[ntp:questions] timing issue with a HP 58534A
hart at ntp.org
Sun Feb 5 19:47:46 UTC 2012
On Sun, Feb 5, 2012 at 19:23, Mark C. Stephens <marks at non-stop.com.au> wrote:
> After capturing data and computing average position for location for
> a few days I put it into timing mode and locked the co-ords in.
> Now the refclock driver says its unreachable, even though I can see
> (looks like) valid data ticking in with -d.
> I have switched the GPS back to survey mode and all is good again,
> looks like I can't use timing mode with this driver.
I'm curious what's going wrong when you have the HP GPS in timing
mode. While it's possible using -d (or -dd, or -ddd, or equivalently
-D2 or -D3) might be needed to isolate the cause, it's likely poking
with ntpq will reveal all the information needed, using:
ntpq -c "rv &1"
ntpq -c "cv &1"
which display the generic peer variables and/or refclock-specific
variables for the first peer/association. If you haven't already,
enable clockstats (such as with "statistics loopstats clockstats" in
ntp.conf) so the actual timecode received is logged, and do so
_before_ switching back to position-locked timing mode. Perhaps the
difference that triggers the problem will become obvious.
More information about the questions