ISL and 802.1Q Compared

1

Brief Review of LAN Switching


Learning
—

The switch learns MAC addresses by examining the source MAC address of each
frame the bridge receives. By learning, the switch can make good forwarding choices in the
future.

Forwarding or filtering
—

The switch deci
des when to forward a frame or when to filter (not
forward) it based on the destination MAC address. The switch looks at the previously learned
MAC addresses in an address table to decide where to forward the frames.

Loop prevention
—

The switch creates a l
oop
-
free environment with other bridges by using
Spanning Tree Protocol (STP). Having physically redundant links helps LAN availability, and
STP prevents the switch logic from letting frames loop around the network indefinitely,
congesting the LAN.


Span
ning Tree Protocol

Characterization of Port


Spanning

Tree State

Description

All the root bridge’s ports

Forwarding

The root bridge is always the
designated bridge on all connected
segments.

Each nonroot bridge’s root port

Forwarding

The root port i
s the port receiving the
lowest
-
cost BPDU from the root.

Each LAN’s designated port

Forwarding

The bridge forwarding the lowest
-
cost BPDU onto the segment is the
designated bridge for that segment.

All other ports


Blocking

The port is not used for f
orwarding
frames, nor are any frames received
on these interfaces considered for
forwarding.


ISL and 802.1Q Compared


Function

ISL

802.1Q

Standards body that defines the protocol

Cisco
-
proprietary

IEEE

Encapsulates the original frame

Yes

No

Allows mu
ltiple spanning trees

Yes

No

Uses a native VLAN

No

Yes


VTP Modes

Function

Server
Mode

Client
Mode

Transparent
Mode

Originates VTP advertisements

Yes

No

No

Processes received advertisements and synchronizes VLAN
configuration information with other
switches

Yes

Yes

No

Forwards VTP advertisements received in a trunk

Yes

Yes

Yes

Saves VLAN configuration in NVRAM

Yes

No

Yes

Can create, modify, or delete VLANs using configuration
commands

Yes

No

Yes







2

2950 Trunk Configuration Options wi
th the
switchport mode
Command

Option

Description

access

Disables port trunk mode and does not even attempt to form a trunk on the interface.

trunk

Configures the port into permanent trunk mode and negotiates with the connected device to
decide whethe
r to use 802.1Q or ISL.

dynamic desirable

Triggers the port to negotiate the link from nontrunking to trunk mode. The port negotiates to
a trunk port if the connected device is either in the
trunk
,
dynamic desirable
, or
dynamic
auto
state. Otherwise, the

port becomes a nontrunk port.

dynamic auto

Lets a port become a trunk only if the connected device has the state set to
dynamic
desirable
or
trunk
.


Routing Protocol Terminology

Term

Definition

Routing protocol


A protocol whose purpose is to learn t
he available routes, place the best routes in the
routing table, and remove routes when they are no longer valid.

Exterior routing protocol


A routing protocol designed for use between two different organizations. These are
typically used between ISPs or
between a company and an ISP. For example, a
company would run BGP, an exterior routing protocol, between one of its routers and a
router inside an ISP.

Interior routing protocol


A routing protocol designed for use within a single organization. For examp
le, an entire
company might choose the IGRP routing protocol, which is an interior routing protocol.

Distance vector


The logic behind the behavior of some interior routing protocols, such as RIP and
IGRP.

Link state


The logic behind the behavior of som
e interior routing protocols, such as OSPF.

Balanced hybrid


The logic behind t he behavior of EIGRP, which is more like dist ance vect or t han link
st at e but is different from t hese t wo t ypes of rout ing prot ocols.

Di jkstra S hortest Path
Fi rst (S PF) al gori
thm


Magic mat h used by link
-
st at e prot ocols, such as OSPF, when t he rout ing t able is
calculat ed.

Di ffusi ng Update
Al gori thm (DUAL)


The process by which EIGRP rout ers collect ively calculat e t he rout es t o place in t he
rout ing t ables.

Convergence


The t im
e required for rout ers t o react t o changes in t he net work, removing bad rout es
and adding new, bet t er rout es so t hat t he current ly
-
best rout es are in all t he rout ers’
rout ing t ables.

Metri c


The numeric value t hat describes how good a part icular rout e is.

