here:
TCP/IP("stack") is a single set of protocols that can work on top of
each physical network type.
* IP - Internet Protocol: This is the the part of TCP/IP that
handles addressing, routing, and delivery of messages or "packets."
Each computer on the network has an IP address which is what IP sends
to. IP does not guarantee that it's packets will be recieved, only
that it will make it's best effort to deliver the packet.
* ICMP - Internet Control Message Protocol: Like I said, IP only
makes a "best effort" to deliver a packet. When it can't, it turns to
ICMP which sends a message like "host unreachable," "time exceded,"
and "redirect" to name a few. If you've ever used the ping program,
which is supposed to allow you to test whether you can connect to
another machine, it sends an ICMP Echo Request which prompts the other
machine to send an ICMP message back, thus telling you that the remote
machine is connected.
* IGMP - Internet Group Management Protocol: Like ICMP, IGMP is a
subset part of IP. IGMP is responsible for addressing for multicast
messages - i.e. a message meant for many different hosts on different
networks.
* TCP - Transmission Control Protocol: The best protocol in the
TCP/IP suite if you ask me. When I told you about IP, I said that it
doesn't guarantee delivery, it only makes a best effort. TCP is a
protocol that sits on top of IP and makes sure that each packet is
delivered correctly. It does this by having the receiving machine send
a receipt for each set of packets it receives. TCP also uses this
acknowlegment system to make sure that the packets are received and
reassembled in the order they were sent.
* UDP - UDP is TCP without error-checking or message
order-checking. UDP has less overhead, but is not quite as reliable,
so it isn't used very much. The only application I can think of off
the top of my head is TFTP (Trivial File Transfer Protocol) which is
used mainly for sending files over a LAN a short distance. It doesn't
have TCP's error checking, so it takes up less bandwidth, but if
something gets lost, no one knows it.
* ARP / RARP - Address Resoluton Protocol / Reverse Address
Resolution Protocol: ARP is used to convert IP addresses into MAC
addresses, while RARP converts MAC addresses to IP addresses.
MAC addresses for addressing is used in Ethernet.
The Internet Assigned Numbers Authority (IANA) has reserved the
following three blocks of the IP address space for private internets:
* 10.0.0.0 - 10.255.255.255 (10/8 prefix)
* 172.16.0.0 - 172.31.255.255 (172.16/12 prefix)
* 192.168.0.0 - 192.168.255.255 (192.168/16 prefix)
Request for IP addresses should be sent to one's ISP who then
registers with one of the following regional registries:
* APNIC (Asia-Pacific Network Information Center)
<http://www.apnic.net>
* ARIN (American Registry for Internet Numbers )
<http://www.arin.net>
* RIPE NCC (Reseau IP Europeens) <http://www.ripe.net>
* INTERNIC (Internet Network Information Center)
<http://www.internic.net>
**INTERNIC is the top level Internet registry. It also administers
the North America region
"IP number" is a four byte value that, by convention, is expressed by
converting each byte into a decimal number (0 to 255)
The end user can install TCP/IP on a personal computer without any
knowledge of either the corporate or regional network. Three pieces of
information are required:
1. The IP address assigned to this personal computer
2. The part of the IP address (the subnet mask) that distinguishes
other machines on the same LAN (messages can be sent to them directly)
from machines in other departments or elsewhere in the world (which
are sent to a router machine)
3. The IP address of the router machine that connects this LAN to
the rest of the world.
In the case of the PCLT server, the IP address is 130.132.59.234.
Since the first three bytes designate this department, a "subnet mask"
is defined as 255.255.255.0 (255 is the largest byte value and
represents the number with all bits turned on). It is a Yale
convention (which we recommend to everyone) that the router for each
department have station number 1 within the department network. Thus
the PCLT router is 130.132.59.1. Thus the PCLT server is configured
with the values:
* My IP address: 130.132.59.234
* Subnet mask: 255.255.255.0
* Default router: 130.132.59.1
MORE DETAILED:
The IP address is the number that is assigned to your computer's
Internet interface (modem / network card) as an ID on the Internet or
any other TCP/IP Network for that matter. It is "attached" to the MAC,
or machine hardware address for your computer's modem or network card.
The format for IP addresses is four sets of no more than three numbers
separated by periods (xxx.xxx.xxx.xxx).
If a given device has more than one network interface (i.e. two
network cards and a modem) AND each has it's own IP in the network, it
is known as multi-homed. A good example of this would be a router, or
a hardware firewall. Each device has at least two network interfaces,
one for the Internet side, and one for the LAN side, each with it's
own IP address.
It would be very hard to route data over a TCP/IP network if the IP
addressing system didn't have some organization to it. As you know,
the Internet is a large network of smaller networks, so it would make
sense to have a particular range of IPs that are known to be assigned
to each network. This makes physical and logical routing much easier.
