My very own N+ Blog!!

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

Er - 10Base-5 doesn't use BNC!

The sockets on network cards on PCs and the like are 15 pin D-types. (Known as AUI connectors sometimes)

The connectors for the RG8 cable are N-Type.

Harry.

 

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

oops!!You're quite right,the AUI 15 pin connection is used for 10Base5,BNC removed.The 10Base5 section already states use of the AUI,I forgot to read my own post first,lol.

 

Ethernet-the be all and end all of all things networking...BOING...WRONG!!!!I'd love to learn nothing more than ethernet but there are a few other technologies that are still about in todays newtorking environments and if I choose to ignore them I run the risk (and murphys' law) that it will appear on my exam and cost me precious marks.. .

Ethernet?No!Thats next door sir!!!


Token Ring

This technology is still around,utilised mainly over WANS and for reasons beyond my caring it is widely used in financial institutions,dunno why,it just is.It is a STAR/RING topology and will run at 4Mbps and 16 Mbps*.It will run over STP and UTP.Run it over STP and you can have up to 260 nodes on the ring and each node can be a maximum of 100 metres from the MAU (MAU,MsAU-take your pick it's the same thing).Run token ring over UTP and you limit yourself to 72 nodes and a maximum length of 45 metres from node to MAU (MAU,MsAU-take your pick it's the same thing,still).Your TOKEN RING network will run as fast as the slowest man,so to speak,so if you have ten nodes running at 16Mbps and one running at 4Mbps then ALL the nodes will run at 4 Mbps.It uses a technology called 'token passing' whereby a node will wait for a 'free' token,grab it,add data to it pass it IN ONE DIRECTION around the ring until the recieving node gets it.The recieving node then strips off the data and releases another 'free' token onto the network after the transmission is completed.This means that unlike CSMA/CD,token passing makes for NO COLLISIOINS!!cool.A break in the ring however will bring your token ring newtork to a stanstill.Token ring networks use TYPE1 (IDC/UDC)connectors at one end of the cable and a 9 pin female connector at the other,OR RJ45,depending on the configuration. *faster token ring is avaialable but for N+ know about 4Mbps and 16Mbps technologies.

ARCNet

STAR/STAR topology (I'm having difficulty trying to picture that one!)-Token Passing-2.5Mbps over RG62 cable to a length of 600 metres.

LocalTalk

uses DIN connectors and can be daisychained.

ATM

Runs over fibre or STP/UTP,uses fixed length frames over a point to point connection.

FDDI/CDDI

As previously explained,FDDI-a dual ring,high redundancy technology mainly implemented over WANS.A node can be up to 2 kilometres from the next one and a ring can be up to 100 kilometres in diameter.CDDI is the same technology run over Copper.

 

great idea.. I wish I had the motivation to do a similar series.. maybe one day


Anyhow, Arcnet.. think it's dead do ye ? Google it .. I dare ye.... still used .. yup, in building control systems! Just think, your heating could be arcnet based!

 

HOLY CRUD!!THIS ARCNET IS EVERYWHERE!!!!!!

Arcnet is a real dark horse,It wouldn't go amiss to have a little further insight into Arcnet,so if you want to-take a gander at the following link.

http://www.arcnet.com/abtarc.htm

EDIT:- while we're at it let's take a quick overiview of the others too..

LocalTalk..http://kb.iu.edu/data/aidj.html

ATM..http://www.webopedia.com/TERM/A/ATM.html

Thank you for that one Clyde..

 

Oh dear,oh deary deary dear deary me.. I have stumbled across a subject that has eclipsed the complexities of cabling and made it look like the four line crossword in this months beano (if you're wondering what Beano is,google it you young adolescent fool!).To be honest,it is a minefield of information that can only lend itslef to the theory side of things,since most of it is prettymuch transparent to poeple like you and me and only a handfull of its components are meant to be seen,alas it is another subject that will DEFINATELY be strongly represented on my exam I'll wager and one that I must master,as best I can.


Protocols...

This subject is best researched FULLY by N+ wannabes and I'm only going to present the very basic stuff here in a very shorthand way.I'd strongly urge you to read your study material slowly,carefully and thoroughly and don't leave the subject until you know it inside out.One subject however,the dreaded SUBNETTING,will be mentioned but in no way explained here,you really must study subnetting as a seperate subject.Network+ 2005 will expect you to know the purpose of subnetting and its application in a network environment.

