides station-to-station connections to form a ring segment. The Ring-In and
Ring-Out connectors of a MSAU can be chained to other MSAUs to form a complete ring
topology.

Token Ring Bridging

To form larger networks, Token Rings are interconnected with bridges. Although a transparent
bridge (or one that forwards frames based solely on MAC addresses) can be used, IBM
designed and introduced Token Ring differently. Source-route bridges are used to forward
frames between rings, based on a predetermined path. The source station includes the exact
ring-and-bridge path within the frame so that specific bridges will forward the frame to the
appropriate rings. Rings must be uniquely numbered and identified with the campus network,
with a number between 1 and 4095. Bridges, however, do not have to be unique across the
network, as long as two bridges with the same number do not connect to the same ring. Bridges
are numbered 1 through 15.

The steps to determine the path a frame should take are as follows:

Step 1 The source station first sends a test frame to see if the destination
is on the local ring. If the destination responds, the source knows
that it is local. If there is no response, the source station will send
an all routes explorer (ARE) frame, which will cause all bridges to
forward the frame to all rings.
Step 2 Within the frame is a routing information field (RIF). The RIF
carries a record of bridges and rings traversed along the way. As
the frame is forwarded, each bridge will append its bridge number
and the next ring number to the RIF in the frame.
Step 3 The destination will then reply to each ARE frame it receives, so
that the source will receive a confirmation of every possible path
to the destination.

76 Chapter 3: Basic Switch and Port Configuration

For future transmissions, the source can choose the path it thinks is best (quickest response,
least number of bridge hops, largest maximum transmission unit (MTU), combination of
factors, and so forth). These frames will contain the exact path desired by the source station, in
the form of a RIF.

In certain scenarios, hybrid bridging can be provided. Sometimes, both source-route bridging
(SRB) and transparent bridging must occur. Here, source-route transparent bridging (SRT)
forwards a frame according to a RIF, if present, or according to MAC address tables if it finds
no RIF.

As in Ethernet switching, Token Rings can also be segmented by dividing a ring across several
switch ports. While this feature increases the available bandwidth on a ring segment, it requires
more in-depth forwarding decisions. Token ring switching, or more properly termed source-
route switching, forwards frames according to a combination of MAC addresses and RIF
contents.

Source-route switching differs from other forms of bridging in that it only looks at the RIF and
never updates or adds to the RIF. Instead, the switch learns route descriptors, or the ring/bridge
combinations that specify the next-hop destinations from incoming frames. The source-route
switch then associates the route descriptors and MAC addresses (if needed) with outbound
ports closest to the destination. When subsequent frames are received on other ports, the route
descriptor is quickly indexed to lookup the outbound port.

In this fashion, source-route switching supports parallel source-route paths to destinations. The
number of MAC addresses to be learned is lessened, because route descriptors point to the next-
hop ports. The actual operation of source-route switching is much like virtual LANs with
Ethernet. For this reason, further discussion of source-route switching is presented in
Chapter 4, “VLANs and Trunking.”

Table 3-5 summarizes the attributes of each type of Token Ring connectivity and segmentation
method.

Table 3-5 Token Ring Segmentation Methods

Method
Forwarding
Decision
Frame
Modification Ring Numbering
Transparent bridging MAC address N/A
Source-route bridging RIF RIF Ring numbers must be unique
among bridge ports.
Source-route transparent
bridging
MAC address or RIF RIF Ring numbers must be unique
among bridge ports.
Source-route switching Route descriptor Ring numbers can be same
across switch ports (single ring
can be segmented on several
ports).

Connecting Switches 77

Connecting Switches

Switch deployment in a network involves two steps: physical connectivity and switch
configuration. This section describes the connections and cabling requirements for devices in a
switch block. Cable connections must be made to the console port of a switch in order to make
initial configurations. Physical connectivity between switches and end users involves cabling
for the various types of LAN ports.

Console Port Cables/Connectors

A terminal emulation program on a PC is usually required to interface with the console port on
a switch. Various types of console cables and console connectors are associated with each Cisco
switch family.

All Catalyst switch families use an RJ-45-to-RJ-45 rollover cable to make the console
connection between a PC (or terminal or modem) and the console port. A rollover cable is made
so that pin 1 on one RJ-45 connector goes to pin 8 on the other RJ-45 connector, pin 2 goes to
pin 7, and so forth. In other words, the cable remains flat while the two RJ-45 connectors point
in opposite directions.

