Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
The table below describes the different unshielded twisted pair (UTP)
cable types (categories). Ethernet specifications use the following pins (Tx is a pin used for
Type Connector Description transmitting and Rx is a pin used for receiving):
Phone RJ-11 Used to connect a PC to a phone jack in a wall • Pin 1: Tx+
cable outlet to establish a dial-up Internet connection. • Pin 2: Tx-
Has two pairs of twisted cable (a total of 4 wires). • Pin 3: Rx+
Cat 3 RJ-45 Designed for use with 10 megabit Ethernet or 16 • Pin 4: Unused
megabit token ring.
• Pin 5: Unused
Cat 5 RJ-45 Supports 100 megabit and 1 gigabit Ethernet and • Pin 6: Rx-
ATM networking.
• Pin 7: Unused
Cat 5e RJ-45 Similar to Cat 5 but provides better EMI protection.
Supports 1 and 10 gigabit Ethernet (gigabit • Pin 8: Unused
connections require the use of all four twisted
pairs). Coaxial Cable Facts
Coaxial cable is an older technology that is usually implemented with a
Cat 6 RJ-45 Supports high-bandwidth, broadband bus topology. It is not suitable for ring or star topologies because the
communications. ends of the cable must be terminated. It is composed of two
Cat-6 RJ-45 Used for data transmissions. Supports up to 600 conductors, which share a common axis, within a single cable.
STP MHz and used in Ethernet, Fast Ethernet,
Gigabit Ethernet, Token Ring, and 155 Mbps ATM. Coaxial cable is built with the following components:
• Two concentric metallic conductors:
The table below describes the two types of connectors used with o The inner conductor, which carries data signals. It is
twisted pair cables. made of copper or copper coated with tin.
Connector Description o The mesh conductor is a second physical channel that
also grounds the cable. It is made of aluminum or copper
• Has 4 connectors coated tin.
RJ-11 • Supports up to 2 pairs of wires • The insulator, which surrounds the inner conductor, keeps the
• Uses a locking tab to keep connector secure in signal separated from the mesh conductor. It is made of PVC
outlet plastic.
• Used primarily for telephone wiring • The mesh conductor, which surrounds the insulator and grounds
the cable. It is made of aluminum or copper coated tin.
• Has 8 connectors • The PVC sheath, which is the cable encasement. It surrounds
RJ-45 • Supports up to 4 pairs of wires and protects the wire. It is made of PVC plastic.
• Uses a locking tab to keep connector secure in
outlet Coaxial cable has the following advantages and disadvantages:
• Used for Ethernet and some token ring connections Advantages • Highly resistant to EMI (electromagnetic
Each type of UTP cable can be substituted for any category below it, interference)
but never for a category above. For example, Cat 6 can be substituted • Highly resistant to physical damage
for a task requiring Cat 5e; however, neither Cat 5 nor Cat 3 should be
used for this particular task.
Disadvantages • Expensive
• Inflexible construction (difficult to install)
Making Cable Facts • Unsupported by newer networking standards
Twisted pair cables remain one of the primary ways that computers
connect to a network. Computers connect to the network through a hub
or switch with a straight-through cable. Computers can connect directly The table below describes the different coaxial cable grades.
to one another using a crossover cable. Resistance
Grade Uses
Rating
The table below illustrates both straight-through and crossover cable
configurations. • 10Base2 Ethernet networking (also known
Cable Description RG-58 as Thinnet) 50 ohms
There are two standards for creating
straight-through cables: • Limited to 185 meters
• T568A--To use this standard, RG-59 Cable TV and cable networking 75 ohms
arrange the wires from pins 1 to 8 RG-6 Satellite TV 75 ohms
in each connector in the following
order: GW, G, OW, B, BW, O, • 10Base5 Ethernet networking (also known
BrW, Br. as Thicknet)
RG-8 50 ohms
• T568B--To use this standard,
arrange the wires from pins 1 to 8 • Limited to 500 meters
in each connector in the following
order: OW, O, GW, B, BW, G,
BrW, Br.
