HOME MCSE CISCO COMPTIA LINKS

Server+ Study Guide

TOPOLOGIES
Types Characteristics Used in
Star Connected to a centralized hub, disadvantage is dividing bandwidth. 10Base5, 10Base-T
Bus Connected to a backbone. Single cabling, ends are terminated. Signal can be sent in both directions or sent unidirectional. Difficult to troubleshoot. Small networks 10Base5, 10Base2
Mesh Each computer is connected to every other computer. Routers are used to find the best path on network. Ends not terminated. WAN configurations
Ring Connected in a circle, difficult to troubleshoot, signal degeneration low. Token-passing

OPERATING SYSTEM
Types Characteristics Used Best By
MS Windows NT

Uses a Directory containing information used by users, groups and computers to manage certain resources.

Primary file system: New Technology File System (NTFS).

Windows NT Server

Reason: because of common NTFS file system and optimised to work best with each other.
Novell NetWare

Uses a Directory Service to manage resources.

Primary file system: a combination of File Allocation Table (FAT) and Directory Entry Table (DET). Provides a text based and command prompt console at the server.

 DOS, Windows 3.11, Windows 9x, Windows NT Workstation.
Unix   Sun Sparc Stations (Unix specific clients)

PROTOCOLS
Types Characteristics OSI Layer
IPX (Internetwork Packet Exchange)

Novell NetWare's default protocol.

Packet sizes:
Ethernet = ± 1500 bytes.
Token Ring = ± 4KB.

Performs addressing and routing functions and need some configuration.

 Network layer
IP (Internet Protocol)

DOD standard designed for ARPAnet. Responsible for addressing and routing packets between hosts.

Two models:
DOD with 4 layers
OSI with 7 layers which is Windows NT and Unix's default protocol.

Configuration needed.

 Network layer (for DOD model - Internet layer)
NetBEUI Protocol made for Microsoft. Has fast packet delivery with little configuration. Not routable. Used in small networks. It operates on the Network and Transport layers of the OSI model. Network and Transport layers
TCP (Transmission Control Protocol) Reliable, connection-oriented protocol for sending large amounts of data with a lot of ACK overhead. Routable.  
UDP (User Datagram Protocol) Unreliable, connectionless-oriented protocol for sending small amounts of data without ACK overhead  
DNS Domain Name System  
IP (Internet Protocol) Addresses and routes packets between hosts. Connectionless protocol  
ARP Address resolution protocol  
FTP (File Transfer Protocol) Fast and error-free. Transfers files from host to host.  
SMTP (Simple Mail Transfer Protocol) Transfers mail files from a mail client to mail server prior to the destination  
POP3 (Post Office Protocol ver. 3) Transfers mail files from a mail server to a mail client form  
IMAP Interactive mail access protocol  
ICMP Internet control message protocol  
RIP Routing information protocol  
OSPF Open shortest path first  
SNMP (Simple Network Management Protocol) Management tool to monitor and control remote network devices.  
HTTP (HyperText Transfer Protocol) Used on the World Wide Web to transfers files from server to a web browser  

RAID's
Types Characteristics
Level 1 Disk Mirroring A partition is duplicated on another physical disk. Consist of 1 data channel and 2 drives; 1 is used for data and 1 is used for parity. Fault tolerant but expensive and needs lots of disk space.
Level 1 Disk Duplexing

A partition is duplicated on another physical disk, which is connected to another Hard Drive Controller. Uses 2 data channels, 2 data cables and 2 DASD; 1 is used for data and 1 for parity. Provides faster read speeds than mirroring.
Disk Striping with and without parity (Levels 2 - 5) Bits of data are written to each drive in array in succession. Improves read/write speeds. Parity checking compares the bit string to an odd or even count and relies on an extra parity bit. If not matched, the data string is sent again. No parity usage improves overall data transmission, but should be used when speed is more important than fault tolerance.
Level 2 Data blocks gets broken up and distributed across all drives in array with error checking.
Level 3 Stored as parity, data blocks are broken up and distributed across all drives in array with one drive that stores parity data.
Level 4 Large blocks of data are distributed across all drives in the array.
Level 5 Data and parity information is send separately across all disks in the array. The parity stripe is used for disk reconstruction if a disk fails. Provides best fault tolerance because it uses several drives with block interleaving.