The lower t he value,
t he bet t er t he rout e.


Issues Rel at ed t o Di st ance Vect or Rout i ng Prot ocol s i n Net works wi t h Mul t i pl e Pat hs

Issue

Solution

Mul ti pl e routes to the same
subnet have equal metri cs

Implement at ion opt ions involve eit her using t he first
rout e learned or put t ing
mult iple rout es t o t he same subnet in t he rout ing t able.

Routi ng l oops occur because
updates pass each other over a
si ngl e l i nk

S pl i t hori zon
—
The rout ing prot ocol advert ises rout es out an int erface only if t hey
were not learned fr
om updat es ent ering t hat int erface.

S pl i t hori zon wi th poi son reverse
—
The rout ing prot ocol uses split
-
horizon rules
unless a rout e fails. In t hat case, t he rout e is advert ised out all int erfaces, including
t he int erface in which t he rout e was learned, but

wit h an infinit e
-
dist ance

met ric.

Routi ng l oops occur because of
routi ng i nformati on l oopi ng
through al ternati ve paths

Route poi soni ng
—
When a rout e t o a subnet fails, t he subnet is advert ised wit h an
infinit e
-
dist ance met ric. This t erm specifically appli
es t o rout es t hat are advert ised
when t he rout e is valid, whereas poison reverse refers t o rout es t hat are not
normally advert ised because of split horizon but t hat are advert ised wit h an infinit e
met ric when t he rout e fails.

Counti ng to i nfi ni ty

Hol d
-
do
wn timer
—
After finding out that a route to a subnet has failed, a router
waits a certain period of time before believing any other routing information about
that subnet.

Triggered updates
—
When a route fails, an update is sent immediately rather than
waiti
ng on the update timer to expire. Used in conjunction with route poisoning,
this ensures that all routers know of failed routes before any hold
-
down timers can
expire.




3

RIP and IGRP Feature Comparison

Feature

RIP (Default)


IGRP (Default)

Update timer

30 seconds

90 seconds

Metric

Hop count

Function of bandwidth and delay (the
default). Can include reliability, load,
and MTU.

Hold
-
down timer

180

280

Flash (triggered) updates

Yes

Yes

Mask sent in update

No


No

Infinite
-
metric value

16

4,29
4,967,295


Comparing Link
-
State and Distance Vector Protocols

Feature

Link State

Distance Vector

Convergence Time

Fast

Slow, mainly because of loop
-
avoidance
features

Loop Avoidance

Built into the protocol

Requires extra features such as split
hor
izon

Memory and CPU Requirements

Can be large; good design can
minimize

Low

Requires Design Effort for Larger
Networks

Yes


No

Public Standard or Proprietary

OSPF is public

RIP is publicly defined; IGRP is not


EIGRP Features Compared to OSPF and IG
RP

Feature

EIGRP

IGRP

OSPF

Discovers neighbors before exchanging routing information

Yes

No

Yes

Builds some form of topology table in addition to adding routes to the
routing table

Yes

No

Yes

Converges quickly

Yes

No

Yes

Uses metrics based o
n bandwidth and delay by default

Yes
*

Yes

No

Sends full routing information on every routing update cycle

No

Yes

No

Requires distance vector loop
-
avoidance features

No

Yes

No

Public standard

No

No

Yes




Route summarization
—
Route summarizatio
n reduces the size of the network’s routing

tables by causing a
number of more specific routes to be replaced with a single route that

includes all the IP addresses covered by
the subnets in the original routes.



Variable
-
length subnet masking
—
VLSM occurs w
hen more than one mask is used in a

single Class A, B, or
C network. Although route summarization causes more than one

mask to be used, requiring support for VLSM,
you can also simply design a network to

use multiple subnet masks.


Table 7
-
4
Interior IP Ro
uting Protocol VLSM Support

Routing Protocol

VLSM Support

Sends Mask/Prefix in
Routing Updates

Route Summarization
Support

RIP
-
1

No

No

No

IGRP

No

No

No

RIP
-
2

Yes

Yes

Yes

EIGRP

Yes

Yes

Yes

OSPF

Yes

Yes

Yes


The following list describes a g
eneralized process by which you can summarize a group of subnets into one
summary route. This process attempts to find the “best” summary that includes all subnets, as opposed to finding all
summary routes that include all subnets:


Step 1
Find the longest

part of the subnet numbers that are identical, moving left to right. (For our purposes,
consider this first part the “in common” part.)


