Re: [ntp:questions] create charts
Hi there On 23/08/2020 14:10, Uwe Klein wrote: Anybody else getting "request received" from TheFork and a bunch of "undeliverable" from uscc.net for each posting to comp.protocols.time.ntp ? I got some canned replies from various help desks, including the fork's, claiming that I contacted them. On topic, something like this perhaps? http://www.sput.nl/ntpstats/ Uses bash, grep, awk and rrdtool. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Local Time NTP Server
Hi there On 24/08/2020 16:07, William Unruh wrote: It was renamed because UTC has nothing to do with Greenwich. For historical reasons, the time at Greenwich is the same as UTC. They are not perfectly identical. The difference is however less then one second; GMT is mean solar time. UTC is TAI (atomic time) + leap seconds. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Embedded solutions
Hi there Rob van der Putten wrote: A lot of people however, by an embedded system, hook op a GPS receiver, find that PPS doesn't work and then just give up. Apparently GPSD supports PPS on CTS. So if you already have got an embedded system and a GPS receiver and your 232 cape supports CTS, this might be the way to go. PPS on CTS might be a compile time option though. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Embedded solutions
Hi there Paul wrote: One normally uses a so-called GPIO pin to read PPS on systems that lack a DCD line or a parallel port. E.g. BeagleBone or Raspberry Pi. Obviously. A lot of people however, by an embedded system, hook op a GPS receiver, find that PPS doesn't work and then just give up. It would be nice to have a comprehensive overview of embedded PPS implementations. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Embedded solutions
Hi there Jaap Winius wrote: Has anyone here managed to turn a relatively cheap, ARM-based embedded system with a serial port into a decent stratum 1 NTP server? Thus far I've always attached my GPS and radio time signal receivers to much larger x86 hardware platforms, but those machines have other things to do and make the NTP server less stable than it can be. However, if I were to use dedicated hardware for the NTP server, I'd rather it power- efficient and as cheap as possible. I've looked at the BeagleBone Black (with an RS232 Cape) AFAIK the BBB 232 cape doesn't support DCD, so PPS is not available. and the Wandboard (both ARM platforms), Same problem. but have not had any success with them. There are stories stories of people who have done it with Soekris hardware (x86), but that's much more expensive. AFAIK Soekris has a 'regular' serial port. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] convert_rawdcf: parity check FAILED on Olinuxino Maxi (iMX233)
Hi there Rob wrote: I use: server 127.127.8.0 mode 6 I use server 127.127.8.0 mode 5. The symlink is to refclock-0. Pulses are 100 or 200 ms long. At 8 bits, one stopbit, no parity and 50 bps the 200 ms pulse will cause a framing error; -+ +---+---+---+---+---+---+---+---+- | S | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | S +---+---+---+---+---+---+---+---+---+ . . /|\ /|\ | | Check; Is bit 0?Check; Is bit 1? If not; Framing error If not; Framing error A '0' is coded as a 100 ms pulse (the ports reads 240 / 0xF0); -+ +- | 100 ms | +---+ A '1' is coded as a 200 ms pulse (the port reads 0); -+ +- | 200 ms| +---+ . /|\ | Framing error Using 8 bits + parity prevents framing errors but introduces parity errors instead; -+ +---+---+---+---+---+---+---+---+---+- | S | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | P | S +---+---+---+---+---+---+---+---+---+---+ -+ +- | 100 ms | +---+ -+ +- | 200 ms| +---+ With a 100 ms pulse there are 5 ones; Odd With a 200 ms pulse there are 0 ones: Even. So parity errors are perfectly normal. Regards, Rob ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] PPS and NMEA
Hi there Richard B. Gilbert wrote: The the PPS (assuming we are talking about the Pulse from a GPS Timing Receiver) is specified to be within plus or minus 50 nanoseconds. It depends on the receiver. I think mine is ± 1 µs. Regards, Rob -- The end is near; http://www.ripe.net/internet-coordination/ipv4-exhaustion/ipv4-available-pool-graph ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] PPS and NMEA