Multi-mode and single mode fiber cables are distinct from each other • Used with single and multi-
and not interchangeable. mode cabling
• Composed of a plastic
The table below describes multi-mode and single mode fiber cables. connector with a locking tab
Type Description Mechanical Transfer-Registered • Uses metal guide pins to
Jack Connector (MT-RJ) ensure it is properly aligned
• Transfers data through the core using a single light ray (the
• A single connector with one
ray is also called a mode)
end holds both cables
• The core diameter is 8 to 10 microns and 125 micron
• Uses a ceramic ferrule to
Single cladding
insure proper core
Mode • Supports a large amount of data alignment and prevent light
• Cable lengths can extend a great distance ray deflection
• Full duplex
802.3z
• Connectors are: MT-RJ and LC
The Gigabit Ethernet SX feature is a TCP/IP attachment that offers
• Transfers data through the core using multiple light rays high-speed data transfers using multi-mode optical fiber. 1000BASE-
SX
Multi-
• The core diameter is 50 to 100 microns and 125 micron
cladding USB and FireWire Facts
mode
• Cable lengths are limited in distance (550 meters) You can create a network connection between two PCs by plugging a
• Full duplex USB cable into their USB ports. You can also use software that allows
you to connect multiple PCs through a USB hub. USB is a serial
communication specification. There are two USB versions:
Fiber optic cabling uses the following connector types:
• USB 1.0 runs at 12 megabits per second.
Type Description
• USB 2.0 runs at 480 megabits per second.
Straight Tip (ST) Connector • Used with single and multi-mode
cabling The table below describes the three types of USB connectors.
• Keyed, bayonet-type connector Connector Description
Network Adapter Facts Active VS’s Passive hubs – Active hubs regenerate the
A network adapter connects a host to the network medium. Some signals while passive hubs do not. Active hubs are generally
computers, like laptops, come with built-in network adapters. Other more expensive than passive hubs. Both passive and active
computers use NICs (network interface cards) that plug in to the hubs work at the physical layer (layer1) of the OSI model.
system's expansion slots or which are external to the computer and Use crossover cables and not patch cables to connect
connect through an existing computer port. A common network multiple hubs together.
interface card is one used on an Ethernet network. Switches provide functionality similar to hubs, but typically on
a larger scale and with higher performance (A switch offers
The table below describes the components of an Ethernet NIC. guaranteed bandwidth to each port). Unlike a hub, a switch
Component Description forwards frames only to the intended host, not every host
connected to the switch.
Media These connect the network interface and host to the A switch builds a database based on MAC addresses to
connectors network media. make forwarding decisions.
This visually indicates the network connection status. • The process begins by examining the source address of
Link indicator Green generally indicates a good connection, and red or an incoming packet. If the source address is not in the
an unlit diode indicates a bad connection. forwarding database, an entry for the address is made
A NIC's transceiver is responsible for transmitting and in the database. The port it came in on is also recorded.
receiving network communications. To send signals to the • The destination address is then examined.
network, it converts digital data from a PC to digital Switch o If the destination address is not in the database,
signals. The type of signal the transceiver sends depends the packet is sent out all ports except for the one
Transceiver
on the type of network. A fiber optic NIC sends light on which it was received.
signals; an Ethernet NIC sends electronic signals. To o If the destination address is in the database, the
receive signals, the transceiver converts digital signals
packet is forwarded to the appropriate port if the
from the network to digital data for the PC.
port is different than the one on which it was
MAC Address The MAC address is a unique hexadecimal identifier received.
burned into the ROM (physically assigned address) of o Broadcast packets are forwarded to all ports
every network interface. except the one on which they were received.
• The MAC address is a 12-digit hexadecimal number
(each number ranges from 0-9 or A-F). Eventually, a switch learns the location of all devices on the
• The address is often written as 00-B0-D0-06-BC-AC network. Incoming frames are then sent directly to the switch
or 00B0.D006.BCAC, although dashes, periods, and port to which a specific host is connected.
colons can be used to divide the MAC address parts. Bridge Bridges connect separate media segments (networks) that
• The MAC address is guaranteed unique through use the same protocol. Like a switch, bridges use MAC
design. The first half (first 6 digits) of the MAC addresses to determine a frame's destination and to build a
address is assigned to each manufacturer. The table of device addresses and their corresponding segments.