You remember the OSI model?If you don't remember it..GO AWAY..and learn it,please.You need to know the functions of the Network,Transport and Session layers in order to understand how and why each protocol works at these levels.To remind you and this is JUST a reminder..

SESSION layer5-end to end connections,here is where the session is maintained,other sessions monitored and maintained and where sessions are ended (torn down as I like to say).

TRANSPORT Layer4-chops up data into smaller chunks called SEGMENTS and adds a sequence number to the packet.It gets the data ready for transporting in other words.

NETWORK Layer3-Deals with routing,predominently IP routing,this is where IP addresses are added to our packet.Since routing is performed here it is where cross platform communications are dealt with.


A protocol must be bound to a NIC in order to communicate using that protocol,this is called BINDING,binding the protocol to a NIC,see?At least one protocol must be bound to a NIC.This is done in various ways,depending on your operatng system ,in XP go to My Network Places/Proprties and properties of the desired NIC to add the protocol to.

NetBIOS/NetBEUI

NetBIOS/NetBEUI (NetBEUI is pronounced 'netbooey') are a protocol stack from many moons ago,but are still very much alive in todays network environments.In fact if you ever install XP you will be asked to pop a 'name' into a dialogue box that cannot be more than 15 characters long and contain no symbols,this is in fact a NeTBIOS name and must confrom to NeTBIOS naming conventions as such.

NeTBIOS functions at the SESSION layer and handles sessions based on the computers NeTBIOS name and service number (research netbios service numbers at this point!)

NeTBEUI functions at the TRANSPORT layer and breaks up out data into smaller chunks,only it now skips the network layer altogether (since IP addresses are not the naming convention in use) and talks directly to the LLC sub layer of the DATA LINK layer (Layer2)

NeTBIOS and NeTBEUI use what is called a 'flat name space',where the names MUST be unique to each host.This works fine for small networks but bugger that for a game of soldiers if you're dealing with thousands of hosts!Names in that scenario become an administrative nightmare.

These are NOT ROUTABLE,since no IP address is present,only names.

IPX/SPX

Think IPX/SPX-think NOVELL

NOVELL is NOT a client operating system persay,it is a SERVER operating system.The clients wanting to talk to a NOVELL server are configured accordingly.

SPX works at the TRANSPORT layer
IPX works at the NETWORK layer

Know that Novell uses an 8 digit number for what is called an INTERNAL NETWORK NUMBER (INN)and a hex string for EXTERNAL NETWORK NUMBERS (ENN) eg.13243546.a1 b2 c3 d4 e5 a6.The INN is effectively the subnet.INN numbers define addresses in a local network while ENN numbers define addressess in an outside newtork.

RIP and SAP are also NOVELL protocols,research them. RIP- http://en.wikipedia.org/wiki/Routing_Information_Protocol SAP- http://en.wikipedia.org/wiki/Service_Advertising_Protocol

To get your PC to talk to a NOVELL server you will need to install a client such as NWLink. http://en.wikipedia.org/wiki/NWLink

IPX over ethernet uses the following frametypes..
802.2
802.3
ETHERNET ii
SNAP

SPX is connection orientated
IPX is connectionless

TCP

Runs at the SESSION layer,using ports in the range of 1024-49151 (hmm,check that I think it could be wrong).The first 1024 ports are set aside as 'common' ports.Running at the TRANSPORT layer,TCP chops up our data and adds sequence numbers. http://en.wikipedia.org/wiki/Transmission_Control_Protocol

IP

Runs at the NETWORK layer to provide a 'best effort' connection to network devices in order to create a network. http://en.wikipedia.org/wiki/Internet_Protocol

DLC

An older protocol and one that you only need to know can be installed,apparently.

AppleTalk

Used by.....Apple....or MAC to you and I.Uses NBP to bind a systems name to its AppleTalk address.Install AppleTalk if you wish to network with other AppleTalk systems,duh...obviously! http://en.wikipedia.org/wiki/Appletalk

I'm sure I've missed a few,but my head is about to implode so I'm happy with it as it is.NEXT....we will be exploring TCP/IP and all of its subnetting glory,as promised..