Hi there ksprabha wrote: Kindly let me know what is the difference between NTP with PPS and NTP with out PPS. NMEA can be a bit off with some GPS receivers. PPS tends to be very accurate. When using RS232, NMEA is send via a data line. PPS via the DCD status line. A few graphs; http://www.sput.nl/ntpstats/parents/ Using the following setup; NMEA and PPS; http://www.sput.nl/time/garmin.html DCF77 [1]; http://www.sput.nl/time/dcf77.html With these receivers DCF77 so more accurate then NMEA. PSS is more accurate then DCF77 [1] http://en.wikipedia.org/wiki/DCF77 Regards, Rob -- The end is near; http://www.ripe.net/internet-coordination/ipv4-exhaustion/ipv4-available-pool-graph ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there Edward T. Mischanko wrote: Question: If I'm using a GPS (Garmin 18x LVC) for a reference clock, Is the NMEA time stamp already in UTC? Yes. It does GPS time as well; http://www8.garmin.com/manuals/425_TechnicalSpecification.pdf If my time stamp is already UTC, then won’t I double leap if I have a leap file specified in my ntp.conf? No. From the NTPD doc; If the latest leap is in the past, nothing further is done other than to install the TAI offset. If the leap is in the future less than 28 days, the leap warning bits are set. If in the future less than 23 hours, the kernel is armed to insert one second at the end of the current day. If the kernel is enabled, the leap is done automatically at that time; otherwise, the clock is effectively stopped for one second at the leap. Additional details are in the The NTP Timescale and Leap Seconds white paper. If none of the above provisions are available, dsependent servers and clients tally the leap warning bits of surviving servers and reference clocks. When a majority of the survivors show warning, a leap is programmed at the end of the current month. During the month and day of insertion, they operate as above. In this way the leap is is propagated at all dependent servers and clients. 'surviving servers and reference clocks' are those with a '+' or '*' in a 'ntpq -p' overview. Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there Pierre Dubuc wrote: You can get it from one of the NIST mirrors: ftp://utcnist2.colorado.edu/pub/ It hasn't been updated yet. This one is; ftp://utcnist.colorado.edu/pub/ It's leap-seconds.3535228800 Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there unruh wrote: Not sure what you are saying. "it does"-- what does? And what does 'right' mean here. 'right' is just a directory name. 'it does' means the file contains leap second info; file /usr/share/zoneinfo/posix/Europe/Amsterdam /usr/share/zoneinfo/posix/Europe/Amsterdam: timezone data, version 2, 13 gmt time flags, 13 std time flags, no leap seconds, 180 transition times, 13 abbreviation chars file /usr/share/zoneinfo/right/Europe/Amsterdam /usr/share/zoneinfo/right/Europe/Amsterdam: timezone data, version 2, 13 gmt time flags, 13 std time flags, 24 leap seconds, 180 transition times, 13 abbreviation chars I have no idea how to extract leap second info from this file. Does ntpd actually use files from the /usr/share/zoneinfo/right/ tree? I'm used to files like leap-seconds.3427142400. If you want to run the French Revolutionary time, 100sec/min, 100min/hr, 10 hr/day, etc, go ahead. Months in alphabetical order would be nice. That you will be out of step with everyone else is your problem. All I am saying is the "announcement" that a leap second will occur 6 months from now gives you and everyone else lots of time to plan for it. And the txdata2012a will contain the information necessary for making those plans, whatever they are. ntpd is a way of synchronizing your clock with everyone else's. That is all. What you do with that information is up to you. If you want your machine to add 35 sec to the time it reads from the computer clock, use the leapseconds file to do that, just as you now use the zoneinfo file to translate your machine's UTC to the time you actually live by, including daylight savings. If you want your machine time to be on TAI, you will have to find some ntp server somewhere that is also on TAI to synchronize to. That that might be hard is just the price you pay for being out of step with everyone else. It is only someone who is crazy who says that when they are out of step with everyone else, it is everyone else that should change to get into step with them. It would have to be implemented in a way that makes sense. A new standard perhaps. Is anyone into this sort of thing? Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there unruh wrote: Do not work what way? What I meant was that it would be nice to run TAI instead of UTC. It is announced now, it occurs Jun 30. The tzdata database contains a file called "leapseconds" which contains all of the leapseconds which have occured or are know to occur in the future. In 'right' (based on the International Atomic Time) it does, in 'posix' (based on the Coordinated Universal Time) it doesn't. Does anyone use 'right'? Is this supported by NTPD? Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there unruh wrote: Probably due to the lawsuit. It will presumably be in the next release (tzdata2012a) of the tzdata database. www.iana.org/time-zones There is no great hurry. It is still 6 months off. Unfortunately, leap seconds don't work this way. Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Leap second