manufacturer determines the rest of the address, This also allows a bridge to prevent messages within a
assigning a unique value which identifies the host media segment from crossing over to another segment. This
address. A manufacturer that uses all the addresses keeps the network from wasting bandwidth by eliminating
unnecessary traffic between segments. If a bridge does not
Note: There are also some switches (called Layer 3 switches) that
Ethernet Facts
have built-in router functionality. These switches examine the logical
Ethernet is the most popular networking architecture for LANs. It offers
network address (instead of the MAC address) to switch packets
high performance at a low cost and is easy to install and manage.
between networks.
The following table describes details of a wireless networking Note: Many wireless access points include ports (or
architecture. hubs, switches, or routers) to connect the wireless
network to the wired portion of the network.
Characteristic Description
FHSS uses a narrow frequency band Wireless Standards
and 'hops' data signals in a predictable Radio frequency wireless networking standards are specified by
sequence from frequency to frequency various IEEE 802.11 committees.
over a wide band of frequencies.
Because FHSS hops between Wireless networking technologies
frequencies, it can avoid interference
Frequency Data Transmission
on one cable as it shifts to another. Standard Frequency Topology Range
Hopping Speed Type
Hopping between frequencies
Spread
increases transmission security by
Spectrum IEEE
making eavesdropping and data
Signaling (FHSS) 802.11 2 Mbps 2.4 GHz FHSS or DSSS Point-point 30 m
capture more difficult.
Method Legacy
Because FHSS shifts automatically
between frequencies, it can avoid
IEEE
interference that may be on a single 11
802.11b 2.4 GHz DSSS with CCK Point-point 30 m
frequency. FHSS applies to 802.11 is WiFi
Mbps
still in use with Bluetooth.
Direct- The transmitter breaks data into pieces IEEE
54
Sequence and sends the pieces across multiple 802.11a 5 GHz OFDM Point-point 30 m
Mbps
Spread frequencies in a defined range. DSSS WiFi
Spectrum is more susceptible to interference and
(DSSS) less secure then FHSS. >20 Mbps:
IEEE
54 OFDM, <20
802.11g 2.4 GHz Point-point 30 m
• Works in peer-to-peer mode WiFi
Mbps Mbps: DSSS
without a WAP (the wireless NICs with CCK
in each host communicate directly
with one another) IEEE
540
802.11n 2.4 GHz MIMO Point-point 50 m
Ad hoc
• Uses a physical mesh topology WiFi
Mbps
• Cheap and easy to set up but
cannot handle more than four Bluetooth 2 Mbps 2.45 GHz FHSS Scatternet 10 m
hosts
• Requires special modifications to 100
reach wired networks kbps~ 100 GHz ~ Point-point
Infrared Baseband 1m
Topology 16 1000 THz LOS
• Employs a WAP that functions like Mbps
a hub on an Ethernet network
IEEE 2 GHz ~ 11
• Uses a physical star topology 802.16
75
GHz, 66 BPSK
Point-point
30 km
Mbps Cellular
• You can easily add hosts without WiMax GHz
Infrastructure increasing administrative efforts
(scalable)
The actual range depends on several factors such as; the greater the
• Allows you to connect easily to a distance, the weaker the signal. As the distance between devices
wired network increases, the data transfer rate drops. The distances listed here are
• Requires more planning to rough maximums assuming no obstructions. For communications at
implement effectively the stated speed in a typical environment (with one or two walls), the
actual distance would be roughly half of the maximums.
Wireless networks use Carrier Sense Media
Access/Collision Avoidance (CSMA/CA) to control
Note: Some newer 802.11g devices can use multiple channels (dual-
media access and avoid (rather than detect) collisions.
band) to effectively double the data transfer rate to 108 Mbps.
Collision avoidance involves implementing the following
However, dual-band wireless is especially susceptible from
practices:
interference from other wireless devices (such as phones). Wireless
Media Access
• If a host detects traffic on the network, it equipment does not come with enabled security features. You must
experiences a longer back-off time than hosts on a enable the types of security you want to implement.
wired network before attempting to transmit again.