OSI LAYERS
Layer Types Characteristics Protocols Hardware Devices Services
Layer 7 - Application Provides network services (eg. messaging and print services) and handles network access, flow control and error recovery SMB and NCP Gateways Telnet, FTP use TCP, TFTP, NFS, SNMP, SMTP use TCP
Layer 6 - Presentation Present data in a form usable by the application layer. Redirector resides here. Also responsible for encrypting data and translation of character sets. NCP Gateways Telnet, FTP use TCP, TFTP, NFS, SNMP, SMTP use TCP
Layer 5 - Session Establish and maintain a session. Computers have synchronization between them.   Gateways Telnet, FTP use TCP, TFTP, NFS, SNMP, SMTP use TCP
Layer 4 - Transport Responsible for reliable communication. Includes error control, repackaging and dividing of messages. Have connection-oriented transmission and end-to-end reliability. SPX, TCP, UDP and NetBEUI Gateways TCP/SPX - UDP
Layer 3 - Network Responsible for logical network addressing. Includes translating system names into addresses, addressing, routing, transmission problems and reassembling data. IPX, IP, ICMP, ARP, RARP, RIP, OSPF, EGP, IGMP, NetBEUI, DLC, and DecNET Routers and Brouters IP/IPX, ARP/RARP, ICMP, IGMP
Layer 2 - Data link

Transmits data from network to physical layer. Responsible for logical topology and MAC addressing.
LLC (Logical Link Control) - defines service access points and controls link control.
MAC (Medium Access Control) - communicates with adapter card.

Bridging, physical address resolution, and media access description.

 HDLC Bridges, switches and brouters Ethernet, Token Ring
Layer1 - Physical Changes binary data into electrical pulses on the physical medium. Also defines cables, cards, and physical aspects.   NICs, cables, repeaters, hubs, patch panels Ethernet (CSMA/CD), Token Ring

Network Component Concepts - Physical Layer
Types Description Used for
Hubs

Provides a central attachment point for network cabling.

Three types:
passive - absorbs some signal with no electronics to process data signal
active - cleans signal with electronics to amplify signal
intelligent - managed and switching hubs
LANs or WANs
Switches Intelligent hubs with bridging capabilities that contains circuitry that routes signals between ports on the hub. Switch filters traffic through MAC addresses. Creates sessions on ports within the hub. Reduces bandwidth waste. Used when upgrading to 100Mb Fast Ethernet
MAUs Connects main cabling structure to devices. Also adds fault tolerance. Token Ring network
Transceivers Connects different Ethernet nodes together to create a segment.  
Repeaters Regenerates signals between similar network segments. Passes broadcast storms. LAN

Data Link Concepts - Data Link Layer
Types Description
Bridges Segments networks. Packets are forwarded based on the destination address. Forwards all protocols and renew signals at packet level. Uses RAM to build a routing table based on hardware addresses. Some are capable of connecting dissimilar network topologies.
MAC addresses Each address is unique and is used by devices to direct their packets to other devices.

Network Concepts - Network Layer
Types Description
Router Routes packets across multiple networks. Uses RAM to build a routing table based on network addresses (IP address). Shares status and routing information to other routers to provide better traffic management and bypass slow connections. Will not pass broadcast traffic. Are slower than bridges due to complex functions. Can accommodate multiple active paths between LAN segments.
Brouter Will act as a router for specified protocols (ie. Network layer) and as a bridge for other specified protocols (Data Link MAC sublayer).
Routable Packets are sent beyond a single LAN/WAN segment, but with a non-routable protocol, packets will remain on the originating LAN segment.
Static Routing Interaction is required to fiill the routing tables and provide accurate IP addressing.
Dynamic Routing Use information from neighbouring routers to fill routing tables. Human error factor is greatly reduced.
Unique Network IDs Prevents confusion between devices and properly directs packets/datagrams.
Default Gateway The entry and exit point of a subnet.
Subnetworks Created to provide security and/or reduced traffic over a WAN or congested networks.

Transport Concepts - Transport Layer
Types Description
Connectionless Internal nodes along the message path do not participate in error correction and flow control. Connectionless protocols include: UDP, NetBEUI, IP, TFTP, NFS
Connection-oriented ACK checks that the host received each segment of the message for reliable delivery service. If not received - data gets send again; if not delivered correctly - transport layer initiate retransmission or inform upper layers. To ensure packet delivery, it uses segmentation, flow control, and error checking.