Hi there Danny Mayer wrote: This is the first time that I remember a leap second being added in the middle of the year instead of the end of December. Am I wrong? From leap-seconds.3427142400; 2272060800 10 # 1 Jan 1972 2287785600 11 # 1 Jul 1972 2303683200 12 # 1 Jan 1973 2335219200 13 # 1 Jan 1974 2366755200 14 # 1 Jan 1975 2398291200 15 # 1 Jan 1976 2429913600 16 # 1 Jan 1977 2461449600 17 # 1 Jan 1978 2492985600 18 # 1 Jan 1979 2524521600 19 # 1 Jan 1980 2571782400 20 # 1 Jul 1981 2603318400 21 # 1 Jul 1982 2634854400 22 # 1 Jul 1983 2698012800 23 # 1 Jul 1985 2776982400 24 # 1 Jan 1988 2840140800 25 # 1 Jan 1990 2871676800 26 # 1 Jan 1991 2918937600 27 # 1 Jul 1992 2950473600 28 # 1 Jul 1993 2982009600 29 # 1 Jul 1994 3029443200 30 # 1 Jan 1996 3076704000 31 # 1 Jul 1997 3124137600 32 # 1 Jan 1999 3345062400 33 # 1 Jan 2006 3439756800 34 # 1 Jan 2009 Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
[ntp:questions] Leap second
Hi there There's a leap second coming up; http://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat Where do I get a leap seconds file (leap-seconds.3550089600?)? Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Google and leap seconds
Hi there John Hasler wrote: Well, we have machines on and around Mars that need to keep time and that we need to synchronize with. In any case, the point is that planetary days and years are intervals to be measured, not clocks. I understand and share the desire to localize the representation of time, but that should be done explicitly. It made sense to synchronize our clocks with the Earth when we knew that it was a more stable oscillator than any we could build, but that is no longer true. Leap seconds don't correct an error: they jigger stable clocks so that they track a known drifty one. They should be treated as localization, not as error correction. Put them in tzdata. Indeed. Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Google and leap seconds
Hi there Richard B. Gilbert wrote: It's unfortunate that the earth DOES NOT rotate exactly 360 degrees in exactly 24. hours. This bit of poor design causes all sorts of problems. It's about 361°, 23:56:04.1 for 360°. Leap seconds are just one of the symptoms! It not just the earth's rotation. It's it's orbit as well. No two orbits are identical. Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Garmin 18 LVC high offset and jitter
Hi there Miguel Gonçalves wrote: Found out what the problem is... The device has definetely lost its configuration but isn't able to keep the saved configuration. I went to the configuration software and used the option "Get Configuration from GPS" then changed the PPS length to 200 ms then "Send Configuration to GPS". When I do it again it stays at 20 ms. I have an unusable GPS on the roof. :-( Can anyone help please? Use minicom to configure the Garmin. Tell NTPD not to send any data to the Garmin. Furthermore, check the data it's sends. Disable sentences you don't need. See; http://www.sput.nl/time/garmin.html Regards, Rob -- Wim T. Schippers for president http://nl.wikipedia.org/wiki/Bestand:Universiteit_Twente_Verzonken_Torentje_van_Drienerlo.jpg ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] Tobit LAN!Time DCF77 receiver not working
Hi there David Lord wrote: > I saw it as CTS on brown and RI on Orange Correct. > All refclocks docs I've checked, by no means all, expect serial > data on RxD. Rewiring the plug might help; Brown and Orange probably have opposite polarity; +-+ +-++ 12 V | | | | ---+ +---+ +--- - 12 V ---+ +---+ +--- + 12 V | | | | +-+ +-+- 12 V > < 0.1 or 0.2 seconds <> 1 or 2 seconds So the thing to do is to connect the positive going pulse to RXD (2). This will probably work with the following NTPD setting; server 127.127.8.0 mode 5 fudge 127.127.8.0 time1 0.220 refid DCF77 After a few minutes 'ntpq -p' should yield something like; remote refid st t when poll reach delay offset jitter = +GENERIC(0) .DCF7. 0 l 61 64 377 0.000 0.151 0.002 Regards Rob -- Nowadays people know the price of everything, and the value of nothing Oscar Wilde ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Tobit LAN!Time DCF77 receiver not working