• Every transmission must be acknowledged. As Wireless antennas are either Omni-directional or directional. Omni
every frame is acknowledged by the receiving host, directional antennas provide a 360 degree dispersed wave pattern
other hosts receive a message indicating that they while directional antennas range is directional towards the transmitter.
must wait to transmit. Omni-directional antenna work best outdoors in open areas and are
“Vertical” antennas.
Devices Devices on a wireless network include:
• A wireless NIC for sending and receiving signals. The table below describes common wireless security features.
• A wireless access point (WAP) is the equivalent of Feature Description
an Ethernet hub. The wireless NICs connect to the
SSID (Service The SSID is used to group several wireless devices and
WAP, and the WAP manages network
Set Access Points as part of the same network and to
communication.
Application Layer 7: The application refers to communication services Examples include: TCP, UDP and SPX.
to applications and is the interface between the network and the
application. This layer is responsible for providing network services Network Layer 3: The Network layer describes how data is routed
such as: file services, print services, and messaging services. across networks and on to the destination. Network layer functions
include:
Application layer functions typically include identifying communication
partners, determining resource availability, and synchronizing • Routing occurs at this layer.
communication. When identifying communication partners, the • Maintaining addresses of neighboring routers.
application layer determines the identity and availability of • Maintaining a list of known networks.
communication partners for an application with data to transmit. When
determining resource availability, the application layer must decide • Data is placed into packets.
whether sufficient network resources for the requested communication • Determining the next network point to which data should be sent.
exist. In synchronizing communication, all communication between Routers use a routing protocol to take into account various factors
applications requires cooperation that is managed by the application such as the number of hops in the path, link speed, and link
layer. reliability to select the optimal path for data.
Examples include: Telnet, HTTP, HTTPS, FTP, TFTP, SFTP, Internet Packets forwarded from the Transport to the Network layer become
browsers, POP3, NTP, NNTP, DNS, NFS, SSH, SNMP, SMTP datagrams and network-specific (routing) information is added. Network
gateways, IMAP4, LDAP, LPR, X.400 mail and FTAM. layer protocols then ensure that the data arrives at the intended
destinations. Some error control and flow control is performed at this
Presentation Layer 6: This layer provides independence from level. The following protocols operate at this layer: IP,.
differences in data representation by translating from application to
network format, and vice versa. The presentation layer works to The Network layer uses logical addresses for identifying hosts and
transform data into the form that the application layer can accept. This making routing decisions. The type of addresses used are determined
layer formats and encrypts data to be sent across a network, providing by the protocol.
freedom from compatibility problems. It is sometimes called the syntax • IP uses IP addresses that identify both the logical network and
layer. host addresses
Logical Link Control: The Logical Link Control (LLC) layer provides an
interface between the MAC layer and upper-layer protocols. LLC
protocols are defined by the IEEE 802.2 committee. The LLC sublayer
is responsible for:
• Maintaining orderly delivery of frames through sequencing
• Controlling the flow or rate of transmissions
• Ensuring error-free reception of messages by retransmitting
• Converting data into an acceptable form for the upper layers
• Removing framing information from the packet and forwarding the
message to the Network layer
• Provide a way for upper layers of the OSI model to use any MAC
layer protocol
Defining Service Access Points (SAPs) by tracking and managing When you configure a computer to connect to the network, you must
different protocols. configure the appropriate protocols so that the computer can
communicate with other hosts on the network. Often the choice of the
Examples include: switches, bridges, wireless access points protocol suite to use depends on the network operating system and the
Physical Layer 1: Responsible for placing the network data on the wire, services that must be provided to network clients.
by changing binary data into electrical pulses on the physical medium
(also known as ‘signal encoding’). The physical topology is defined at Be aware of the following facts regarding protocol suite support and
this level. features:
Examples include: Hubs, Repeaters, NICs, Transceiver, Token Ring • Virtually all operating systems today provide native (built-in)
MAU support for TCP/IP.
• Most older versions of some operating systems used a different
protocol as the default protocol suite. For example, older NetWare
servers used IPX/SPX, while Mac OS systems used AppleTalk.
