[ntp:questions] Re: mystery leap second

wa6zvp wa6zvp at gmail.com
Sat Dec 31 16:01:06 UTC 2005


Wolfgang S. Rupprecht wrote:
> Can someone shed any light on what is going on here?  Could
> clock.sjc.he.net have jumped the gun and inserted a leap second a day
> early?  I've never seen it off by more than a millisecond or two
> before and the offset of almost exactly a second is very suspicious,
> especially since we are so close to a leap second event.
>
>      remote           refid      st t when poll reach   delay   offset  jitter
> ==============================================================================
> *GPS_ONCORE(0)   .GPS.            0 l   14   16  377    0.000    0.004   0.002
> -bonnet.wsrcc.co 25.96.172.67     2 u   34   64  377    0.250   -0.473   0.390
> xclock.sjc.he.ne .CDMA.           1 u    7   64  377   15.278  -999.34   0.953
>  LOCAL(0)        LOCAL(0)        10 l   45   64  377    0.000    0.000   0.001
>
> -wolfgang
> --
> Wolfgang S. Rupprecht                http://www.wsrcc.com/wolfgang/
> Direct SIP URL Dialing: http://www.wsrcc.com/wolfgang/phonedirectory.html

Sure looks like a software bug in the reference clock.  They do not
have additional backup references
to protect against false ticking.  Here is the current status:

$ ntpq -c cv clock.sjc.he.net
status=0606 clk_badtime, last_clk_badtime
device="EndRun Technologies Praecis CDMA",
timecode="2005 365 15:41:02.644978", poll=415464, noreply=1,
badformat=0, baddata=1, fudgetime1=0.000, stratum=0, refid=CDMA,
flags=2

I suppose it could be an error in the actual reference clock in the
cellsite this thing
is listening to which is likely a GPS rcvr.

Humm. . . My PCS phone is a second slow right now, but I've never
actually
checked it to the second before. . .

 Roger (at hmc.edu)




More information about the questions mailing list