Hi there Marc-Andre Alpers wrote: > I have no RS232 tester. The cover of the receiver is sealed. No screws. > http://666kb.com/i/bf7g1grgo1sbus2c1.jpg > > The connector inside: > http://666kb.com/i/bf7fxqd8cr17q7d8x.jpg That's female on the left and male on the right? > http://666kb.com/i/bf7g2xelg4xqppx3l.jpg That's male on the left and female on the right? > The transistor type is: > MC78L > 05ACP > M535 That's a 78L05; A 100 mA +5 V stabilizer IC. The circuit appears to be as follows; M F DCD 1 - 1 RXD 2 - 2 TXD 3 - 3 DTR 4 - 4 GND 5 --+-- 5 | | +--+-> Red GND | +-+-+ ++ S +---> Black +5V F | M+---+ DSR 6 |78L05 RTS 7 --+-- 7 CTS 8 > Orange RI 9 > Brown The circuit 'expects' NTPD to raise RTS and read from CTS or RI. As far as I can tell NTPD does raize RTS but does not read from CTS or RI. Regards, Rob -- Nowadays people know the price of everything, and the value of nothing Oscar Wilde ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Tobit LAN!Time DCF77 receiver not working
Hi there Marc-Andre Alpers wrote: > I can give you a Picture: > http://666kb.com/i/bf7er4kfwd3feibip.jpg Definitly not a a Conrad. Anyway, some specs would be nice. Lacking those a bit of reverse engineering. Have you tried a RS232 tester? Which LEDs are on? Which colour? Which one blinks? Regards, Rob -- Nowadays people know the price of everything, and the value of nothing Oscar Wilde ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Tobit LAN!Time DCF77 receiver not working
Hi there Marc-Andre Alpers wrote: > The Clock works correctly on windows with the programm DCF77_32.exe provided > on this site: http://www.rrs-web.net/in3her/dcf77_32.html This is about a Conrad DCF77 receiver. A Conrad DCF77 receiver doesn't have a LED. And NTPD usually receives data on RXD. Maybe you can post a link to the receiver you're using. Regards, Rob -- Nowadays people know the price of everything, and the value of nothing Oscar Wilde ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Which Linux kernels have PPS support
Hi there I read that kernel PPS support is now part of vanilla kernels (no patch required). Since when? Which is the oldest kernel with PPS support? Regards, Rob -- Nowadays people know the price of everything, and the value of nothing Oscar Wilde ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there pc wrote: > FWIW, it's possible that the antenna/receiver modules are manufactured > by this company: > > http://www.hkw-elektronik.de/ http://www.hkw-elektronik.de/englisch/products/products.php http://www.hkw-elektronik.de/englisch/products/assemblys.php Regards, Rob -- There are 709 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there Rob van der Putten wrote: > This will probably work; > >+-> DATA >| >| >| / c > +-+ b |/ > TXD >-+ 4k7 +--*-*---|BC547B > +-+ | | |\ > | | | \| e > | | -| > +++ --+-- | > 47 kΩ | | /|\ | > | | --+-- | > +++| 1N| > | | 4148 | > | | | > | | | > GND >--*-*---*-> GND If you want to build four, just multiply the resistor values by four (it will probably still work with 10 x these values). > You have to write software to drive it. This is for a parallel port; http://www.captain.at/rtai-dcf77-simulator.php Regards, Rob -- There are 707 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there Pete Ashdown wrote: > The clocks take input from a module like this: > > http://clausurbach.de/shop/product_info.php?info=p50_dcf77-module-for-our-nixie-clock-kits.html > > It looks pretty much line an antenna to me. So what I'd like is a preferably > cheap option to duplicate that signal to wire directly to the clocks instead, > sourcing from my Linux box. This will probably work; +-> DATA | | | / c +-+ b |/ TXD >-+ 4k7 +--*-*---|BC547B +-+ | | |\ | | | \| e | | -| +++ --+-- | 47 kΩ | | /|\ | | | --+-- | +++| 1N| | | 4148 | | | | | | | GND >--*-*---*-> GND You have to write software to drive it. Regards, Rob -- There are 707 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there Uwe Klein wrote: > a generic serialport sending with 50 or 25 Baud that switches a 77.5kHz > modulation signal to 75% ? It might be possible to drive the clocks with a TTL signal. > ( and a little bit of code that sends the DCF code bits every second ) With a PLL you can start the pulse on the second. Regards, Rob -- There are 703 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there Pete Ashdown wrote: > I've bought a number of clocks from Germany that use DCF-77 for time > set/synchronization. Needless to say, here in the states I can't use DCF-77. > I have see a number of GPS -> DCF-77 converters, but since I have a GPS synced > NTP server, I'd rather just pull time from the client Linux box next to my > desk. Does anyone have or know of a USB or serial device that can output a > DCF-77 signal to a number of clocks (4)? You could send a series of pulses from a serial port to the clocks. Regards, Rob -- There are 703 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] PC DCF-77 signal emulation?