To connect the PC end, the rollover cable plugs into an RJ-45 to DB-9 or DB-25 “Terminal”
adapter (or a DB-25 “Modem” adapter for a modem connection). At the switch end, the rollover
cable plugs directly into the RJ-45 jack of the console port. This situation is true for the Catalyst
1900, 2820, 2900, 3500, 2926G, 2948G, 4912G, 5000 Supervisor IIG/III/IIIG, and the 6000
switches.

On the Catalyst 4003, 5000 Supervisor I/II, and the 8500 switches, the rollover cable must
connect to an RJ-45 to DB-25 “Modem” adapter. These switches have a DB-25 console port
connector that is a female DCE.

Once the console port is cabled to the PC, terminal, or modem, a terminal emulation program
can be started or a user connection can be made. The console ports on all switch families require
an asynchronous serial connection at 9600 baud, 8 data bits, no parity, 1 stop bit, and no flow
control.

Ethernet Port Cables/Connectors

Catalyst switches support a variety of network connections, including all forms of Ethernet. In
addition, Catalyst switches support several types of cabling, including UTP and optical fiber.

On Catalyst 1900 and 2820 series switches, the Ethernet ports are fixed-speed with 12 or 24
10BaseT and one or two 100BaseTX or 100BaseFX ports. The 10BaseT ports can be connected
only to other 10BaseT-capable devices (including 10/100 autosensing devices), and the
100BaseX to other 100BaseX-capable devices. The 10BaseT and 100BaseTX ports use
Category 5 UTP cabling and RJ-45 connectors.

78 Chapter 3: Basic Switch and Port Configuration

The 100BaseFX ports use two-strand multimedia fiber (MMF) with SC connectors to provide
connectivity. The SC connectors on the fiber cables are square in shape. These connectors snap
in and out of the switch port connector as the connector is pushed in or pulled out. One fiber
strand is used as a transmit path and the other as a receive path. Therefore, the transmit fiber on
one switch device should connect to the receive fiber on the other end.

The remainder of the Catalyst switch families support 10/100 autosensing (using Fast Ethernet
autonegotiation) and Gigabit Ethernet. Switched 10/100 ports use RJ-45 connectors on
Category 5 UTP cabling to complete the connections. These ports can be connected to other
10BaseT, 100BaseTX, or 10/100 autosensing devices. UTP cabling is arranged so that RJ-45
pins 1,2 and 3,6 form two twisted pairs. These pairs are connected straight through to the far
end.

In order to connect two 10/100 switch ports back-to-back, as in an access layer to distribution
layer link, a Category 5 UTP crossover cable must be used. In this case, RJ-45 pins 1,2 and 3,6
are still twisted pairs, but 1,2 on one end connect to 3,6 on the other end, and 3,6 on one end
connect to 1,2 on the other end.

NOTE Because UTP Ethernet connections use only pairs 1,2 and 3,6, some cable plant installers only
connect these pairs and leave the remaining two pair positions empty. While this move provides
Ethernet connectivity, it is not good practice for future needs. Instead, all four pairs of the RJ45 connector should be connected end-to-end. For example, a full four-pair UTP cable plant can
be used for either Ethernet or Token Ring connectivity, without rewiring. (Token Ring UTP
connections use pairs 3,6 and 4,5.) Also, to be compatible with the new IEEE 802.3ab standard
for Gigabit Ethernet over copper, all four pairs must be used end-to-end.

Gigabit Ethernet Port Cables/Connectors

Gigabit Ethernet connections take a different approach by providing modular connectivity
options. Catalyst switches with Gigabit Ethernet ports have standardized rectangular openings
that accept Gigabit Interface Converters (GBICs). GBIC modules provide the media personality
for the port so that various types of cables can be connected. In this way, the switch chassis is
completely modular and requires no major change to accept a new media type. Instead, the
appropriate GBIC module is hot-swappable and is plugged into the switch to support the new
media. GBICs are available for the following Gigabit Ethernet media:

. 1000BaseSX GBIC.short wavelength connectivity using SC fiber connectors and MMF
for distances up to 550 meters (1804 feet).
. 1000BaseLX/LH GBIC.long wavelength/long haul connectivity using SC fiber
connectors and either MMF or single-mode fiber (SMF); MMF can be used for distances
up to 550 meters (1804 feet) and SMF can be used for distances up to 10 km (32,810 feet).