It happens that by reading the first part of an IP you can determine
the network it is located on, but this is where it gets a little
tricky. Because of the fact that different size networks need
different numbers of IPs, three basic IP address classes were created
- Class A, Class B, and Class C.
* Class A IPs use the first set of numbers to identify the network
(10.xxx.xxx.xxx). These addresses range from 1.xxx.xxx.xxx to
127.xxx.xxx.xxx, and are reserved for large networks and backbones
like Comcast since each network contains over 16 million hosts.
* Class B IPs use the first two sets for a network ID
(172.16.xxx.xxx). These addresses range from 128.xxx.xxx.xxx to
191.xxx.xxx.xxx, contain 65,534 hosts per network.
* Last, but not least, we have the Class C IPs. They use the first
three sets for a network ID (192.168.0.xxx), and range from
192.xxx.xxx.xxx to 223.xxx.xxx.xxx. Each network contains 254 hosts.
At this point, you may be saying, "Pete, why do the Class C addresses
stop at 223.xxx.xxx.xxx? Can't I have a 346.xxx.xxx.xxx IP address?"
Well, no you can't. The reason is the way that computers handle
numbers. Binary numbers (0's and 1's) are used in IP addresses (an
8-bit binary number for each set of 3 numbers to be precise). If you
are familiar with binary numbering, then you know that an 8-bit binary
number can only handle the numbers 0-255.
To answer the original question, most of the IP numbers above
223.xxx.xxx.xxx are either reserved for research or for "future use".
Another thing that limits the number of useable IPs is the number of
IPs that are reserved for special functions like troubleshooting. Here
is a brief list of the ones I can think of off the top of my head, and
that I use a lot:
* 255.xxx.xxx.xxx - Addresses starting with 255 are reserved for
calculations involved in subnetting
* xxx.xxx.xxx.0 - Addresses that end with a single 0 are called
network addresses. They are a shorthand way of describing a whole
network. For example, my home network (192.168.1.1 - 192.168.1.254)
could be described as 192.168.1.0.
* 127.0.0.1 - This address is very useful. It's the loopback
address, and essentially, this tells the computer to send to itself.
It is used to determine whether your TCP/IP stack is correctly configured.
* xxx.xxx.xxx.255 - Any address that ends in 255 is a broadcast
address and will send to all nodes on your subnet/network.
There is also a special set of IPs that have been set aside for
"private network" use. They will never be included on the Internet,
and there is one range for each IP address class:
* Class A - 10.0.0.0 -> 10.255.255.255
* Class B - 172.16.0.0 -> 172.31.255.255
* Class C - 192.168.0.0 -> 192.168.255.255
If you are just setting up a small LAN you can use the 192.168.0.0
range like I did. My LAN uses 192.168.1.1 - 192.168.1.254 so I have up
to 254 numbers to assign. Lucky for me, I only have 13 computers, and
only 6 of them are connected to the network (for now :-) so I don't
need to worry about assigning all the IPs by hand. If you are setting
up more computers than I am, and you're lazy (a lot like I am) you may
want to consider using DHCP.
About Dynamic Host Configuration Protocol:
If you've ever dialed-up for Internet service, you've probably used
DHCP and not known it. If you've dialed-up and didn't understand
anything I said above in the TCP/IP section, then you definitely used
DHCP.
As the name suggests, DHCP automatically assigns an IP (and other
settings if you so choose) to your computer when it boots up.
Essentially, you set up one of your computers on the network to have a
DHCP server, and give it a range of IP addresses to take care of, for
example our 192.168.1.1 - 192.168.1.254 network. Each of the other
computers is set to "Obtain an IP address automatically" and when it
boots up, it probes the network looking for a DHCP server. The server
responds, and sends the computer an IP to use from the list of IPs it
has to administer. The IP is given on a lease, typically of a week or
so, that that computer is allowed to use the IP for. When the time
expires, the computer must request that the lease be renewed, or it
gets kicked off of the network.
DHCP, while not used in smaller networks, is good to be aware of in
case you have a smaller number of IPs than computers, thus requiring
the IPs to be shared among them, or when you have a HUGE number of
computers and don't feel like going to the trouble of assigning IPs to
each one.
SUBNET:
Lets say for a second that we want to keep it as 1 network of 254
hosts and that we have been assigned to use the Class C network
192.168.1.0. Why 254 hosts or "combinations" and not 256 as an 8-bit
binary number should allow? Well, remember that we can't use
192.168.1.0 since that is the network identifier address, and we can't
use 192.168.1.255 since that is the broadcast address. We can assign
192.168.1.1-192.168.1.254, or 254 hosts.
Notice how the first three octets of our subnet mask are set to 255?
255 in binary is 11111111. The last octet is 0. 0 in binary is
00000000. See a pattern forming? Octets that are all 1's in binary
identify the part of the address that identifies the network, while
those that are all 0 identify usable host slots.
For more subneting go to: http://noc.skilmnet.net/learn/tcpip/routing.html
Ipsens
Tidak ada komentar:
Posting Komentar