[3474] in testers

home help back first fref pref prev next nref lref last post

Re: sun4 [8.2.0]: time syncronization

daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Sat May 23 00:17:18 1998

From: Jonathon Weiss <jweiss@MIT.EDU>
To: Greg Hudson <ghudson@MIT.EDU>
Cc: Jonathon Weiss <jweiss@MIT.EDU>, testers@MIT.EDU
In-Reply-To: Your message of "Sat, 16 May 1998 13:20:43 EDT."
             <199805161720.NAA25132@small-gods.MIT.EDU> 
Date: Sat, 23 May 1998 00:17:12 EDT

> miki said you have a workaround in place for this; what was it?

a while 1 that slept for a minute and ran fs checks.  I just needed
something to let me use the machine, I'm not even sure I'd call this a
work-around tho. :-)

> miki also gave me a paper suggesting that there can be a conflict
> between xntpd and the "dosynctodr" variable.  If you put
> 
> 	set dosynctodr = 0
> 
> in /etc/system, and disable your other workarounds, do things work
> better?

Yeah, it works somewhat better.  When I booted the machine, it came
up, xntpd sync'd to the reouter and things looked like they were going
to be happy.  However, now, every 7 minutes or so I get 3 syslogs in
rapid succession:

SYSLOG, speaker-for-the-dead, daemon.info, root@SPEAKER-FOR-THE-DEAD 
May 21 12:32:04 [speaker-for-the-dead] xntpd[422]: synchronized to 18.177.0.1, stratum=2

SYSLOG, speaker-for-the-dead, daemon.notice, root@SPEAKER-FOR-THE-DEAD 
May 21 12:32:05 [speaker-for-the-dead] xntpd[422]: time reset (step) 1.386642 s

SYSLOG, speaker-for-the-dead, daemon.info, root@SPEAKER-FOR-THE-DEAD 
May 21 12:32:05 [speaker-for-the-dead] xntpd[422]: synchronisation lost

The next thing I'm going to try is not touching dosynctodr, but not
running xntpd either.

[Later] I tried this, and after one or two clock sets, afs stopped
mucking with the clock, and the time remains correct.

	Jonathon


home help back first fref pref prev next nref lref last post