From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vc0-f171.google.com (mail-vc0-f171.google.com [209.85.220.171]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 1004820221E for ; Mon, 20 Aug 2012 14:27:07 -0700 (PDT) Received: by vcdd16 with SMTP id d16so13625962vcd.16 for ; Mon, 20 Aug 2012 14:27:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=43rj3e7MQSgNzhNgqsciOj7AcdywiSVjjcoziVZOUG0=; b=N96fnxSP5wElCDB7FJBh0f9mZeEuqUotL1u6Mw137K1KwUILHtOGdUgPki+orvH7sx l0iMXqJTnWM6QTwVKYzWCga2PzaG9Nd6YtzENlvSmtC25zJivXIKSUpK5gsIDvjS2217 h5TUrwC+xrz7LoXQdnY7+4kdmVPT3mZSimIic0NWHDp1lHh1UvsKI/mQmMihMD/olTJK KaN9MHDVM59YSUcP2FmdYrZ0J/rvwtV5bAppQb3F9atBdW7f4UNNUYb5p8EJRqIZvi37 82PJw9OclClVsB01phi/JEdFl6DE/FmQkPE4/aMest3IDZCjvraOZ9vZiefL2e/g5ucn iqPQ== MIME-Version: 1.0 Received: by 10.52.95.46 with SMTP id dh14mr9693532vdb.114.1345498027031; Mon, 20 Aug 2012 14:27:07 -0700 (PDT) Received: by 10.58.231.234 with HTTP; Mon, 20 Aug 2012 14:27:07 -0700 (PDT) In-Reply-To: References: <502E064C.50305@etorok.net> <502E9609.5040800@etorok.net> <9246.1345321014@sandelman.ca> Date: Mon, 20 Aug 2012 17:27:07 -0400 Message-ID: From: George Lambert To: david@lang.hm Content-Type: multipart/alternative; boundary=20cf3071cc68f27dfe04c7b92bd9 Cc: cerowrt-devel@lists.bufferbloat.net Subject: Re: [Cerowrt-devel] cerowrt 3.3.8-17: nice latency improvements, some issues with bind X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 20 Aug 2012 21:27:08 -0000 --20cf3071cc68f27dfe04c7b92bd9 Content-Type: text/plain; charset=ISO-8859-1 Good point - that is just how I do it on my server network, sorry. reminds me of a good point Q: what is the different between in theory and in practice? A: In theory, NOTHING. ;-) so we test and iterate until we get it right in practice. G. On Mon, Aug 20, 2012 at 4:48 PM, wrote: > On Mon, 20 Aug 2012, George Lambert wrote: > > Check and set the time by syncing to NTP Servers - not user supplied times >> if the network >> is available. to see if they have set times > those set by NTP Server >> > > In theory you are right, in practice you are not. > > it's not uncommon for systems to point at a local set of timeservers (GPS > based for example), sometimes things go wrong with those servers, and so > people configure a local fallback (because they need the clocks on the > systems to remain consistant for things like kerberos to keep working). > This leads to a failure mode where if something goes wrong on that system, > the time can get set via NTP to some time in the future. > > There needs to be a way to recover from such conditions. > > The recent problems that people had with leap seconds is an indication > that even if you do use Internet NTP servers, sometimes things go wrong. > > David Lang > > -- P THINK BEFORE PRINTING: is it really necessary? This e-mail and its attachments are confidential and solely for the intended addressee(s). Do not share or use them without approval. If received in error, contact the sender and delete them. --20cf3071cc68f27dfe04c7b92bd9 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Good point - that is just how I do it on my server network, sorry.=A0
<= br>
reminds me of a good point=A0

Q: wha= t is the different between in theory and in practice?=A0
A: In th= eory, NOTHING.=A0

;-) so we test and=A0iterate until we get it right in p= ractice.=A0=A0


G.


On Mon, Aug 20, 2012 at 4:48 PM, <davi= d@lang.hm> wrote:
On Mon, 20 Aug 2012, Georg= e Lambert wrote:

Check and set the time by syncing to NTP Servers - not user supplied times<= br> if the network
is available. to see if they have set times > those set by NTP Server

In theory you are right, in practice you are not.

it's not uncommon for systems to point at a local set of timeservers (G= PS based for example), sometimes things go wrong with those servers, and so= people configure a local fallback (because they need the clocks on the sys= tems to remain consistant for things like kerberos to keep working). This l= eads to a failure mode where if something goes wrong on that system, the ti= me can get set via NTP to some time in the future.

There needs to be a way to recover from such conditions.

The recent problems that people had with leap seconds is an indication that= even if you do use Internet NTP servers, sometimes things go wrong.

David Lang




--
P THINK BEFO= RE PRINTING: is it really necessary?

This e-mail and its attachments= are confidential and solely for the
intended addressee(s). Do not share= or use them without approval. If received in error, contact the sender
and delete them.
--20cf3071cc68f27dfe04c7b92bd9--