Unix (Solaris) computers report RTT (in ping) in integer numbers of 
miliseconds. (Until we put in a fix, that will be next semester).

Linux computers report RTT (in ping) to either microsecond accuracy or
sometimes 10 microsecond accuracy.

Students in CIS 656 (spring 2003) can use afs1, ..., afs59 for 
nslookup, ping, traceroute experiments etc. These are all Unix boxes.

In addition, students in CIS 656 (spring 2003) can use the linux box
lafite.njit.edu  .  This computer can only be accessed remotely.
(ssh, telnet, rlogin, rsh).

For the ping excercise for 03/01/2003:

find two computers on the same network.

(e.g., check that afs1, ..., afs36 are all on one subnetwork. I have
NOT checked all of them!). (I think afs49 - afs59 are all on another
subnetwork: check). To be really safe, find two with addresses that
are different only in the last bit. To be really, really safe, double
check using traceroute.
(There may still be an ethernet switch or bridge between the computers.
That can not be checked using level 3 tools.)

Do ping between the computers you just chose.

A perfect choice would be from
lafite.njit.edu   to   mouton.njit.edu
(check the addresses!). These are two linux boxes in the Internet Lab.
(You can log in only to lafite). 
However, currently mouton is down with a bad card. Hopefully this 
will soon be repaired. Try it.

argerich-41 ott>: ping mouton
no answer from mouton

Lafite does not have a ``man ping'' page, and ping is slightly different 
on linux than on unix. All you need to know is:

lafite-43 ott>: ping -s mouton
Usage: ping [-LRUbdfnqrvVaA] [-c count] [-i interval] [-w deadline]
            [-p pattern] [-s packetsize] [-t ttl] [-I interface or address]
            [-M mtu discovery hint] [-S sndbuf]
            [ -T timestamp option ] [ -Q tos ] [hop1 ...] destination

If you are about to start doing ping excercises, do
(prompt>:) who
This will tell you who else is logged on to the same computer. 
If more than a few people are logged on, get out (logout or exit) 
and postpone your experiment. Too many people doing the experiment
at the same time will foul up the data.  
Corollary: Log out as soon as you are done, or have to stop to think.

Do not do too many pings! Say 10 pings per packet size is plenty!
Question: If those 5 or 10 pings do not give the same RTT, which RTT 
do you use? The Minimal? Maximal? Mean? Something else?