From owner-freebsd-hackers  Thu Oct 23 01:57:11 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id BAA11712
          for hackers-outgoing; Thu, 23 Oct 1997 01:57:11 -0700 (PDT)
          (envelope-from owner-freebsd-hackers)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id BAA11687;
          Thu, 23 Oct 1997 01:57:05 -0700 (PDT)
          (envelope-from bde@zeta.org.au)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.7/8.6.9) id SAA29257; Thu, 23 Oct 1997 18:53:26 +1000
Date: Thu, 23 Oct 1997 18:53:26 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199710230853.SAA29257@godzilla.zeta.org.au>
To: freebsd-hackers@FreeBSD.ORG, freebsd-stable@FreeBSD.ORG, joes@seaport.net
Subject: Re: Interesting behaviour from sysctl(kern.boottime)
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>But, I've come to notice the following:  (five executions worth of data:)
>
>----------------- cut here --------------
>System time is (877528801) Wed Oct 22 07:00:01 1997
>System Booted at (877407594) Mon Oct 20 21:19:54 1997
>                             ^^^^^^^^^^^^^^^^^^^
>Difference is: 121207 seconds
>-----
>System time is (877532401) Wed Oct 22 08:00:01 1997
>System Booted at (877407591) Mon Oct 20 21:19:51 1997
>                             ^^^^^^^^^^^^^^^^^^^
>Difference is: 124810 seconds
>-----
>...
>I run xntpd.  Could that be a factor? (updating the clock ticks or whatever?)

Yes, it changes the boot time whenever it calls settimeofday() to step
the clock.  settimeofday() always adjusts `boottime' when it adjust
`time'.  This is correct when both were wrong originally, but completely
wrong if the clock has drifted.  A drift of 3 seconds per day is good
if xntpd is _not_ used, but xntpd not step the clock if it is correctly
configured and the system is always connected.

Bruce