Also note that Neal came up with several hundred traceroute servers
(beyond the ones John used in the initial study), via Altavista.
Perhaps if you look at all of them, you'll come up with a reasonable
number that support LSRR.
This could also give us a way of independently validating
our traceroute numbers, which is something we really need to do.
tom
--------
Date: Mon, 14 Sep 1998 16:08:59 -0700 (PDT)
From: Kenichi Ishikawa <ishi@cs.washington.edu>
To: tom@cs.washington.edu
Subject: path characteristics measurement with LSRRed packets.
Message-ID: <Pine.BSF.3.95LJ1.1b3.980914160716.10582A-100000@grad-pc29.cs.washington.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO
I have tried to do measuring, but there were two problem, as expected.
1) many end hosts and intermediate routers prohibit forwarding LSRRed
packets.
Condition:
I have picked out 44 hosts (same as John's traceroute servers)
and I exculde never reachable 1 host. (192.41.37.39)
Result:
i) How many end hosts can receive LSRRed packets?
16 hosts can.
27 hosts can't. (3 target hosts prohibit receipt,
24 intermediate routers prohibit forwarding)
ii) How many end hosts can forward LSRRed packets?
11 hosts can.
32 hosts can't. (4 target hosts prohibit forwarding,
28 intermediate routers prohibit forwarding)
2) UW filters LSRRed packets, therefore I have tried to measure them
using MCI 56Kbps dialup line. However large delay of 56K dialup was
a big noise for measurement. I think I need more fast line.
-- Kenichi Ishikawa