 

At this point in time I would like to thank Mrs.Malnomates for allowing me to study as much as I shouldn't and allowing me to thoroughly clean the cooker,inside and out then be presented with a 'well done' eight pack of Carling Black Label,cold,unsullied and all mine....muwahahahaaaa...but anyways...cheers..

 

...HIC...!!

There's nothing like watching my son play rugby on a cold,windy Sunday morning with a generous helping of hangover symptoms,after the lager came...JD...my bestest buddy.

This snippet of information is just that-a snippet-covering a few soundbytes for TCP/IP that I must familiarise myself with and so should you!

TCP/IP..in a nutshell

TCP/IP is by far the most prominent protocol suite on the market to date,it is evrywhere,even in places it never used to be in,everywhere.There are many protocols that run under TCP/IP and to go through them all would be far too much of an effort (I have a hangover remember?),but worthwhile investigating nonetheless.Here are a few little bullet notes that seem the most likely to appear in conversation,in exams and on the back of lavatory doors,ladies or mens,whatever floats your boat.

IP addresses

A 32 bit,4 octet address,seperated by dots (or period)- for example 192.168.1.5.It is a decimal representation of a binary number between o & 255 for each octet.Computers do not know decimal,they speak binary,end of sports,but sane humans do not,they speak decimal,kapiche?Each IP address on any network MUST be unique to that network or an IP CONFLICT will occur,meaning you end up in an unholy row with your staff and two users who cannot get onto the network.

Subnet mask

In addition to your IP address you will also have a subnet mask,again it is a 32 bit,4 octet address that defines which subnet your IP belongs to and is a decimal representation of a set of binary numbers between 0 & 255.A subnet mask can be classed & classless

Basic IP address rules

NO two Identical IP addresses may exist on the same network.This includes the subnet mask.However the IP address can be duplicated as long as the subnet mask is different (subnetting).

IP addresses cannot end in all 1's or all 0's (we're talking binary here)since 11111111(255) is a broadcast address and 00000000(0) denotes the network ID.Know this.

Subnetting

Subnetting allows for increased performance on a network thus increasing network efficiency.It can creat 'networks within networks' and is most often implemented to increase the number of hosts that your network currently supports.You are not required to know how to subnet for N+ but I would strongly advise you learn it,since it is the ONLY way that you will fully understand the implications and implementations of subnetting.Research subnetting and classless subnetting in the CF forums or here http://en.wikipedia.org/wiki/Subnetting

IP address classes

There are FIVE current IP address classes.They are A,B,C,D,E.We concern ourselves with only A,B & C,end of.

CLASS A-1.0.0.0 to 126.0.0.0 subnet mask 255.0.0.0
CLASS B-128.0.0.0 to 191.255.0.0 subnet mask 255.255.0.0
CLASS C-192.0.0.0 to 223.255.255.0 subnet mask 255.255.255.0

There are also what are called 'PRIVATE' IP addresses.These addresses are not routable (i think) hence the need for routers to perform NAT.These adresses are (with subnets)

CLASS A-10.0.0.0 to 10.255.255.255 subnet mask 255.0.0.0
CLASS B-172.16.00 to 172.31.255.255 subnet mask 255.240.0.0
CLASS C-192.168.0.0 to 192.168.255.255 subnet mask 255.255.0.0

127.0.0.0/8 is the loopback address,know this.

KNOW THESE ADDRESSES LIKE YOU KNOW YOUR OWN!!!!!

Be aware of CIDR addressing.CIDR addresses show the network portion of an IP address by using a slash / and the number of 1's used in the binary subnet mask..eg 192.168.100.56/24-denoting 24 1's (255.255.255.0) in the subnet.Another example is 223.112.34.119/8-this is a CLASS A address,we know that because the /8 means eight 1's in the subnet mask=255.0.0.0 (an IP address with the subnet 255.0.0.0 is CLASS A)


GATEWAY

A hardware or software device that will allow access to another network or computer,no faff.

DNS

Web addresses are usually in a format known as a FQDN (fully qualified domain name) like-www.malsmells.com-but your computer does not know what that is in reality,it needs to know what IP address corresponds to that name in order to send packets to that address.DNS does just this,DOMAIN NAME SYSTEM resolves domain addresses to IP addresses.