• Older operating systems without native TCP/IP support enabled
TCP/IP communications by either installing the protocol stack or
through a process known as encapsulation or tunneling. With this
process, non-TCP/IP packets are re-packaged as TCP/IP packets
at the sending device. The receiving device strips off the TCP/IP
headers to reveal the original packets.
• Addressing as referred to in this table refers to logical host and
network addresses (addresses used at the Network layer). Do not
confuse logical addresses with physical (MAC) addresses. Be
aware, however, that some protocols (such as IPX/SPX) use the
MAC address as the logical host address.
• IPX/SPX must also be configured with a Data Link layer frame
type. The frame type specifies the format of the frames.
IPv6 Facts
The current IP addressing standard, version 4, will eventually run out
of unique addresses, so, a new system is being developed. It is named
IP version 6 or IPv6. You should know the following about IPv6:
Subnetting Facts
Subnetting is the process of dividing a large network into smaller • Full implementation should be around 2015.
networks. When you subnet a network, each network segment (called • The new version will dramatically increase address availability:
a subnet) has a different network address (also called a subnet o IPv6 will provide about 3.4 x 1038 globally unique addresses.
address). In practice, the terms network and subnet are used o IPv6 provides 79,228,162,514,264,337,593,543,950,336
interchangeably to describe a physical network segment with a unique
times as many addresses as IPv4.
network address.
• The new IP address is a 128-bit binary number. A sample IPv6 IP
From a physical standpoint, subnetting is necessary because all address looks like:
network architectures have a limit on the number of hosts allowed on a 35BC:FA77:4898:DAFC:200C:FBBC:A007:8973.
single network segment. As your network grows, you will need to o Bits are divided into eight groups of 16-bit hexadecimal
create subnets (physical networks) to: sections.
• Increase the number of devices that can be added to the LAN (to o Each group is represented as a hexadecimal number
overcome the architecture limits) between 0 and FFFF.
• Reduce the number of devices on a single subnet to reduce o Hex values are separated by colons.
congestion and collisions o Leading zeros can be omitted in each section.
• Reduce the processing load placed on computers and routers o Addresses with consecutive zeros can be expressed more
• Combine networks with different media types within the same concisely by substituting a double-colon for the group of
internetwork (subnets cannot be used to combine networks of zeros. For example:
different media type on to the same subnet) FEC0:0:0:0:78CD:1283:F398:23AB
FEC0::78CD:1283:F398:23AB (concise form)
Subnetting is also used to efficiently use the available IP addresses.
For example, an organization with a class A network ID is allocated • IPv6 addresses are 4 times as large as IPv4 addresses (without
enough addresses for 16,777,214 hosts. If the organization actually optional fields, addresses are only twice as large).
uses only 10,000,000 host IDs, over 6 million IP addresses are not • The network ID part of the address is hierarchical and includes
being used. Subnetting provides a way to break the single class A identifiers for various levels of the network from top level network
network ID into multiple network IDs. segments down to an organization's specific network segment
• Subnetting uses custom rather than the default subnet masks. IDs.
For example, instead of using 255.0.0.0 with a Class A address, • IPv6 allows the addition of header extensions. Flexible packet
you might use 255.255.0.0 instead. headers can:
• Using custom subnet masks is often called classless addressing o Include optional fields and other extensions
because the subnet mask cannot be inferred simply from the o Increase IPv6 from 2 times to 4 times larger than IPv4,
class of a given IP address. The address class is ignored and the through the addition of optional fields
mask is always supplied to identify the network and host portions o Allow IETF (Internet Engineering Task Force) to adapt the
of the address. protocol changes in underlying network hardware or to new
• When you subnet a network by using a custom mask, you can applications
divide the IP addresses between several subnets. However, you • In general, IPv6 bases node IDs on physical addresses.
also reduce the number of hosts available on each network.
• Multicast IPv6 addresses always begin with a binary 1111 1111
The following table shows how a Class B address can be subnetted to (hexadecimal FF.)
provide additional subnet addresses. Notice how by using a custom • Following is the IPv6 local loopback address: 0:0:0:0:0:0:0:1 or :1
subnet mask the Class B address looks like a Class C address. (concise form.)
Default Example Custom Example
Additional features of IPv6 are displayed in the table.
