Re: SNTP implementation recommendations (was "Re: [ntpwg] SNTP clients")
mayer@gis.net Thu, 28 April 2005 17:06 UTC
Return-Path: <mayer@gis.net>
X-Original-To: ntpwg@lists.ntp.isc.org
Delivered-To: ntpwg@lists.ntp.isc.org
Received: from localhost (localhost [127.0.0.1]) by ntp1.ntp.isc.org (Postfix) with ESMTP id B471539A69 for <ntpwg@lists.ntp.isc.org>; Thu, 28 Apr 2005 17:06:09 +0000 (UTC) (envelope-from mayer@gis.net)
Received: from ntp1.ntp.isc.org ([127.0.0.1]) by localhost (ntp1.isc.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 73417-03 for <ntpwg@lists.ntp.isc.org>; Thu, 28 Apr 2005 17:06:07 +0000 (UTC)
Received: from gis.net (mx03.gis.net [208.218.130.11]) by ntp1.ntp.isc.org (Postfix) with ESMTP for <ntpwg@lists.ntp.isc.org>; Thu, 28 Apr 2005 17:06:01 +0000 (UTC) (envelope-from mayer@gis.net)
Received: from gis.net ([208.218.130.24]) by mx03.gis.net; Thu, 28 Apr 2005 13:05:59 -0400
Sender: mayer@gis.net
From: mayer@gis.net
To: plonka@doit.wisc.edu, ntpwg@lists.ntp.isc.org
Subject: Re: SNTP implementation recommendations (was "Re: [ntpwg] SNTP clients")
X-Mailer: Galaxy AstroMail used by mayer
X-Originating-IP: 198.22.153.244
Date: Thu, 28 Apr 2005 13:05:58 -0400
Priority: normal
Message-id: <427117f6.39c.3f56.15800@gis.net>
X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on ntp1.isc.org
X-Spam-Status: No, hits=0.4 tagged_above=-999.0 required=5.0 tests=NO_RDNS2, NO_REAL_NAME, TW_TP
X-Spam-Level:
Cc:
X-BeenThere: ntpwg@lists.ntp.isc.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: mayer@gis.net
List-Id: IETF Working Group for Network Time Protocol <ntpwg.lists.ntp.isc.org>
List-Unsubscribe: <https://lists.ntp.isc.org/mailman/listinfo/ntpwg>, <mailto:ntpwg-request@lists.ntp.isc.org?subject=unsubscribe>
List-Archive: <https://lists.ntp.isc.org/mailman/private/ntpwg>
List-Post: <mailto:ntpwg@lists.ntp.isc.org>
List-Help: <mailto:ntpwg-request@lists.ntp.isc.org?subject=help>
List-Subscribe: <https://lists.ntp.isc.org/mailman/listinfo/ntpwg>, <mailto:ntpwg-request@lists.ntp.isc.org?subject=subscribe>
X-List-Received-Date: Thu, 28 Apr 2005 17:06:10 -0000
Probably the most important thing to include is that KOD packets need to be obeyed. In IETF terms this is a MUST. There are far to many NTP/SNTP clients that do not know or care about KOD packets and continue to make requests are they are told to go away. In addition there are clients that will continuously send requests as frequently as 1 second apart and will ignore that fact that the servers are dropping their packets or telling it to go away, see above. The CSIRO incident a year or so ago points out that when CSIRO changed their NTP server address, not only did the traffic not go away, the traffic actually increased as clients became even more insistent in getting their time from a non-existent server. SNTP clients, in short, need to be good net citizens and understand both responses and lack of responses. This goes for ntp servers too. Danny ----- Original Message Follows ----- > Hi Jason, > > On Thu, Apr 28, 2005 at 09:37:24AM -0400, Jason Terry wrote: > > > > I am trying to improve an SNTP client implementation. I was > > wondering if anyone could point me to a resource that has > > information on the real-world issues developers face when > implementing an SNTP client. > > I made some comments/recommendations in this BCP RFC (numbers yet to > be assigned): > > > http://www.ietf.org/internet-drafts/draft-ietf-grow-embed-addr-05.txt > > Also, I believe one of the vendor consortiums (perhaps DOCSIS?) > codified some of our recommendations in their specs, but I've lots the > reference. If someone knows where I can find it, please let me know. > > In addtion to RFC2030: > > ftp://ftp.rfc-editor.org/in-notes/rfc2030.txt > > we worked on this updated draft which is probably closest to what you > want: > > > http://www.eecis.udel.edu/~mills/database/rfc/draft-mills-sntp-v4-00.txt > > Dave > > P.S. You might also find this backgroun info pertient - in my > technical report here: > > http://www.cs.wisc.edu/~plonka/netgear-sntp/ > > and in the related presentations, for instance at LISA and NANOG: > > http://net.doit.wisc.edu/~plonka/lisa/lisa2003/ > http://www.nanog.org/mtg-0310/plonka.html > > -- > plonka@doit.wisc.edu http://net.doit.wisc.edu/~plonka ARS:N9HZF > Madison, WI _______________________________________________ > ntpwg mailing list > ntpwg@lists.ntp.isc.org > https://lists.ntp.isc.org/mailman/listinfo/ntpwg
- [ntpwg] SNTP clients Jason Terry
- SNTP implementation recommendations (was "Re: [nt… Dave Plonka
- Re: SNTP implementation recommendations (was "Re:… mayer