WINS

Just like DNS,but instead of translating domain names WINS translates NetBIOS names to IP,no faff.


DHCP


An automated system of assigning IP addresses to your network hosts.Static IP addresses that you would configure manually are just fine in a small network (but with a few exceptions),however becomes a nightmare as your network expands.A DHCP server will assign from a 'pool' of addresses and offer a 'lease' time to the recieving host,say 7 days for example.After 7 days the DHCP assigned address would be removed from the host and renewed with another address from the pool (AND NOT UNUSUALLY THE SAME ADDRESS IF NO-ONE ELSE HAS BAGGED IT!).Be familiar with DHCP,WINS and DNS,thet are very important and will appear in your troubleshooting scenarios on the N+ exam,I'll put money on it!

 

A few thoughts on this latest installment of an excellent series...

Er - no. The actual packets contain no mask info - so two machines with the same IP address is an error.

Actualy - it reduces the number of hosts, as each subnet causes the loss of a Network address and a Broadcast address. The point of subnetting is organizational.

The 'C' in CIDR stands for 'classless'. The 223. example can't be a Class A because it is in the wrong number range for that. What it *is* is a '/8'.

I must say that this whole set of posts is great though!

Harry.

 

I blame the Jack daniels,but thank heavens for Harry I say!

the 223. address,thats a stupid mistake on my part,but nicely spotted H.

Organizational..my new subnetting buzzword and make it yours poeple..nice one H

The IP conflicts..It's all clear now fella,I was unaware that the subnet isn't part of the packet,though it makes perfect sense if you look at it!Nice one H..

todays lesson?leave JD for the grown ups kids..

 

Just brilliant Mal.

Though I do wish you would put a space after your punctuation marks, as that would make it much easier to read mate

 

noted..

 

Drink JD after you've studied and not before ;)

 

Sound advice from Wizard..STUDY SOBER....

I'm taking a short break tonight, I have an all important skittles match to play and look forward to putting the names on the chalkboard and instinctively writing out a table of TCP/IP protocols instead of scores. Tomorrow seems a good day to tackle Network Operating Systems and the ins and outs of how they do their magic. Note my new format (as requested by Mr.Bluerinse) with spaces after my punctuation.

 

So,what is the one thing that our networks all have in common,the one thing that holds the whole kit and kaboodle together? Protocols?NO Cables?NO Nudey desktop wallpaper?MAYBE.

The answer is simple-the OPERATING SYSTEM.We need to know a little about a few of these operating systems in order to progress through N+, how they do their magic and how they manage to break down the communications barrier that would normally have them sitting at opposite ends of the classroom, so let us get on...

Network Operating Systems -NOS

We are going to be taking a brief study of four operating systems in particular, they are Microsoft, Novell, Apple and Unix.

Before we delve any further though let us get used to a few well known phrases that we WILL see on the exam..

NOS

The acronym for...you guessed it...Network Operating System. A NOS can provide the ability to SERV-a SERVER-and may also be able to act as CLIENT-aclient accesses resources FROM a server, though not all NOS's are specifically both, see NOVELL.

SERVERS provide resources and serve them to clients.
CLIENTS access resources from the server (or servers).

A good example of this is your web browser,it is the client and it accesses web pages from a web SERVER,shoot me down if this a bad analogy.

PEER to PEER

Peer to peer networks have NO centralised server to access, rather they are comprised of a small number of machines networked together via some sort of media and each client acts as both a server (sharing its resources like folders or internet connection) that other clients on the network access when needed.It is a very simplistic approach in many ways but does make for a relatively insecure scenario given that onus for security of shared resources falls on each particular client.

Client/Server model

The client server model can be resource based, where basic folder permissions are applied, server based where authentication is required at each server on the network and organistional based where a single logon would give access to multiple servers.

Microsoft

The Microsoft 9x familly has share level security (resource based remember?), uses NetBIOS as its native networking protocol which in turn instigates a function called 'Browser Election' (research browser election and Browse master) or utilises WINS (NetBios to IP resolution)

NT, on the other hand, is a server based NOS,incorporating DOMAINS,authentication for login (usernames & passwords) and the very security orientated NTFS file system.NT has a myriad of folder permissions ..