4

Step 2
The summary route’s subnet number has the same value in the “in common” part of the summarized subnets
and binar
y 0s in the second part.

Step 3
The subnet mask for the summary route has binary 1s in the “in common” part and binary 0s in the rest of
the mask.

Step 4
Check your work by calculating the range of valid IP addresses implied by the new summary route,
compa
ring the range to the summarized subnets. The new summary should encompass all IP addresses in the
summarized subnets.


Table 7
-
5
Interior IP Routing Protocol: Classless or Classful?

Routing
Protocol

Classless

Sends Mask/Prefix in
Routing Updates

VLSM
Sup
port

Route Summarization
Support

RIP
-
1

No

No

No

No

IGRP


No

No

No

No

RIP
-
2

Yes

Yes

Yes

Yes

EIGRP

Yes

Yes

Yes

Yes

OSPF

Yes

Yes

Yes

Yes




Classless and classful routing protocols
—
With classful routing protocols, the routing protocol mu
st consider
class rules; classless routing protocols do not. Specifically, classful routing protocols must automatically
summarize routing information at network boundaries, meaning that they cannot support discontiguous
networks. Classful routing protocol
s also cannot support VLSM. Classless routing protocols can support
discontiguous networks, and support VLSM.



Classless and classful routing
—
With classful routing, the only time the default route is used is when a
packet’s destination Class A, B, or C netw
ork number is not in the routing table. With classless routing, the
default is used whenever the packet does not match a more specific route in the routing table.


Range of IP Addresses

Class of Networks

Number of Networks

10.0.0.0 to 10.255.255.255

A

1

172.16.0.0 to 172.31.255.255

B

16

192.168.0.0 to 192.168.255.255

C

256


NAT Addressing Terms

Term

Description

Inside local

In a typical NAT design, the term “inside” refers to an address used for a host inside an enterprise.
An inside local i
s the actual IP address assigned to a host in the private enterprise network. A more
descriptive term might be “inside private,” because when using RFC 1918 addresses in an
enterprise, the inside local represents the host inside the enterprise, and it is a

private RFC 1918
address.

Inside global

In a typical NAT design, the term “inside” refers to an address used for a host inside an enterprise.
NAT uses an inside global address to represent the inside host as the packet is sent through the
outside networ
k, typically the Internet. A NAT router changes the source IP address of a packet sent
by an inside host from an inside local address to an inside global address as the packet goes from the
inside to the outside network. A more descriptive term might be “i
nside public,” because when using
RFC 1918 addresses in an enterprise, the inside global represents the inside host with a public IP
address that can be used for routing in the public Internet.

Outside global

In a typical NAT design, the term “outside” r
efers to an address used for a host outside an
enterprise
—
in other words, in the Internet. An outside global is the actual IP address assigned to a
host that resides in the outside network, typically the Internet. A more descriptive term might be
“outside
public,” because the outside global represents the outside host with a public IP address that
can be used for routing in the public Internet.

Outside local

In a typical NAT design, the term “outside” refers to an address used for a host outside an
enterp
rise
—
in other words, in the Internet. NAT uses an outside local address to represent the
outside host as the packet is sent through the private enterprise network (inside network). A NAT
router changes a packet’s destination IP address, sent from an inside

host to the outside global
address, as the packet goes from the inside to the outside network. A more descriptive term might be
“outside private,” because when using RFC 1918 addresses in an enterprise, the outside local
represents the outside host with a

private IP address from RFC 1918.


5

ICMP Message Types

Message

Purpose

Destination Unreachable

Tells the source host that there is a problem delivering a packet.

Time Exceeded

The time it takes a packet to be delivered has expired, so the packet has b
een discarded.

Redirect

The router sending this message has received a packet for which another router has a better
route. The message tells the sender to use the better route.

Echo

Used by the
ping
command to verify connectivity.



Comparison of FTP
and TFTP

FTP

TFTP

Uses TCP

Uses UDP

Uses robust control commands

Uses simple control commands

Sends data over a TCP connection separate from control commands

Uses no connections because of UDP

Requires more memory and programming effort

Requires le
ss memory and programming effort