Re: [MBONED] mboned: UDP port conflict mtrace/traceroute

[John Kristoff <[e-mail]>]

On Mon, 29 Jul 2019 17:09:22 +0000
Warren Kumari <[e-mail]> wrote:

> Just FYI, I sent email to IANA letting them know that ports 33435 -
> 33534 should probably be listed it as "Known Unauthorized Use".
> >From some archaeology, 33434 is apparently 2^15 + 666, and the  
> "standard" traceroutes use up to 100 ports.

The starting port may be 33434 now, but in my reading of history, it
wasn't always.  If you look at Van Jacobson's original code he has this:

  u_short port = 32768 + 666;

Which indeed does add to 33434.  However, when the probes start it
begins with this:

  send_probe(++seq, ttl, &t1);

and in that routine there is:

  outudp->uh_dport = htons(port + seq);

seq was initialized to zero, but it becomes one before the first probe
is sent, which means the first destination port in the original
traceroute was actually 32768 + 666 + 1 or 33435.

Now all modern traceroute implementations just seem to set the default
port to 33434 and start there.  I don't know that any of this really
matters a whit, but I guess it demonstrates bit errors can
happen.  :-)

The only explanation I've heard for the high numbered port is that it
was choosen as one that would likely, and hopefully not be in use.
I'm not sure if Van intended the 666 part as a joke or not.  Probably a
good question for Internet historians on the [ih] list.

Sorry for veering a bit off topic.

John