Administrator
Power User
User
Backup Operator
Replicator
Guest
Everyone

It utilises domains efficiently (arguably I might add),uses a PDC (primary domain controller) and creates GLOBAL USERS and GROUPS.

Windows 2000PRO/SERVER


Built on the superb technologies of NT and capable of fully supporting the NTFS file structure.Quite why anyone would want W2k to run in FAT is beyond me. It uses DNS naming conventions as opposed to NT's NetBIOS naming conventions. A typical DNS name might be server.malstoilet ,there is a similarity to a web address if you look hard enough and we know by now that DNS resolves domain names to IP,get it?

Active Directory

A storage area that holds information about all the network resources,or that's how I remember it anyway.

Any single active directory must consist of at least one domain.

XP

XP PRO-designed to work in domains
XP HOME cannot join a domain.

Novell Netware

...comes in various flavours..

3.x uses BINDERY to store usernames for a single server.
4.x uses NDS-Novell Directory Services.
5.x uses NDS-Novell Directory Services
6.x uses NDS Novell Directory Services

versions 5.x and 6.x use TCP/IP natively.

NOVELL is NEVER a client OS,it is a DEDICATED SERVER NOS

to connect to a NOVELL server from Windows you install CLIENT SERVICES FOR NETWARE (with no windows server present)on each client -CSNW
to connect to a NOVELL server via a windows server you install GATEWAY SERVICES FOR NETWARE on the windows server-GSNW
for file and printer sharing from a NOVELL server you install FILE AND PRINTER SERVICES FOR NETWARE on the windows server (emulates NOVELL F&P service)-FPNW

Unix, Linux

Suns Solaris
HP UNIX
AIX UNIX
BSD

..this NOS is common in web servers, given that protocols like HTTP,FTP,DNS and ARP originated in UNIX, again correct me if I have this horribly wrong.

Printing from UNIX is via LPR or LPD or CUPS (common UNIX printing system.CUPS has an extension of .ppd

From a command prompt use lp or lpr to send a CUPS print command

UNIX administrators are called 'root'

MAC

Uses AppleTalk which has similarites to NetBIOS but has a 20 character limit for names (NetBIOS has how many????..I know..do YOU?)

AppleTalk is NOT ROUTABLE

AppleTalk incoroporates ZONES

supports TCP/IP through APPLESHARE.

So,an extremely brief insight into NOS and one you need to expand on yourself, since I've got studying to get on with and this chapter alone could take years to write in full.

Sharing and security I'll leave for another post since that is a subject all unto its own.

 

You have listed the built in security groups rather than NTFS permissions.

NTFS permissions..

Full control
Modify
Read and Execute
List folder contents
Read
Write

Also..

This is the best definition I could find in a rush..

Source

 

If you've done your A+ you will be familiar with the command line interface. Network+ wants us to know how to use a few of the utilities available to help troubleshoot connection and data transfer issues whilst using command line tools to best effect.

Simple diagnostics in command line

Before you read any further I am going to make an assumption that you understand exactly what I am referring to when I use the term 'command line' or 'c -prompt'. If you do not know what these terms mean I would strongly advise you learn about them, how to access them and know some basic commands to navigate them, alternatively you may choose to ignore that and go down the pub, in which case mine's a lager and a packet of chicken itchings please.

*for NT/2000/XP volumes use IPCONFIG and for windows 9x use WINIPCFG

IPCONFIG/DISPLAYDNS-guess what?It displays the DNS cache

IPCONFIG/FLUSHDNS-guess what?It flushes the DNS CACHE

Difficult isn't it?

ARP & RARP - ARP is the Address Resolution Protocol that resolves IP to MAC addresses (logical to physical). RARP is the Reverse Address Resolution Protocol which resolves MAC to IP addressing (physical to logical)

NSLOOKUP-type this in and get your current default DNS server infromation.

DIG-if you are running command line in LINUX then DIG will give the same infromation (a bit more actually) as NSLOOKUP in Windows.

IPCONFIG/RELEASE-release the current DHCP assigned IP address.

IPCONFIG/RENEW-renew a DHCP assigned IP address,as if you need to ask!

