Re: ietf-nntp Initial draft FINALLY available
USENET news manager <newsmaster@ucs.cam.ac.uk> Thu, 03 October 1996 09:31 UTC
Received: from cnri by ietf.org id aa29465; 3 Oct 96 5:31 EDT
Received: from PHEASANT.ACADEM.COM by CNRI.Reston.VA.US id aa05644;
3 Oct 96 5:31 EDT
Received: (from majordom@localhost) by pheasant.ACADEM.COM (8.7.5/8.7.3) id
EAA27101 for ietf-nntp-outgoing; Thu, 3 Oct 1996 04:23:17 -0500
X-Authentication-Warning: pheasant.ACADEM.COM: majordom set sender to
owner-ietf-nntp using -f
Received: from academ.com (root@ACADEM.COM [198.137.249.2]) by
pheasant.ACADEM.COM (8.7.5/8.7.3) with ESMTP id EAA27097 for
<ietf-nntp@PHEASANT.ACADEM.COM>; Thu, 3 Oct 1996 04:23:14 -0500
Received: from lyra.csx.cam.ac.uk (news@lyra.csx.cam.ac.uk [131.111.8.36]) by
academ.com (8.7.6/8.7.1) with SMTP id EAA12451 for <ietf-nntp@academ.com>;
Thu, 3 Oct 1996 04:23:11 -0500 (CDT)
Received: by lyra.csx.cam.ac.uk (SMI-8.6/MDTG-V1.1.8@lyra.csx.cam.ac.uk)
id KAA26292; Thu, 3 Oct 1996 10:21:39 +0100
Message-Id: <199610030921.KAA26292@lyra.csx.cam.ac.uk>
Subject: Re: ietf-nntp Initial draft FINALLY available
To: Evan Champion <evanc@synapse.net>
Date: Thu, 3 Oct 1996 10:21:37 +0100 (BST)
Cc: ietf-nntp@academ.com
In-Reply-To: <32532559.4D2B@synapse.net> from "Evan Champion" at Oct 2,
96 10:30:49 pm
From: USENET news manager <newsmaster@ucs.cam.ac.uk>
X-Mailer: ELM [version 2.4 PL22]
Content-Type: text
Sender: owner-ietf-nntp@academ.com
Precedence: bulk
Evan Champion wrote:
>Stan Barber wrote:
>> What is the point of having a IANA assigned port number then? What if I
>> chose to run an HTTP server on port 119?
>
>As I said in my last note, the reason is IMHO for interoperability
>_only_.
>...
>One way to think of it is to divorce protocol from transport. There is
>no technical reason why NNTP _must_ run on 119/TCP (just like there is
>no technical reason why NNTP must run over TCP, or that the servers be
>connected by IP... )
>
>If there is a technical justification for something then it should be
>said with a MUST. If something is being said as a suggestion (even a
>strong suggestion) it should be said with a SHOULD. IMHO MUST does not
>apply here, but SHOULD certainly does.
I agree in general terms, but would place the emphasis slightly differently.
Since port 119/TCP is the standard port, any news software which could not
run on that port would de facto be useless for most purposes, yet without
specifying it as SHOULD, an implementation could choose to use a non-standard
port (for some bizarre reason or other), possibly (especially if no source
code supplied) with no way for you to change it, and claim they conformed to
the standard.
On the other hand, the ability to use an alternative port is very useful for
testing (you can run a second, test server on the same system as your live
server, for example), and some people have tried running INN with innd (to
handle feeds) on a different port, with nnrpd (to handle readers) on the
standard port - by arrangement with their peer feed sites, of course. Full
use of non-standard ports does, of course, require not only the server to
listen on the non-standard port but readers and peer-server feeding
mechanisms to be able to send to it - selecting the target port as required
for a particular connection.
I therefore feel there should be strong encouragement to support that type
of flexibility, rather than mentioning only support for the standard port,
and suggest that the requirements should amount to
* NNTP client and server software MUST support using the IANA-defined
standard port for their NNTP dialogues, and should use that port by
default. [Mmm... does that mean they should ignore the system's
/etc/services or equivalent, if someone changes the nntp definition? I
guess that counts as an explicit local override, with 119/TCP used if
nothing externally defines something different.]
* they SHOULD be capable of using any other valid port (in principle;
clearly not in reality if at the time the port is in use for something
else!) by explicit configuration (and per-connection in the case of
software opening connections to news servers, since different servers
may run on different ports; one overall control is not sufficient).
Or putting that another way, they shouldn't include arbitrary limitations on
which ports can be used, should provide a way to select a non-default port
for listening and (per connection) as the target for reading or feeding
connections, and should default to the standard port for the protocol.
John Line
--
Cambridge University Computing Service - USENET news manager. Usually John Line
newsmaster@ucs.cam.ac.uk (alias {newsmaster,news,usenet}@news.cam.ac.uk)
- ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Ben Polk
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Ben Polk
- Re: ietf-nntp Initial draft FINALLY available Evan Champion
- Re: ietf-nntp Initial draft FINALLY available Chris Caputo
- Re: ietf-nntp Initial draft FINALLY available James Fidell
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Brian Kantor
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Rich Salz
- Re: ietf-nntp Initial draft FINALLY available Brian Kantor
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Ramanan
- Re: ietf-nntp Initial draft FINALLY available Brian Kantor
- Re: ietf-nntp Initial draft FINALLY available Rich Salz
- Re: ietf-nntp Initial draft FINALLY available Rich Salz
- Re: ietf-nntp Initial draft FINALLY available Nat Ballou
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Bob Sloane
- Re: ietf-nntp Initial draft FINALLY available Brian Kantor
- Re: ietf-nntp Initial draft FINALLY available Thomas 'Mike' Michlmayr
- Re: ietf-nntp Initial draft FINALLY available Ben Polk
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available Bob Sloane
- Re: ietf-nntp Initial draft FINALLY available Stan Barber
- Re: ietf-nntp Initial draft FINALLY available William H. Magill
- Re: ietf-nntp Initial draft FINALLY available Brian Kantor
- Re: ietf-nntp Initial draft FINALLY available Evan Champion
- Re: ietf-nntp Initial draft FINALLY available USENET news manager
- Re: ietf-nntp Initial draft FINALLY available Evan Champion
- Re: ietf-nntp Initial draft FINALLY available chris (c.) lewis
- Re: ietf-nntp Initial draft FINALLY available William H. Magill
- Re: ietf-nntp Initial draft FINALLY available Ben Polk
- ietf-nntp XHDR versus XPAT USENET news manager