Re: Successful remote AES key extraction

From: Rob Warnock (rpw3_at_rpw3.org)
Date: 04/18/05 

Date: Mon, 18 Apr 2005 03:53:48 -0500

D. J. Bernstein <djb@cr.yp.to> wrote:
+---------------
| Louis Scheffer wrote:
| > A session of ping -s shows at least 1 ms rms deviation on all servers
| > I tried.
|
| That's excessive even for a cable modem.
+---------------

Doesn't seem at all excessive to me. I have a really good symmetric DSL
connection and I see much worse than 1ms *all* the time, e.g., picking
a server at random:

    20 packets transmitted, 20 packets received, 0% packet loss
    round-trip min/avg/max/stddev = 71.346/73.763/84.398/2.595 ms

It just depends on whether you happen to go through a "busy" peering
point or not. When things get busy, variance gets *BAD*, e.g., the same
remote host as above, a half-minute later:

    20 packets transmitted, 20 packets received, 0% packet loss
    round-trip min/avg/max/stddev = 71.286/82.635/140.136/19.699 ms

This sort of variation is quite typical, actually.

+---------------
| When I send pings several miles from UIC to the University of Chicago,
+---------------

*HAH!* You're probably not going through more than a couple of routers.
[Use "traceroute" to see exactly how many.] Try a path that goes halfway
across the country, something with over a dozen hops that goes through
at least two major peering points, and you'll see what I mean.

-Rob

-----
Rob Warnock <rpw3@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607