Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 31 Aug 2001 18:34:55 -0700
From:      "Bruce A. Mah" <bmah@FreeBSD.ORG>
To:        Hroi Sigurdsson <hroi@asdf.dk>
Cc:        freebsd-hackers@FreeBSD.ORG
Subject:   Re: Should URL's be pervasive. 
Message-ID:  <200109010134.f811Ytc02837@intruder.bmah.org>
In-Reply-To: <3B902B68.E15DFA3E@asdf.dk> 
References:  <56710000.999257324@lobster.originative.co.uk> <Pine.BSF.4.10.10108310943310.81728-100000@mail.matriplex.com> <20010831154300.B27173@dragon.nuxi.com> <3B902B68.E15DFA3E@asdf.dk>

next in thread | previous in thread | raw e-mail | index | archive | help
--==_Exmh_-2016360560P
Content-Type: text/plain; charset=us-ascii

If memory serves me right, Hroi Sigurdsson wrote:
> David O'Brien wrote:
> 
> > Do you know how traceroute works??  For one, the destination host cannot
> > be listening on the port used.  And you know that each progressive
> > traceroute packet sent out bumps the destination port by one, to help
> > trace the ICMP "time exceeded" / "port unreachable" responses.
> 
> Only UDP traceroute requires that the port not be listened to (otherwise
> you wouldn't get a packet back). If using TCP you will get a packet back
> (RST/SYNACK/ICMP-portunreachable etc) whether the port is open or not.
> In theory at least..

. o O (I can't believe I'm going to participate in this discussion.)

We're not talking theory here.  Have you actually tried this in a real 
program?

People are proposing to do various hacks to existing utilities and
applications without having any real idea of the impact of their ideas,
or even in some cases what semantic sense these hacks have.

As David explained, the proposed changes to traceroute and ping make no
sense.  Please folks, try to understand what these utilities do before
suggesting making changes just to make everything gratuitously look like
a Web browser.  "TCP/IP Illustrated Volume 1" is a good start.

Even if doing so made sense in any way, adding the required
functionality is not anywhere near as easy as naively saying "this is
the way I think it should work" or "just adding a URI/URL-parsing
library".  If you want to see why, you can download and read the source
code to my pchar utility (which implements a superset of traceroute's
functionality), and look at how much code is required to deal with the
fact that (hint, hint) raw sockets don't receive TCP segments.

Bruce.



--==_Exmh_-2016360560P
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (FreeBSD)
Comment: Exmh version 2.3.1+ 05/14/2001

iD8DBQE7kDs/2MoxcVugUsMRAsGYAKDy3tQAs1npWg+eaDI402hYX8RYFwCgtNNK
Zn3QTyUWuYoOCY61WNa3amY=
=8/jY
-----END PGP SIGNATURE-----

--==_Exmh_-2016360560P--

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200109010134.f811Ytc02837>