Connecting Switches 79

. 1000BaseZX GBIC.extended distance connectivity using SC fiber connectors and
SMF; can be used for distances up to 70 km and even to 100 km when used with premium
grade SMF.
. GigaStack GBIC.uses a proprietary connector with a high-data-rate copper cable with
enhanced signal integrity and electromagnetic interference (EMI) performance; provides
a GBIC-to-GBIC connection between stacking Catalyst switches or between any two
Gigabit switch ports over a short distance.
CAUTION The fiber-based GBICs always have the receive fiber on the left SC connector and the transmit
fiber on the right SC connector. These GBICs could produce invisible laser radiation from the
transmit SC connector. Therefore, always keep unused SC connectors covered with the rubber
plugs and do not look directly into the SC connectors.
Figure 3-2 illustrates a fiber-based GBIC module and how one is installed in a Gigabit Ethernet
switch port.
Figure 3-2 Gigabit Interface Converter

Token Ring Port Cables/Connectors

Catalyst switches support UTP Token Ring connections. These ports operate at either 4 or 16
Mbps, in several half and full-duplex modes. RJ-45 connectors on Category 5 UTP cabling use
twisted pairs 3,6 and 4,5. These pairs are connected straight through to the far end.

80 Chapter 3: Basic Switch and Port Configuration

Switch Management

Cisco Catalyst switch devices can be configured to support many different requirements and
features. When a PC is connected to the serial console port, configuration is generally done with
a terminal emulator application on the PC. Further configurations can be performed through a
Telnet session across the LAN or through a web-based interface. These topics will be covered
in later sections.

Catalyst switches support one of two types of user interface for configuration: Cisco IOS-based
commands, and set-based, command-line interface (CLI) commands. The IOS-based
commands (found in Catalyst 1900/2820, 2900XL, and 3500XL) are similar to many IOS
commands used on Cisco routers. However, the CLI commands (found in 2926G, 4000, 5000
and 6000) use set and clear commands to change configuration parameters. Both types of user
interface are discussed in the sections that follow.

Identifying the Switch

All switches come from the factory with a default configuration and a default system name or
prompt. This name can be changed so that each switch in a campus network will have a unique
identity. This option can be useful when you are using Telnet to move from switch to switch in
a network.

Setting the Hostname/System Name on an IOS-Based Switch

To change the host or system name on an IOS-based user interface, enter the following
command in configuration mode:

Switch(config)# hostname hostname

The hostname is a string of 1 to 255 alphanumeric characters. As soon as this command is
executed, the system prompt will change to reflect the new hostname.

NOTE Configuration changes made on IOS-based switches apply only to the active running
configuration, stored in RAM. To make the changes permanent, in effect even after a power
cycle, remember to copy the switch configuration into the startup configuration, stored in
NVRAM. You can do this by using the copy running-config startup-config command.

Setting the Hostname/System Name on a CLI-Based Switch

To set the system name on a CLI-based user interface, the system prompt is changed with the
following command:

Switch(enable) set system name name-string

Switch Management 81

As soon as this command is executed, the system name and the prompt will change to reflect
the new value. This prompt is displayed at the beginning of every CLI line.

Passwords and User Access

Normally, a network device should be configured to secure it from unauthorized access.
Catalyst switches offer a simple form of security by setting passwords to restrict who can log
in to the user interface. Two levels of user access are available: regular login, or EXEC mode,
and enable login, or privileged mode. EXEC mode is the first level of access, which gives access
to the basic user interface through any line or the console port. The privileged mode requires a
second password and gives access to set or change switch operating parameters or
configurations.

Cisco provides various methods for providing device security and user authentication. Many of
these methods are more secure and robust than using the login passwords in Chapter 12,
“Controlling Access in the Campus Environment,” describes these features in greater detail.

Setting Login Passwords on an IOS-Based Switch

To set the login passwords on a Cisco IOS-based switch interface, enter the following
commands in global configuration mode:

Switch(config)# enable password level 1 password
Switch(config)# enable password level 15 password

Here, the EXEC mode password is set with a privilege level of one (1), while the enable
password is set with a privilege level of 15. The password is a string of four to eight
alphanumeric characters. Passwords on these switches are not case-sensitive.

To remove a password, use the no enable password level password command.

Setting Login Passwords on a CLI-Based Switch

Example 3-1 lists the commands you would enter in enable mode to set the login passwords on
a Cisco switch with a CLI-based user interface.

Example 3-1 Setting the Login Passwords on a Cisco Switch

Switch (enable) set password
Enter old password: oldpassword
Enter new password: newpassword
Retype new password: newpassword
Password changed.
Switch (enable) set enablepass
Enter old password: oldenablepassword
Enter new password: newenablepassword
Retype new password: newenablepassword
Password changed.
Switch (enable)