[comp.protocols.time.ntp] What happend

Mills@udel.edu (10/08/90)

Folks,

How can all these tickers be out at the same time? The receivers
on the east coast are doing fine and they are the furthest from the
transmitter.

Dave

----- Forwarded message # 1:

Received: from louie.udel.edu by huey.udel.edu id aa03936; 7 Oct 90 17:25 EDT
Received: from trantor.umd.edu by louie.udel.edu id aa05105; 7 Oct 90 17:23 EDT
Received: by trantor.umd.edu (5.64/1.34)
	id AA14058; Sun, 7 Oct 90 17:16:42 -0400
Received: from hermes.chpc.utexas.edu by trantor.umd.edu (5.64/1.34)
	id AA14054; Sun, 7 Oct 90 17:16:36 -0400
Received: by hermes.chpc.utexas.edu (5.61/SMI-3.2)
	id AA09113; Sun, 7 Oct 90 16:16:47 -0500
Date: Sun, 7 Oct 90 16:16:47 -0500
From: Bill Jones <jones@hermes.chpc.utexas.edu>
Message-Id: <9010072116.AA09113@hermes.chpc.utexas.edu>
To: ntp@trantor.umd.edu
Subject: What happend


fuzz.sdsc.edu, truechimer.cso.uiuc.edu, ncarfuzz.ucar.edu and
wwvb.isi.edu and probly other WWVB clocks are all running at 
stratum 2 and have a lot of baddata errors?

Bill Jones
UT SYSTEM CHPC

----- End of forwarded messages

CASNER@VENERA.ISI.EDU (Stephen Casner) (10/08/90)

Dave,
	I don't know why, but my Spectracom receiver is UNLOCKed,
i.e., not hearing WWVB.  In fact, I'm running the experiment I mentioned
to you, so I have two receivers hooked up, and both are unlocked.  They
have been in that condition long enough that they have drifted apart by
two seconds.

	Perhaps WWVB uses a collection of antennas each radiating in
a different direction, rather than one omnidirectional antenna?  Or
some really strange atmospheric condition?
							-- Steve
-------

Mills@udel.edu (10/08/90)

Steve,

Beats me. In the scheme of things a 60-kHz directive array would probably
represent a significant hazard to air navigation. However, I don't know
how else to explain the situation. Is your receiver in phase lock, as
against timecode okay? 

Dave

CASNER@VENERA.ISI.EDU (Stephen Casner) (10/08/90)

Dave,

Both receivers have lost phase lock, not just timecode sync.

Regarding the antenna hypothesis:  I see in "NBS Special Publication
432" that the WWVB antenna "is a 122-meter, top-loaded vertical
installed over a radial ground screen".  Doesn't sound directional to
me. 
						-- Steve
-------

Mills@udel.edu (10/08/90)

Steve,

Things still look okay here, although your radio still shows "?". I note
that both umd1 and apple show rather large offset errors in the hundreds
of milliseconds, although both show stratum one. Also, istd at SRI shows
large errors, too. Hark, the chronons have landed!

Dave

fair@APPLE.COM ("Erik E. Fair", Your Friendly Postmaster) (10/09/90)

Huh? At this hour, our billboard shows:

     remote           refid      st  when poll reach  delay  offset   disp
==========================================================================
*127.127.3.0     .WWV.            0    25   64  376    10.5   -0.71    0.2
 hpsdlz.sdd.hp.c apple.com        2     1   64  377   109.5    0.83    4.1
+pmax6.osf.org   .WWV.            1   491 1024  376   161.1    2.72    3.1
 norad.arc.nasa. wwvb.isi.edu     2    15   64  377    76.3 -754.12  519.3
 hermes.chpc.ute .WWV.            1    34   64  376   149.4  -11.06    2.7
 BITSY.MIT.EDU   .WWV.            1   995 1024  337   176.1   -2.64    0.9
 SW.MCC.COM      .WWV.            1   964 1024  376   173.5   12.79   66.8
 otc1.psu.edu    .WWV.            1   628 1024  376   209.6   -9.55    3.0
 gw.ccie.utoront suzuki.utcs.uto  2   544 1024  356   289.0    3.51   17.8
 dcn6.udel.edu   .WWV.            1   605 1024  373   467.3   -2.09    5.8
 clepsydra.dec.c 0.0.0.0         16 never 1024    0     0.0    0.00  64000
+fuzz.sdsc.edu   .WWVB.           1   438 1024  337    99.8    5.37    2.6
 wwvb.isi.edu    .WWVB.           1   199 1024  376   134.0   57.96    6.9
+ncarfuzz.ucar.e .WWVB.           1   950 1024  377   146.1   -3.15    2.2
 dcn5.udel.edu   .WWVB.           1   613  256  322   237.4   -1.16  504.5
+truechimer.cso. .WWVB.           1   945 1024  377   153.2   -4.59    4.0
 umd1.umd.edu    .WWVB.           1   949 1024  377   233.6   91.31    5.6
 dcn1.udel.edu   .WWVB.           1   326  512  352   232.0  -12.44    4.0
 suzuki.utcs.uto .CHU.            1   481 1024  377   276.3    5.73   13.4
 128.39.1.149    .ATOM.           1   241   64  170   490.0   30.57 7596.6
 Kermit.Stanford HOLMES.LCS.MIT.  3  1003 1024  376    55.0   -4.05    1.6
 largo.ig.com    apple.com        2    43   64  376    66.6   -6.32    3.6
 blake.u.washing apple.com        2    61   64  376    98.0   -0.15    1.3
 olivej.ATC.Oliv apple.com        2    48   64  376    93.6  -10.64    4.0
 NNSC.NSF.NET    apple.com        2    38   64  376   157.6   -0.23    6.1
 CS.ORST.EDU     apple.com        2    51   64  376   235.4   -8.94  688.1

Hundreds of milliseconds off? I don't think so... Maybe what you saw
was a network routing anomaly?

	Erik E. Fair	apple!fair	fair@apple.com

jones@HERMES.CHPC.UTEXAS.EDU (Bill Jones) (10/09/90)

The WWVB just started working agein around 12 AM.

BIll Jones

Mills@udel.edu (10/09/90)

Erik,

Apple has returned to nominal offsets as of now. I don't doubt that
routing troubles are at the root of the problem; however, of late
those problems are getting more severe and, in the case of umd1,
are severe enough for other primary tickers to toss the buggers
out. Having said that, even umd1 is at the moment showing only
nominal offsets. The trick now is to find the cause of the occasional
warps.

Finally, I note the strange west-coast warp of WWVB seems to have
subsided and now all primary tickers have resumed joy of ion. That, too
needs to be explained.

Dave

Mills@udel.edu (10/09/90)

Erik, (really to Milo),

Has norad.arc.nasa.gov sprung a hydrogen leak? Seems like 750-odd
milliseconds is pretty far to space. That ticker shows stratum 2,
which means its PSTI clock might also have come bum. ET fone home.

Dave