CABLING

Coaxial:
Types Used for Speed & Length Connectors
Thinnet 10Base2 Ethernet cabling 10Mbps & 185m max BNC or RG-59 Broadband transmission -Television Cable
Thicknet 10Base5 Ethernet cabling 10Mbps & 500m max DB15 or N-series

Twisted-Pair:
Types Characteristics Used for Speed & Length Connectors
UTP/STP - CAT 3 Voice or data, cable segments to workstations or printers 10BaseT Ethernet cabling 10Mbps & 100m max RJ45
UTP/STP - CAT 5 Voice or data, backbone and cable segments 100BaseTX Ethernet cabling 100Mbps & 100m max RJ45
UTP (Unshielded Twisted Pair) Cheap and easy to install. Sensitive to crosstalk Falls under CAT 2 to 5 100Mbps & 100m max RJ45, RJ11, RS232 and RS449
STP (Shielded Twisted Pair) Easy to install, bit more expensive. Crosstalk is reduced via foil around the wiring. Falls under CAT 2 to 5 500Mbps & 100m max RJ45, RJ11, RS232 and RS449

Fibre-Optic:
Characteristics Used for Speed & Length Connectors
Great backbone 10BaseF Ethernet cabling 10Mbps & 2 000m max SMA
100BaseFX Ethernet cabling 100Mbps & 2 000m max

GATEWAYS
Default gateway - an IP address used to forward packets from one subnet to another.
Gateway for connecting dissimilar systems or protocols - grants a workstation a direct connection to the host computer and acts as a messenger between them. Adds expansion and functionality because unlike computer systems can inter-operate.

INTERNET

Top Level Internet Domains
.com - used for profit based companies
.edu - used for schools, universities and colleges
.gov - used for non-secret government agencies
.mil - used for US Military sites
.net - used for network / internet service providers
.org - used for charitable and non-profit organisations

IEEE SPECIFICATIONS
Types Description Characteristics
802.1 Media Access Control Divides OSI layer 2 into two sub layers.
802.2 Logical Link Control (LLC) Manages link control and defines SAPs. Also adds header fields to identify upper-layer protocols. Ensure that MAC frames find their way to the right network layer.
802.3 Carrier Sense/Multiple Access with Collision Detection (CSMA/CD) Provides physical layer options including topologies, media types, data rates and signaling modes. Very similar to the Ethernet standard.
802.5 Token Ring Uses token-passing media access protocol across a physical star and logical ring.

TCP/IP

Fundamentals
Types Description/Facts
IP default gateways
The entry and exit point of datagrams between subnets
Packets are decremented by at least 1 until it reaches 0, to prevent it from travelling forever. If too large - then it gets fragmented.
Must have same subnet mask as the network it resides on.
DHCP (Dynamic Host Configuration Protocol)
Uses BOOTP protocol to communicate with clients and cross routers if the router is RFC 1542 compliant and has BOOTP forwarding enabled.
Add a scope to provide services for additional subnets, range is limited to a particular subnet
Process: client request; server offer; client select; server acknowledges
DCHPREQUEST - a clients attempt to renew after 50% of the lease life has expired.
DHCPREQUEST - a clients attempt to renew again at 87.5% expiration of lease life
IPCONFIG, WINIPCFG - to confirm IP assignments
DHCPRELEASE - sent by client to retain IP assignment or until lease expires
DHCP Relay Agent - forwards DHCP messages between clients and servers
DNS (Domain Name Services)
Resolves DNS host names to IP addresses.
Uses static mapping
CNAME - a method of DNS aliasing (eg. www or ftp server)
Zone transfers can be setup between primary and secondary DNS servers for fault tolerance
WINS (Windows Internet Naming Service)
Resolves NetBIOS computer names to IP addresses.
Eliminates sending of broadcasts for computer name resolution and LMHOSTS files
WINS requests = routable datagrams
Process: WINS client register NetBIOS name and IP at startup; WINS client sends name query request to WINS server to talk to another host; if IP and host name is found the WINS server will send to requesting client; during shutdown, WINS client will send a name release to the WINS server
Uses dynamic mapping, does not support WINS Macintosh
Host names must be 15 char or less, and the UNIQUE UNIX computer names and IPs are entered statically
HOSTS
Syntax = IP address TAB host name. Multiple hosts can be on one line, except where the first instance of the IP address is found
Contains mappings between DNS host names and IP addresses.
Can be up to 256 char long (on NT servers), case sensitive
At least one file on each computer must not be using DNS
Most common used names must be at the beginning of the file
LMHOSTS
Contains mappings between NetBIOS computer names and IP addresses. Not case sensitive
Used by non-WINS clients using NetBIOS broadcasts for NetBIOS name resolution
# - always means 'comment', except when in front of a special command (eg. #dom, #end_alternate)

