Steve, Steve Kostecke wrote: > On 2009-01-22, Martin Burnicki <[email protected]> wrote: > >> Steve Kostecke wrote: >> >> [---=| Quote block shrinked by t-prot: 35 lines snipped |=---] >>> GPS, DCF >>> 39 hopf clock drivers by ATLSoft HOPF >>> (default), >>> GPS, DCF >>> 40 JJY Receivers JJY >>> 42 Zyfer GPStarplus Receiver GPS >>> 43 RIPE NCC interface for Trimble Palisade RIPENCC >>> 44 NeoClock4X NEOL >> >> AFAIK the refid (as reported in the ntpq -p billboard) can only have up >> to 4 characters since it's transported in a 32 bit value in the NTP >> packet, so it's not possible to set it e.g. to "OMEGA" or "DUMBCLOCK". > > Please see the reference IDs listed on the following pages (among > others): > > http://www.eecis.udel.edu/~mills/ntp/html/drivers/driver5.html > http://www.eecis.udel.edu/~mills/ntp/html/drivers/driver33.html > http://www.eecis.udel.edu/~mills/ntp/html/drivers/driver43.html > > I assembled a list based on the Official Distribution Documentation > to provide some useful information to our news-group readers in place > of the usual dismissive reply of "RTFM".
Yes, it's a good idea, but unfortunately there seem to be quite some discrepancies between those docs and the current code. A quick grep over the ntp-dev sources shows that for some refids just the first 4 characters are used on the wire (e.g. "DUMBCLOCK" -> "DUMB") whereas for some refclock types the refid in the code just differs from the docs (e.g. "RIPENCC" -> "GPS\0". > If there are errors in the documentation please contact the > documentation maintainer. We both know it's not easy to submit fixes for the docs, and in case of the refclocks it may be even worse since most refclocks have different maintainers, but the docs are on Dave's server. Martin -- Martin Burnicki Meinberg Funkuhren Bad Pyrmont Germany _______________________________________________ questions mailing list [email protected] https://lists.ntp.org/mailman/listinfo/questions
