(FWD) Re: Servers with Dynamic IPs

Roger Dingledine arma at mit.edu
Mon Aug 9 02:53:06 UTC 2004


----- Forwarded message from owner-or-dev at freehaven.net -----

Date: Sun, 8 Aug 2004 18:47:08 -0400 (EDT)
From: Anna Shubina <ashubina at tahoe.cs.dartmouth.edu>
To: or-dev at freehaven.net
Subject: Re: Servers with Dynamic IPs

That may be mostly true, but it may not be true in 100% of cases. (If it
was true in my case, sorry for ruining the performance of your network.) The
choice of giving a machine a dynamic IP is an administrative choice, which
may be made even for high-bandwidth and low-latency connections because of
perceived administration overhead.

My machine, althing (formerly poblano), just went from having a dynamic IP
to having a static IP. It is sitting on the same subnet of the same
college network, and I do not believe anything changed in terms of its
bandwidth and latency. (True, it did not change its IP often, but it was a
dynamic IP nevertheless.)

I know it caused some problems for you before, but I thought these were
due to DNS and the general unreliability of our whole network for a while,
rather than low bandwidth and high latency?

On Sun, 8 Aug 2004, Roger Dingledine wrote:

> I've always assumed that dynamic IPs implies low reliability, low
> bandwidth, and high latency. Am I wrong?

----- End forwarded message -----



More information about the tor-dev mailing list