Concepts
Types Decimal range Default subnet mask
Class A 0-126 255.0.0.0
Class B 128-191 255.255.0.0
Class C 192-223 255.255.255.0

Ports
Application Port number
HTTP 80/tcp
TELNET 23/tcp
FTP 21/tcp
SMTP 25/tcp
POP3 110/tcp

Utilities
Utilities Description Command syntax to test on Internet
ARP Gathers hardware addresses of local hosts and default gateways. Displays and edit the ARP cache. Translates IP addresses to MAC addresses. arp (at DOS prompt) Displays menu of all possible switches.
TELNET A terminal emulation program to run interactive commands on Telnet. Opens remote sessions with UNIX hosts. Data only passes if a connection is established, if connection breaks, Telnet will inform you. Can be used to test login configuration parameters to a remote host. telnet lexmark.com or telnet 192.146.101.2 (from Start - Run command)
NBTSTAT Used for DNS and WINS name resolution, local cache lookup, and referral to LMHOSTS and HOSTS files. Reports statistics and connections for NetBIOS over TCP/IP. Can't test on Internet.
TRACERT Determines the route a packet takes to reach its destination tracert lexmark.com or tracert 192.146.101.2 (at DOS prompt)
NETSTAT Displays information on current TCP/IP connections netstat (at DOS prompt)
WINIPCFG Identifies IP address conflicts and displays IP-addressing information for local network adapters or a specified NIC. winipcfg (from Start - Run command)
FTP Transfers files between server and client by using TCP. ftp lexmark.com or ftp 192.146.101.2 (at DOS prompt)
PING Checks to see if host is available and active - ie checks presence of remote IP address. ping lexmark. com or ping 192.146.101.2 (at DOS prompt)

NETWORK SECURITY
Share-level security User-level security
A unique password is assigned to each resource. Shares password-protected resources in Windows 95. Shares authorised-resources in Windows NT.
Each user has one password. Preferred over share-level security.

NETWORK IMPLEMENTATION AND INSTALLATION

Safety
Room conditions Normal humidity to prevent ESD. Place equipment in secure room for security reasons
Placing contents and personal items Keep in consideration heat, EMI, TV and radio interference. EMI can be fluorescent lights, elevator motors, large generators or magnets.
Computer equipment Can effect unshielded data cables because of EMI. Faulty computer equipment can lead to faulty network components.
Error messages Can use diagnostic software for faulty computer parts or other problems.

Cabling
Installing an analog modem in a digital jack Risk is high to burn out modem
Using RJ-45 connectors on different cables For 10BASE-T - use 2 pairs of CAT 3 wires, except when upgrading to 100BASE-TX in the future - then use CAT 5 wires.

NETWORK TROUBLESHOOTING

Troubleshooting Tools
Type Characteristics
DVM
Digital Volt Meters		 Used for network cable troubleshooting. Also measures voltage passing.
Network Monitor Checks errors, packet types and traffic on each computer.
Oscilloscope Measures amount of signal voltage per unit of time.
Protocol Analyzer Consists of a built-in Time-Domain Reflector. For troubleshooting purposes, it looks inside the packet to determine the cause.
TDRs Time-Domain Reflectors Sonar-like pulses looks for shorts, breaks or crimps in network cables.

Troubleshooting
Frame Types If incorrect, problems will occur between two systems using IPX/SPX.
TCP/IP General problems are normally caused by incorrect subnet masks and default gateways.
Thinnet Coaxial Cable

Always make sure the cable terminator reads 50Ohms.