IFCONFIG-the LINUX command to show what IPCONFIG and WINIPCFG will show (your TCP/IP settings)

IFCONFIG eth0 down-turn off the primary NIC card
IFCONFIG eth0 up-turn ON the primary NIC card

PING-expensive golfing equipment.

PING-PING followed by either an IP address or name will send an ICMP echo packet to the stated destination and wait for a reply. This will tell you if the destination host is reachable. Be VERY familiar with PING, it is the primary tool for troubleshooting. Go to your command prompt and type PING www.certforums.co.uk and see the results. If you want to continually PING an IP address or name the use the switch -t. PING-t will transmit continuosly until stopped.

NETSTAT-displays current open TCP and UDP connections including the port on which they are being used, very handy.

NBTSTAT-displays the NetBios over TCP/IP connections and stats, similar to NETSTAT but specifically for interrogating NetBIOS connections.

NETVIEW-from what I recall it displays all NetBIOS names in the local segment that are currently connected.

A MAC address is a PHYSICAL address.

An IP address is a LOGICAL address.

In terms of troubleshooting, USE YOUR COMMON SENSE!! I don't mean that in a derogatory way, but look for the simplest thing first. Use PING on the workstation that seems to be troublesome, but also confirm that the problem is local to that machine and try PING (same address) on a neighbouring machine. CompTIA WILL test you on your ability to logically apply your troubleshooting skills, logically and effectively.I know this because my A+ exams had more than a few 'what do you do next' type questions, thankfully I got them all right!

 

Connecting to the internet or to another network via the internet is such a transparent affair dont you think? You are rarely more than one click away from nonsense like this! Nonsense or not-you are now in my grip buddy-reading every word with trembling anticipation that I might actually have learned something this fine evening and wish to share with you my intellectual capacity as a.....shut up Mal.

All Things Internet..

This is a HUGE subject and needs special attention from us N+ wannabes if we are to succeed in both the exam and in the field. So here are my bullet points that I have put together, not all from memroy I admit but there's a lot to cover, so I'm forgiven for using my notepad occasionally, I hope..

Dedicated lines-a dedicated line is ALWAYS connected and has NO telephone number, why would it need one if it never needs to dial or be dialled? Dedicated lines are primarilly utilised by technologies like T1 and T3, more on those a little later.

Dial Up lines-as you would expect from that hugely descriptive name, dial-up lines well....dial-up, end of. Dial-up is usually implemented over PSTN (Public Switched Telephone Network) or ISDN (Integrated Service Digital Network) and POTS (Plain Old Telephone System). V90 and V.92 are familiar standards for Dial Up technologies (or should be, if not I would seriously think about taking up flower arranging instead of IT for a living!) and make for a connection download speed of a WHOPPING 56Kbps, if you're lucky.There are some other standards yoou should get familiar with too....

V42-----error checking
V42bis---data compression
V44-----data compression
MHP5---both error checking & data compression

The protocols used by Dial up are PPP and SLIP

SLIP

-Serial Line Internet Protocol.
-Supports TCP/IP ONLY (no support for NetBEUI or IPX)
-Is NOT compatible with DHCP (bummer huh?)
-Transmits passwords in 'clear text' which is NOT a good idea.

PPP

Point to Point Protocol
-Supports IPX, NetBEUI, IP and DHCP (and tastes like strawberries)
-ALL remote access uses PPP!

ISDN

Intergrated Services Digital Network
-Runs over existing copper lines (POTS lines)
-BRI (Basic Rate Interface) utilises 2 'B' channels and 1 'D' channel to give a transfer rate of 128Kbps.
-PRI (Primary Rate Interface) utilises 23 'B' channels and 1 'D' channel to give a tranfer rate of 1.544Mbps*

*this PRI standard is for North America, European PRI uses 30 'B' channels and 1 'D' channel to give 2048Mbps. Know this.

-a single channel is referred to as a DSO
-ISDN uses Terminal Adaptors or TA's to connect. Think of a TA as an ISDN modem, though a TA is NOT a modem..