Network Address 188.50.0.0 188.50.0.0
Feature Description
Subnet Mask 255.255.0.0 255.255.255.0
Because hardware IDs are used for node IDs, IPv6
# of Subnet Auto-
One 254 nodes simply need to discover their network ID. This
Addresses configuration
can be done by communicating with a router.
# of Hosts per Built-in support for bandwidth reservations which make
65,534 254 per subnet
Subnet Built-in Quality guaranteed data transfer rates possible. (Quality of
Subnet Address(es) 188.50.0.0 (only one) 188.50.1.0 of Service service features are available as add-ons within an IPv4
188.50.2.0 environment, but are not part of the native protocol.)
Well Known
• Assigned for specific protocols and services
• Port numbers range from 0 to 1023
• 10.0.0.1 to 10.255.255.254 With Zeroconf, you should be able to connect two computers and
IP version 4 • 172.16.0.1 to 172.31.255.254 automatically have them be able to communicate. You should also be
able to set up a small network by simply connecting devices and
• 192.168.0.1 to 192.168.255.254 without performing any additional configuration tasks.
IPv6 reserves all addresses beginning with a binary 1111
1110 11 (hexadecimal FEC0::/48) for private IP To enable Zeroconf networking, the following features must be
IP version 6 enabled:
networks. This address range is called the site-local
address range. Feature Description
IP hosts must be able to obtain an IP address without a
ICS Facts DHCP server. The Zeroconf working group has
Internet Connection Sharing (ICS) is a service available on Windows completed the IPv4LL which reserves specifies how a
systems that enables multiple computers on a single small network to device uses autoconfiguration to assign itself an IP
access the Internet by sharing one computer's connection. With ICS, addresses on the 169.254.0.0 network (mask of
most configuration tasks are completed automatically. When using IPv4 Link-
Local 255.255.0.0). IPv4LL is currently implemented as follows:
ICS:
Addresses • Automatic Private IP Addressing (APIPA) on
• The ICS system is configured as a NAT router, a limited DHCP (IPv4LL) Microsoft systems.
server, and a DNS proxy (name resolution requests from the
private network are forwarded to DNS servers on the Internet). • Implementations on Linux, Mac OS, and other
devices such as printers.
• The IP address for the private interface is automatically changed
to 192.168.0.1 with a mask of 255.255.255.0. Note: IPv6 supports link-local addressing by design.
• The default gateway of the ICS system is set to point to the Host Name IP hosts should be able to perform IP address-to-host
Internet connection. Resolution name resolution without a DNS server. Current
• Hosts on the private network should use DHCP for address and implementations include:
DNS server information. • Multicast DNS (mDNS) used by Mac OS.
• The ICS system uses DHCP to deliver the following information to • Link-local Multicast Name Resolution (LLMNR)
hosts on the private network:
Terminal Services can be used to remotely manage servers Windows NetWare To connect a Windows or Linux
or to run applications on the server. system to a NetWare server:
Linux
Remote Remote Desktop is a Microsoft service that uses Terminal • On the Windows client install
Desktop Services technology to allow you to remotely access any either the NetWare Client for
Remote Desktop-enabled system. For example, you can Windows or the Microsoft Client
enable Remote Desktop on a computer at your office. From Services for NetWare
Protected
o PEAP-EAP-TLS. This method uses • Based on Point-to-Point Protocol (PPP)
certificates (either on the local system or
Extensible
on a smart card).