Hi there David Lord wrote: > Conrad DCF77 module here works ok on MSF. I had to swap xtal and aerial. > If you can find the type of receiver chip and see the xtal and aerial > connections it's probably worth looking up if chip will pickup WWV. I > can't find my Conrad data but I think that also does WWV (output is > just demodulated carrier). http://en.wikipedia.org/wiki/DCF77 http://en.wikipedia.org/wiki/WWV_(radio_station) Regards, Rob -- There are 703 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Radioclock connection via serial to usb adaptor on Linux or BSD?
Hi there David Lord wrote: > I've recently been trying various radioclocks, gps, DCF77 and now MSF. > I have problems with reception being in a valley with horizon over > most of 360deg being 50m or so above the house. Cloudbase is often > below height of surrounding hills. > > On good days Garmin GPS-16-LVC has given around 2us (NMEA + PPS), > Conrad module DCF77 reception over about 1000km is lost most > evenings and early mornings but otherwise about 850us (Radioclkd2 > + SHM), and Conrad module with 60kHz and frame aerial for MSF > has done 65hr at reach of 377 for 175us (Radioclkd2 + SHM). Those > offsets are mid values rather than averages. > > I'd now like to try via a serial to usb for use with notebook. My > Vodafone mobile broadband connection regularly has differential > latency of maybe +/- 2 seconds and even chrony can't make sense > of this. > > From what I've read there are many serial to usb devices that > require their own drivers to provide full rs232 control lines. > PPS and radioclkd2 all require at least usable DCD and for > radioclkd2 I need one or both CTS and DSR as well. > > Can anyone here advise regarding serial to usb adapters on > Linux/BSD? AFAIK USB (ESP 232 over USB) causes timing problems. You are probably better off with syncing over ethernet. Regards, Rob -- There are 691 days until Central Registry IPv4 address exhaustion. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] DCF77 Problem
Hi there Rob wrote: > No you should not touch the pulse length, it conveys the time information. > 50 baud is the correct setting for the port. If missing a stopbit on each '1' is OK. > There have been various problems in the parse driver that cause things > like trash being written to the logfile, ntpd exiting on bad input, etc. > It will depend on the version of ntpd that you use, 4.2.4 > if you have any > problems like that. For some time, I had to run a "watchdog" process > that restarts ntpd when it has crashed. It doesn't crash. It keeps processing the other clocks. > But lately this has not been > a problem. > > It is normal to get things like this in the logfile: > 25 Feb 18:50:00 ntpd[4540]: parse: convert_rawdcf: parity check FAILED for > "-#-#---###M-S1-4p---81-p12-8--124--4---2--1---p_" > > 25 Feb 18:50:00 ntpd[4540]: PARSE receiver #0: FAILED TIMECODE: > "-#-#---###M-S1-4p---81-p12-8--124--4---2--1---p" (check receiver > configuration / wiring) That's not the problem. Once I have a line like above, it doesn't go back to normal DCF processing any more. Unless I restart NTPD. > When it happens too often or all the time, you will need to find a better > place for the receiver, or align its direction. If have a scope and a rs232 tester connected. The signal looks fine. The pulse lengths are OK. It can count from 0 to 58 with the pulses and get a 2 s interval. The signal is so clean that I get a 1 or 2 µs jitter on DCF. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] DCF77 Problem
Hi there Rob van der Putten wrote: > sput:~# stty -a speed 50 baud; rows 0; columns 0; line = 0; At 50 baud 9 bits (start + 8 data) is 180 ms. The max pulse length is just under 200 ms, so there is no stopbit. Is this OK? Should I build a circuit to reduce the max pulse length? Is it possible that interference confuses NTPD in a way that it can't recover from? It this a bug? Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] DCF77 Problem
Hi there Rob wrote: > Of course. You can do it at any time, I got the above output from my > own DCF77 receiver port. sput:~# stty -a ; quit = ; erase = ; kill = ; eof = ; eol = ; eol2 = ; swtch = ; start = ; stop = ; susp = ; rprnt = ; werase = ; lnext = ; flush = ; min = 1; time = 0; -parenb -parodd cs8 -hupcl -cstopb cread clocal -crtscts -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl -ixon -ixoff -iuclc -ixany -imaxbel -iutf8 -opost -olcuc -ocrnl -onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 -isig -icanon -iexten -echo -echoe -echok -echonl -noflsh -xcase -tostop -echoprt -echoctl -echoke Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] DCF77 Problem