DSL

Digital Subscriber Line

-ADSL (Asynchronous DSL) has fast download speeds and limited upload, figures are constantly changing so I won't bamboozle you with transfer rates that might be out of date by the time we take the exam!
-SDSL (Synchronous DSL) has the same upload and download speeds and is expensive
-to seperate DSL signals from your POTS signals (your phone signal in other words) you would install FILTERS at each connection point.
-DSL uses PPPoE (Point to Point Protocol over Ethernet)

Cable

A very fast and increasingly popular method of connection that transmits data over TV coax cable.The drawback of cable is the contention rate, where several local cable users share the available bandwidth. This is a real pain in the preverbials if you live next door to a serial downloader..

Sattelite

-A solution if you are out in the wilds!Sattelite is a great way to communicate from outlying areas that would otherwise have no means of connection.You see a LOT of news reports coming in on sattelite connections, especially reports from warzones.
-One Way connection is SATTELITE in and PSTN out
-Two Way connection are SATTELITE in and SATTELITE out
-Sattelite dishes need to point in a southerly direction, so I'm led to believe.

T1

-A digital connection that has a CSU and DSU unit at both ends of the connection. The CSU provides protection from lightening strikes and electrical interference, while the DSU controls timing (synchronisation), I'd check on that if I were you but I think thats about right!
-it is a POINT to POINT connection.
-T1 uses DS1 frametypes
-transfer rates of 1.544Mbps
-uses TDM (time division multiplexing) to split frames across mulltiple channels
-T1 uses a DEDICATED connection (no phone number,remember?)

T3

Same specifications as T1 apart from a transfer speed of 43Mbps.

E1

The European standard equivalent of T1. It is rated at 2Mbps (T1 is 1.544Mbps)

E3

The european standard equivalent of T3 (surprisingly enough.. ) that has a rate of 45Mbps

J1/J3

The Japanese T-standard equivalents.
-J1 is identical to T1 (1.544Mbps and the rest of the gumph)
-J3 rates at 32Mbps

***SONET/SDH, FRAME RELAY and ATM are no longer part of the Network+ syllabus,but research them regardless, if only for prosterity.

Optical Carriers

Know these standards..

OC1-----51.8Mbps
OC256---13.2Gbps which incidentally is 256 x 51.8

If you are asked an OC standard (and there are lots of them), then simply multiply the standard by 51.8, or by 50 to give you a ruddy good guess at the answer.

Packet Switching

X.25-thats what you need to memorise for packet switching....X.25 (fast communication across high speed digital lines that are NOT leased lines)

Remote Access

-across private Dial Up -one system must be a RAS server while the client system uses DUN (Dial Up Networking)

Authentication protocols to know..

RADIUS-Remote Authentication Dial In User Service-authentication protocol for Dial UP
PAP-Password Authentication Protocol, sends clear text passwords, avoid it like the plague if you can!
SPAP-SHIVA PAP-PAP with encryption.
CHAP-Challenge Handshake Authentication Protocol-an improvement on PAP but the challenge is one way.
MSCHAP-Microsofts CHAP greatly improves authentication in both MSCHAP v.1 and MSCHAP v.2
EAP-Extensible Authentication Protocol- strong security protocol that uses digital certificates in its encryption
Kerberos-a VERY strong security protocol that issues 'tickets' to verify user authentication.

VPN

Virtual Private Network

-Uses encryption to pass data over the internet to the reciever, connecting to the IP address of your VPN device on your LAN
-Uses PPtP (Point to Point Tunnelling Protocol) which literally wraps one protocol inside another before transmitting and L2TP which is Cisco's Layer 2 Tunnelling Protocol.

ICS

Internet Connection Sharing

You should be familiar with ICS if you have studied your A+, but for those that don't know ICS...

ICS allows one client with an internet connection to share that connection with other clients in a LAN.
-It uses DHCP and DNS, NAT, AUTO DIAL and API's

Make sure you know all of these acronyms thoroughly and apologies for not translating ALL of them, but I'm knackered and need a rest for tonight.

 

Very interesting read mate!

Note quite true, Windows 2000 server and probably Server 2003 can use SLIP if the remote clients don't support PPP but clearly PPP is preferred.

I am not sure if it's Cisco's protocol or not but again it's the preferred VPN solution in Windows servers, it uses IPSec for encryption.

here a good link Windows 2000 remote access