• Uses standard authentication protocols, such
Authentication as CHAP or PAP
Protocol (PEAP) o PEAP-MS-CHAP v2. This method uses
certificates on the server, but passwords • Supports TCP/IP only
on the client. Use this method when the Point-to-Point • Encapsulates other LAN protocols and carries
client does not have a certificate. Tunneling Protocol the data securely over an IP network
• One of the most effective wireless security (PPTP) • Does not encrypt data (used in conjunction
solutions. with Microsoft Point-to-Point Encryption for
encryption)
• Centralizes control of remote access
authentication (in multiple remote access server
• Is supported by most operating systems and
networks). servers
o All remote access policies are maintained • L2TP is making PPTP obsolete
on a single Radius server. • Offers mutual authentication
RADIUS (Remote o All other Network Access Servers (NASs) Layer 2 Forwarding
are RADIUS clients. (L2F) • Does not encrypt data
Authentication
Dial-In User • Uses the MD-5 encryption method to encrypt • Merged with PPTP to create L2TP
Service) password information. • Can use certificates for authentication
• A platform independent method. • Uses IPSec for encryption (requires
Layer Two Tunneling
• An AAA (authentication, authorization and certificates)
Protocol (L2TP)
accounting) protocol for applications such as • Supports multiple protocols (not just IP)
network access or IP mobility. It is intended to
work in both local and roaming situations • Not supported by older operating systems
Note: You can also use tunnel mode if only one end device
understands IPSec. For example, the source device can
generate the IPSec packet, and the destination router can Most of the time, you will perform backups using a strategy that
remove the IPSec information before forwarding it on to the combines backup types. The following table compares common
destination host. The destination host can then send a normal backup strategies.
packet in response, with the router adding IPSec information Strategy Backup Characteristics Restore Characteristics
before forwarding it back to the original source host.
Requires large tapes for each
As you consider implementing IPSec, keep in mind the following: To restore, restore only the
backup.
Full Backup last backup. This is the
• IPSec is supported natively by all versions of Windows since Takes a long time to perform
fastest restore method.
Windows 2000. However, earlier operating systems like Windows each backup.
98 require additional client software in order to use IPSec. Perform a full backup
• IPSec (or any other encryption system) creates additional server periodically (for example once
processor load, as the encryption process involves computations. a week), followed by
To restore, restore the full
Before implementing IPSec you should first determine if your incremental backups every
Full + backup and every
servers can easily accommodate this extra workload. other day.
Incremental subsequent incremental
• Implementing encryption also increases the amount of network backup.
traffic that is created, as in addition to the normal traffic, there is Incremental backups are quick
additional traffic associated with the encryption process. to perform. This is the fastest
backup method.
Backup and Restore Facts To restore, restore the last
Most backup methods use the archive bit on a file to identify files that Differential backups take full backup and the last
need to be backed up. When a file is modified, the system Full + progressively longer to differential backup.
automatically flags the file as needing to be archived. When the file is Differential complete as time elapses Next to a full backup, this
backed up, the backup method may reset (clear) the archive bit to since the last full backup. is the fastest restore
indicate it has been backed up. method.
Note: Do not combine incremental and differential backups.
The following table shows the type of data backed up using each
backup method. Keep in mind the following facts about doing backups:
Resets Archive • Backup user data more often than system state data (it changes
Backup Type Backs Up
Bit? more frequently).
Full
Backs up all files regardless of the
Yes • Backup system state data and applications whenever you make a
archive bit. system change.
Incremental
Backs up files on which the archive bit
Yes • During a system state backup, all system configuration
is set. information is backed up (system data cannot be backed up
Backs up files on which the archived selectively in portions).
Differential No
bit is set. • Files backed up from one system might not restore to another
Backs up all files regardless of the system. Restore to a system running the same OS.
Copy No • Be sure to test your backup and restore strategy. It does no good
archive bit status.
to back up your data if you can't restore it.
Since parity is used, a RAID 3 stripe set can withstand a single disk
failure without losing data or access to data.
Advantages
• RAID-3 provides high throughput (both read and write) for
large data transfers.
• Disk failures do not significantly slow down throughput.
Advantages
Disadvantages
• RAID 0 offers great performance, both in read and write • This technology is fairly complex and too resource intensive
operations. There is no overhead caused by parity controls.
to be done in software.
• All storage capacity can be used, there is no disk overhead.
• Performance is slower for random, small I/O operations.
• The technology is easy to implement.
RAID 5
Disadvantages RAID 5 is the most common secure RAID level. It is similar to RAID-3
RAID 0 is not fault-tolerant. If one disk fails, all data in the RAID 0 array except that data are transferred to disks by independent read and write
are lost. It should not be used on mission-critical systems. operations (not in parallel). The data chunks that are written are also
larger. Instead of a dedicated parity disk, parity information is spread
RAID 1: mirroring across all the drives. You need at least 3 disks for a RAID 5 array.