Hi there Rob wrote: > So do a "stty -a > Then you should get this output: > > speed 50 baud; rows 0; columns 0; line = 0; > intr = ; quit = ; erase = ; kill = ; eof = > ; > eol = ; eol2 = ; swtch = ; start = ; stop = > ; > susp = ; rprnt = ; werase = ; lnext = ; > flush = ; min = 1; time = 0; > parenb -parodd cs8 -hupcl -cstopb cread clocal -crtscts > -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl -ixon > -ixoff > -iuclc -ixany -imaxbel -iutf8 > -opost -olcuc -ocrnl -onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 > ff0 > -isig -icanon -iexten -echo -echoe -echok -echonl -noflsh -xcase -tostop > -echoprt > -echoctl -echoke > > If not, something has messed up your port. Your task to find out what > it is. Assuming of course, that ntpd doesn't reset the tty to the original values when stopped. Or should I do this with ntpd running? Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] DCF77 Problem
Hi there Rob wrote: > Most likely some stupid program has changed the settings of the COM port. > (especially the baudrate) > > For example, in SuSE Linux when you are so unfortunate to click on the > "hardware information" icon in YaST, everything is messed up in the process > of detecting what is attached to your serial ports. There is no qui on this box. It's a server. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] DCF77 Problem
Hi there From the syslog; Feb 24 15:24:05 sput ntpd[28039]: parse: convert_rawdcf: INCOMPLETE DATA - time code only has 3 bits Feb 24 15:24:05 sput ntpd[28039]: PARSE receiver #0: interval for following error message class is at least 00:01:00 Feb 24 15:24:05 sput ntpd[28039]: PARSE receiver #0: FAILED TIMECODE: "--" (check receiver configuration / wiring) Feb 24 15:24:07 sput ntpd[28039]: parse: convert_rawdcf: INCOMPLETE DATA - time code only has 2 bits Feb 24 15:24:12 sput ntpd[28039]: parse: convert_rawdcf: INCOMPLETE DATA - time code only has 2 bits Feb 24 15:24:18 sput ntpd[28039]: parse: convert_rawdcf: INCOMPLETE DATA - time code only has 6 bits This goes on for hours. Once this happes I can only fix things by restarting the NTPD. Any ideas? Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Sudden drop in frequency after software update
Hi there Adam Myrow wrote: > It would be interesting to know what kernel version those who are having > trouble with unstable drift in Linux are using. I am using kernel > 2.6.27.7, 2.6.26 > and it is very stable, varying no more than 5 PPM, even across > reboots. It should be noted that I rebuilt my kernel with the timer > frequency set to 100HZ, It's 250. Timekeeping is not a primary concern on this box. > and made sure that dynamic ticks was disabled. > This seems to produce the lowest drift values. I can't find that option. > In an older 2.6 kernel, > I noticed the drift rate jumping all over the place as described. > However, it appears that this may have finally been fixed. It's pretty stable except for this perticular reboot on this perticular box. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Sudden drop in frequency after software update
Hi there Brian Inglis wrote: > Is spread specturm clock signal generation (EMI reduction) disabled for > the CPU and buses in the firmware? These are the BIOS options; CPU Spread Spectrum [Auto] Allows you to enable or disable the CPU spread spectrum. Configuration options: [Auto] [Disabled] PCIE Spread Spectrum [Auto] Allows you to enable or disable the PCIE spread spectrum. Configuration options: [Auto] [Disabled] They are both set to 'Auto' (default). Checking these involved a reboot. And now 'frequency' seems to be returning to it's original value. > That can cause these variations. > ISTR it being mentioned in the FAQ. URL? Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Sudden drop in frequency after software update
Hi there One of my Debian Lenny boxes more then halved it's 'frequency' after a software update (among others, kernel and ntpd). It used to be 43 ppm and is now below 17 ppm and still dropping. Is this normal? Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] My extra second ...
