1997-09-30 - Re: remailer latencies

Header Data

From: “David E. Smith” <dave@bureau42.ml.org>
To: cypherpunks@Algebra.COM
Message Hash: 1fd7737ab26684fa7e250c0a2787a3ea441a8bd6538a16ea2f1ff2a9dd23fcb7
Message ID: <Pine.LNX.3.96.970930191318.151D-100000@bureau42.ml.org>
Reply To: N/A
UTC Datetime: 1997-09-30 19:41:11 UTC
Raw Date: Wed, 1 Oct 1997 03:41:11 +0800

Raw message

From: "David E. Smith" <dave@bureau42.ml.org>
Date: Wed, 1 Oct 1997 03:41:11 +0800
To: cypherpunks@Algebra.COM
Subject: Re: remailer latencies
Message-ID: <Pine.LNX.3.96.970930191318.151D-100000@bureau42.ml.org>
MIME-Version: 1.0
Content-Type: text/plain



Monty,

Just so you know that the less-than-100-percent message rate doesn't
necessarily mean that messages are getting lost:

Last update: Tue 30 Sep 97 15:09:32 EDT
remailer  email address                        history  latency  uptime
-----------------------------------------------------------------------
bureau42 remailer@bureau42.ml.org         -----------+  3:31:14 100.00%

It's a fluke; I just happened to look at this list (by fingering
rlist@anon.lcs.mit.edu) while I'd been on the net for a while, and so no
messages were in transit.  Mere seconds later, the EFGA list showed my uptime
as 99.97% and Raph's list showed a meager 94.08% (it looks like somewhere,
about three days ago, messages got dropped somehwere).

On balance, the remailer network is rather more reliable than you seem to
give it credit for.  (Side note: the more reliable remailers, with bureau42
being a notable exception :) tend to be faster too.  The top remailers as of
_right_now_ all have reliability above 99.90% and delay times less than
twenty-five minutes.  Chain a few of those together and your message will
almost certainly arrive in less than an hour. Few would argue that that's an
unacceptable delay. Messages routinely take longer than that to be delivered,
due to random nameserver burps and other "normal" factors.)

dave (bureau42 gearhead)






Thread