Data are stored twice by writing them to both the data disk (or set of A RAID 5 array can withstand a single disk failure without losing data
data disks) and a mirror disk (or set of disks). If a disk fails, the or access to data. Although RAID 5 can be achieved in software, a
controller uses either the data drive or the mirror drive for data hardware controller is recommended. Often extra cache memory is
recovery and continues operation. You need at least 2 disks for a RAID used on these controllers to improve the write performance.
1 array.
The following table compares how a single break in the network affects
Media tester device-to-device communication for specific topologies.
Topology Effect
Use a media tester to make sure that a cable is A break in the network bus means that the end of the network
unbroken and that all cables are connected to bus is no longer terminated. For this reason, a break in the
the correct pins inside the connector. Bus bus typically means that no devices can communicate.
Identifying the location of the break is difficult on a true bus
network.
A break in a cable in a star means that the device connected
Tone generator to the central device (hub or switch) through that cable can
Star
no longer communicate on the network. All other hosts will be
A tone generator sends an electronic signal on a able to communicate with all other devices.
wire or cable. Use a tone generator to locate the
A break in the ring means that messages can only travel in
other end of a specific cable. Generate the tone
one direction (downstream) up to the break. Computers can
on one end of the cable, then test the other ends Ring
send messages downstream to other devices, but because of
of many cables until you detect the tone.
the break will not be able to receive any responses.
A break in one ring in a dual ring configuration has no effect
Time Domain Reflector Dual Ring on communications. A decrease in bandwidth might result,
(TDR) but data can be sent on the other ring.
Like a tone generator, a TDR sends signals on a A break in a single link in a mesh topology has no effect on
cable or a wire. Use a TDR to get information communications. Data can be routed to the destination
Mesh
about the cable such as its length and to identify device by taking a different (sometimes longer) path through
the distance to the break in a cable. the mesh topology.
Interpreting ipconfig
You can use ipconfig /all to troubleshoot IP configuration problems.
Loopback plug Following is sample output from the ipconfig /all command:
A loopback plug reflects a signal from the Windows 2000 IP Configuration
transmit port on a device to the receive port on Host Name . . . . . . . . . . . . : NY-DEV-WRK3
the same device. Use the loopback plug to verify Primary DNS Suffix . . . . . . . : westsim.com
that a device can both send and receive signals. Node Type . . . . . . . . . . . . : Broadcast
A failure in the loopback test indicates a faulty IP Routing Enabled. . . . . . . . : No
network card. A successful loopback test means WINS Proxy Enabled. . . . . . . . : No
the problem is in the network cabling or other DNS Suffix Search List. . . . . . : westsim.com
connectivity devices.
Ethernet adapter Local Area Connection:
One valuable troubleshooting method is to keep
a set of components that you know are in proper Connection-specific DNS Suffix . : westsim.com
Known good spares functioning order. If you suspect a problem in a Description . . . . . . . . . . . : 3Com 3C920 Integrated Fast Ethernet
component, swap it with the known good Controller (3C905C-TX Compatible)
component. Physical Address. . . . . . . . . : 00-06-5B-1C-92-B8
DHCP Enabled. . . . . . . . . . . : Yes
Often, physical problems are intermittent and might go away even Autoconfiguration Enabled . . . . : Yes
before you take corrective action. If the problem appears to be a IP Address. . . . . . . . . . . . : 192.168.1.141
physical problem, check the following: Subnet Mask . . . . . . . . . . . : 255.255.255.0
• Verify that connectors and components are securely fastened and Default Gateway . . . . . . . . . : 192.168.1.1
DHCP Server . . . . . . . . . . . : 192.168.1.20
that connectors are clean
DNS Servers . . . . . . . . . . . : 192.168.1.20
• Check for EMI and other atmospheric conditions that might be 192.168.1.27
causing communication problems. Lease Obtained. . . . . . . . . . : Monday, April 18, 2005 7:46:41 AM
o For wired networks, verify that cables are not near Lease Expires . . . . . . . . . . : Monday, April 18, 2005 11:46:41 AM
fluorescent lights or other sources of interference.