Hi there Steve Kostecke wrote: > All my Linux systems had a fine time. None of them locked up / crashed / > rebooted / etc. > > The kernels involved included: > > 2.6.24-etchnhalf.1-amd64 > 2.6.18-5-486 > 2.6.16-2-486 > 2.6.18-5-k7 > 2.6.18-4-powerpc > 2.4.16-k7 What about the hardware (Intel /AMD)? Regards, Rob -- Don't abolish the leap second ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] My extra second ...
Hi there Unruh wrote: > Apparently a number of Linux machines completely locked up at the leap > second. Problems in the kernel ntp.c code apparently. One of mine did; I have two Debian Lenny boxes. Kernel 2.6.26-1-486 on a AMD Athlon, and 2.6.26-1-686 on Core 2 Quad. Both are based on Linux 2.6.26-12 (identical source). 2.6.26-1-486 locked up, 2.6.26-1-686 did not. Neither did my Sarge 2.4.27-2-686 Pentium II or Sid 2.6.20-1-686 Pentium III. Apart from that everything went smoothly. Apparently my ISPs clocks messed up again. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Question about 2008 leap second
Hi there Unruh wrote: > It depends. Some people regard the leapsecond as counting > 23:59:0 23:59:1 ...25:59:59 23:59:60 0:0:0 > while I think the leapsecond standard is actually > 23:59:0 23:59:1 ...25:59:59 0:0:00:0:0 > Ie it occurs on teh first second of the new year, rather than the last > second of the old. I assume BBC (and ntp ) follow the former. AFAIK, the first is the standard, the second the implementation. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Question about 2008 leap second
Hi there Richard B. Gilbert wrote: > I predict that the leap second will be ignored by at least 98% of the > world! Most people cannot set their clocks to the correct second and > most would not bother if they could. > > The leap second will be important to astronomers, navigators, the NIST > and a few other people and institutions. I read that, during the previous leap second, in the countdown to the new year on BBC TV, the last 'second' lasted two instead of one seconds. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Garmin GPS 18LVC Setup but questions on best way
Hi there David Woolley wrote: > V.24 doesn't specify electrical characteristics. I suspect you mean > V.28. The standard was 'split' at some point (I haven't looked into this stuff for years). > RS232 C is also 25 volts, open circuit, although drivers for both > standards are not required to be able to achieve this. > Totem poll TTL high is 2.4V at 400 micro Amps. LS TTL is 2.7V. > (National Semiconductors Data Book, 1976) > > CMOS is 4.95V minimum into infinity, at 5V Vdd, or 4.6 volts at 300 > micro amps. The CMOS case corresponds to a load resistance that is a > little higher than that of an RS232 C receiver, so one would expect > somewhat lower an output voltage into the real load. Figures for 4000 > series CMOS, from 1975 RCA COS/MOS Integrated Circuits Data Book. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Garmin GPS 18LVC Setup but questions on best way
Hi there Unruh wrote: > Yes, well... The RS232 standard says that teh signal levels are -12V and > +12V and that the absolute minimum be -5V and +5V. However, many serial > chip makers have bent those standards and the serial port may or may not > respond to the 0,5 level that the PPS output actually delivers. There is > absolutely no reason why it should. That signal is completely out of spec. > If yours does work, it is because your serial port manufacturer severely > bent the rules. But many (most?) do. But it could well be that the serial > port is flakey on especially the 0V instead of -5 to -12 V end. In V24 it's -25 ... 25 V [1]. -3 ... 3 V isn't defined. Most level translators however are completely happy with 'TTL' [2] signals. This is not bending the specs, it's within the specs. Not being happy with 'TTL' is also within the specs. If you want to make sure have a look at the data sheets of your level translators. Most consider < 0.8 V low and > 2 V high. > The parallel port specs state that the signal levels are 0V and 5 V which > is exactly what the garmin delivers. An alternative is converting the Garmin signals to RS232. [1] Most level translators can cope with -15 ... 15 V [2] TTL high is actually Ca 3.5 V. CMOS high is 5 V (when supplied with 5 V power). Regards, Rob -- Anglo-Saxon management is a memetic virus ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Garmin GPS 18LVC Setup but questions on best way
