I'm trying write down the big picture:

Ok, so I fire up my browser and type www.google.com into the address line and
press enter:

1-my computer must first resolve www.google.com into an IP address. Some sort of
nslookup command is called which sends a nslookup command inside a TCP? packet
to the name server. Another TCP packet is received my computer with the result:
A.B.C.D

2-my computer looks at my subnet address and realizes that I am not on the same
network as google's A.B.C.D, so my computer sends out a HTTP GET request (again
encapsulated inside a TCP packet, which is encapsulated inside an IP packet,
which gets an Ethernet header and trailer which goes out ethernet card to the
local area network, through a switch to a router (gateway) to it's destination.

3-If I *was* on the same subnet as A.B.C.D my computer would have send out an
ARP packet asking for the MAC address of the host with IP address A.B.C.D. that
server would have replied with it's MAC address, AA:BB:CC:DD:EE:FF and then my
computer would have sent out a HTTP GET request inside a TCP inside a IP inside
an Ethernet packet addressed to AA:BB:CC:DD:EE:FF.

Questions:
Is ARP used in scenerio #2? 
When is the RARP protocol used?
Why do we have both IP and MAC addresses? Wouldn't just one of the two do
nicely? (I guess routers would have to route based on MAC address...)

Ideas?


Obviously, it's been a while since 460. 

:-)


Ryan


____________________
BYU Unix Users Group 
http://uug.byu.edu/ 
___________________________________________________________________
List Info: http://uug.byu.edu/cgi-bin/mailman/listinfo/uug-list

Reply via email to