Message-Id: <aad2ea6f120210041e6d@[198.114.157.23]>
Date: Tue, 25 Oct 1994 15:40:44 -0400
To: Patrik Faltstrom <paf@nada.kth.se>
From: John Curran <jcurran@nic.near.net>
Subject: Re: Port for URN->URC resolution
At 2:28 AM 10/25/94, Patrik Faltstrom wrote:
>At 04.25 94-10-25, Rich Salz wrote:
>>It is foolish to think that a network service is secure because the
>>server is listening on a particular port, or because the server requires
>>clients to originate on a particular port. ...
>
>True, but I do want on my computer that a service like this is running
>on a secure port even though it is not as secure as the word "secure port"
>implies.
>
>IF my server daemon dies, I don't want any user on my system to be able to
>start a new daemon.
I believe that we should obtain a port which is less than <1024, so that those
people who have operating system protections against binding to such ports
can use such facilities. Additionally, many sites (for better or worse) may
wish to block access to the internal URN resolution services using commonly
accepted techniques (such packet filtering "low ports" from external access).
/John