From: Kragen Sitaker (kragen@pobox.com)
Date: Sun Nov 26 2000 - 13:56:57 PST
Grlygrl writes:
> hey, i JETTED this thing out WAY earlier but because of the
> kragen factor it just WAFTED in like a paper airplane.
To sum up: it spent one minute on AOL (which *could* be my fault --- my
machine goes offline for a minute or two about four times a day), ten
seconds on kragen.dnaco.net, six minutes on xent. It wafted like a
paper airplane, but it's not the kragen factor to blame --- THIS time.
It's been kragen.dnaco.net to blame before, though.
Detailed dissection of the specimen follows:
>From fork-owner@kragen.dnaco.net Sun Nov 26 12:14 EST 2000
Received: from wormwood.pobox.com (wormwood.pobox.com [208.210.124.20])
by april.dnaco.net (8.9.1/8.9.1) with ESMTP id MAA29038
for <kragen@dnaco.net>; Sun, 26 Nov 2000 12:14:26 -0500 (EST)
Received: from XeNT.ics.uci.edu (xent.ics.uci.edu [128.195.21.213])
by wormwood.pobox.com (Postfix) with ESMTP
id 23F63728DC; Sun, 26 Nov 2000 12:14:24 -0500 (EST)
Received: from kragen.dnaco.net (kragen.dnaco.net [207.238.206.70])
by XeNT.ics.uci.edu (8.8.5/8.8.5) with SMTP id JAA06999
for <FoRK@xent.ics.uci.edu>; Sun, 26 Nov 2000 09:13:07 -0800 (PST)
Received: (qmail 14107 invoked by alias); 26 Nov 2000 17:07:59 -0000
Delivered-To: fork@kragen.dnaco.net
Received: (qmail 14103 invoked from network); 26 Nov 2000 17:07:58 -0000
Received: from imo-r10.mx.aol.com (152.163.225.10)
by kragen.dnaco.net with SMTP; 26 Nov 2000 17:07:58 -0000
Received: from Grlygrl201@aol.com
by imo-r10.mx.aol.com (mail_out_v28.33.) id t.f9.4f103ad (26116)
for <fork@kragen.dnaco.net>; Sun, 26 Nov 2000 12:07:01 -0500 (EST)
This says it was handed off to AOL's mail at 17:07:01 UTC, received on
kragen.dnaco.net at 17:07:58 UTC, received at xent at 17:13:07 UTC,
dumped in my mailbox at pobox at 17:14:24 UTC, and then forwarded to
So it looks like the kragen factor is indeed to blame. Strange, since
kragen.dnaco.net is running qmail. Usually mail gets delivered immediately.
But what about clock skew?
sleep 2 | telnet kragen.dnaco.net daytime
sleep 2 | telnet wormwood.pobox.com daytime
sleep 2 | telnet xent.ics.uci.edu daytime
Trying 207.238.206.70...
Connected to kragen.dnaco.net.
Escape character is '^]'.
26 NOV 2000 16:20:56 EST
Connection closed by foreign host.
Trying 208.210.124.20...
telnet: Unable to connect to remote host: Connection refused
Trying 128.195.21.213...
Connected to xent.ics.uci.edu.
Escape character is '^]'.
Sun Nov 26 13:26:01 2000
Connection closed by foreign host.
Oh ho! Xent's clock is set some five minutes ahead of
kragen.dnaco.net's. So actually, it couldn't have sat on
kragen.dnaco.net for more than ten seconds or so.
kragen.dnaco.net sets its clock by NTP. If we assume that NTP is
working and AOL sets their clocks correctly, the mail sat on AOL for
roughly a minute before ending up on my machine, then ten seconds or so
more on my machine, then an undetermined amount of time on xent.
So, the question is, what's the clock skew between my machine and
april.dnaco.net? Less than one second:
april:/home/kragen/ date; telnet kragen.dnaco.net daytime; date
Sun Nov 26 16:38:14 EST 2000
Trying 207.238.206.70...
Connected to kragen.dnaco.net.
Escape character is '^]'.
26 NOV 2000 16:38:15 EST
Connection closed by foreign host.
Sun Nov 26 16:38:15 EST 2000
So it really did take six and a half minutes from when it arrived on my
machine to when it arrived in my dnaco mailbox. If, as it appears,
wormwood.pobox.com's clock is fairly accurate, essentially all of the
delay is on xent.
--
<kragen@pobox.com> Kragen Sitaker <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we possess
ourselves.
-- Gandalf the Grey [J.R.R. Tolkien, "The Two Towers", Bk 3, Ch. XI]
This archive was generated by hypermail 2b29 : Sun Nov 26 2000 - 14:00:44 PST