Sorry to ask this, but you double-triple checked that you are running the correct, self-complied binary, right? After a "make install" it should be in /usr/local/sbin while Ubuntu's ntpd is in /usr/sbin ...
Regards, Heiko Am 27.08.2010 15:40, schrieb Burkhardt, Frank: > Dear Heiko, > > thanks for your hint. > >> You definitely have no ATOM_PPS refclock support in your ntpd but it >> looks like your kernel supports PPS and it works fine. Please re-run >> configure and check its output. I would suspect that the configure >> script looks for certain include files in your kernel header dir(s) and >> does not find them. That results in disabling ATOM at configuration time >> (can be verified by looking into the generated config.h file of NTP). > > I checked the configure process but it has no errors, concerning ATOM: > > "checking ATOM PPS interface... yes" > > Also config.h shows: > > #define CLOCK_ATOM 1 > > And config.log displays: > > configure:29089: running /bin/bash ./configure --disable-option-checking > '--prefix=/usr/local' '--enable-ATOM' --cache-file=/dev/null --srcdir=. > ac_cv_var_atom_ok=yes > > Make shows no serious errors. > > Cheers, > Frank > > > > >> -----Ursprüngliche Nachricht----- >> Von: questions-bounces+frank.burkhardt=fokus.fraunhofer...@lists.ntp.org >> [mailto:questions- >> [email protected]] Im Auftrag >> von Heiko Gerstung >> Gesendet: Freitag, 27. August 2010 14:12 >> An: [email protected] >> Betreff: Re: [ntp:questions] ntptime didn´t show ppssignal >> >> Hi Frank, >> >> Am 23.08.2010 14:30, schrieb Burkhardt, Frank: >>> Dear Listmembers, >>> >>> first of all sorry for my bad english, but I'm not a native speaker >> and not trained so well ;-) >>> >>> I tried to get ntpd working with a pps signal (atom driver) under >> Ubuntu 10.04 since one week without success. I'm using a self compiled >> kernel 2.6.34.3 with pps support and pps_client ldisc built in. "pstest" >> shows following data, so I think it is generally working: >>> >>> r...@chaser:/home/fast# /export/pps-tools/ppstest /dev/pps0 >>> trying PPS source "/dev/pps0" >>> found PPS source "/dev/pps0" >>> ok, found 1 source(s), now start fetching data... >>> source 0 - assert 1282565685.198781684, sequence: 352669 - clear >> 1282213025.923967757, sequence: 10 >>> source 0 - assert 1282565686.198843801, sequence: 352670 - clear >> 1282213025.923967757, sequence: 10 >>> source 0 - assert 1282565687.198921096, sequence: 352671 - clear >> 1282213025.923967757, sequence: 10 >> >> That shows that your PPS support is working fine on kernel level. >> >>> I compiled ntpd with --enable-atom option, but when I start ntpd it >> shows following error in the log: >>> >>> 16 Aug 10:36:00 ntpd[17210]: refclock_newpeer: clock type 22 invalid >> This clearly shows that you have no ATOM_PPS refclock support. >> >>> 16 Aug 10:36:00 ntpd[17210]: 127.127.22.0 interface 127.0.0.1 -> >> (null) >>> >>> But i think the problem has maybe nothing to do with ntpd directly so >> far, because ntptime shows only PLL, NANO and not PPSXXXX in the status >> field: >>> >>> ntp_gettime() returns code 0 (OK) >>> time d01ce7a9.c30c885c Mon, Aug 23 2010 14:27:21.761, (.761910415), >>> maximum error 257004 us, estimated error 458 us >>> ntp_adjtime() returns code 0 (OK) >>> modes 0x0 (), >>> offset -42.572 us, frequency 17.326 ppm, interval 1 s, >>> maximum error 257004 us, estimated error 458 us, >>> status 0x2001 (PLL,NANO), >>> time constant 6, precision 0.001 us, tolerance 500 ppm >> I do not see this in the output of ntptime on my (PPS-enabled) system, I >> guess ntptime does not know about the newer PPS implementation. >> >>> So is the kernel not detecting the PPS, even though ppstest shows the >> signal etc.? >>> >>> I don´t know how I can proceed, miss I something else, do someone have >> any hints for me? >>> >>> Many thanks in advance and kind regards from Berlin, >>> Frank >>> >> >> You definitely have no ATOM_PPS refclock support in your ntpd but it >> looks like your kernel supports PPS and it works fine. Please re-run >> configure and check its output. I would suspect that the configure >> script looks for certain include files in your kernel header dir(s) and >> does not find them. That results in disabling ATOM at configuration time >> (can be verified by looking into the generated config.h file of NTP). >> >> Best Regards, >> Heiko >> >> _______________________________________________ >> questions mailing list >> [email protected] >> http://lists.ntp.org/listinfo/questions _______________________________________________ questions mailing list [email protected] http://lists.ntp.org/listinfo/questions