Hi there George R. Kasica wrote: > Did you need to use two physical serial plugs or a splitter or just do > this with symlinks in the OS? Two plugs. Regards, Rob -- Anglo-Saxon management is a memetic virus ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Garmin GPS 18LVC Setup but questions on best way
Hi there George R. Kasica wrote: > If I switch to the following settings I can seem to get NEMA data but > then I lose the PPS function which hurts the accuracy far more. Do you > know if shm can somehow allow both with some type of setting - ideally > that is what I'm trying to accomplish through the shm driver or > something else without hacking the kernel, etc?? I used to use a two serial port setup; ttyS0 for PPS using SHM and ttyS1 for NMEA. This requires a gps0 -> ttyS1 symlink; server 127.127.20.0 fudge 127.127.20.0 time1 0.172 server 127.127.28.0 minpoll 4 prefer fudge 127.127.28.0 refid PPS ttyS1 gets all the signals, ttyS0 just the PPS. This worked better then GPSD, so I think I will return to this setup. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Garmin GPS 18x LVC huge offset between PPS and NMEA -- up to 700 ms
Hi there Some of my experiences with this Garmin and GPSD; The Garmin isn't necessaryly set to defaults. Reset with: $PGRMO,,4 Then set the output pulse length to 200 ms; $PGRMC,9, These should be terminated with a CrLf pair. If connected directly to Minicom, the Garmin appears to be online once per second. This is because the PPS is tied to DCD. The circuit below can be used to avoid line status confusion; Garmin Minicom MaleFemale 9P 25P RXD-RXD 23 TXD-TXD 32 GND-GND 57 +-DCD18 | *-DTR4 20 | +-DSR66 +-CTS75 | +-RTS84 SH--SH Start Minicom with '-o' to avoid sending an init string. Exit Minicom with 'Ctrl-A,Z,Q' instead of 'X' to avoid a reset. GPSD wants to change theses settings. Disable this in /etc/default/gpsd with; DAEMON_OPTS="-b -n" The NMEA data is off by some 170 ... 180 ms. This time keeps changing all the time; server 127.127.28.0 minpoll 4 fudge 127.127.28.0 time1 0.178 refid GPSa server 127.127.28.1 minpoll 4 prefer fudge 127.127.28.1 refid PPSa NTPD want to make shore that GPS time source isn't talking nonsense. This means you have to use an other time source as well. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] basic questions about the leapsecond
Hi there David L. Mills wrote: > I am told the file is on all NTP servers operated by NIST. See the list > of public servers at NIST or www.ntp.org. ftp://ntp-a.boulder.nist.gov/pub/leap-seconds.3427142400 works Thanks! Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] basic questions about the leapsecond
Hi there Hal Murray wrote: > That site is unlikely to be down for long. It's still down. I can ping time.nist.gov, but it won't FTP. > Are you behind a NAT box? I need to use the passive mode for ftp. No. I also tried the shell box at my ISP. Same result. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] basic questions about the leapsecond
Hi there Antonio M. Moreiras wrote: > 1 - download ftp://time.nist.gov/pub/leap-seconds.3427142400 Is there an other source? This site appears to be down. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there Steve Kostecke wrote: > Pipermail is just Mailman's archive mechanism. It has _nothing_ to do > with the gateway. Anyone using the gateway? > | Content-Type: text/plain; charset=UTF-8; format=flowed > | Content-Transfer-Encoding: 8bit I don't think news servers do charset conversions. > I am using slrn packaged for Debian Stable (version 0.9.8.1pl1-28) in > both a 'screen in ssh in an xterm' session and at an 80 column x 60 line > console (using using a font chosen for readability) and on the standard > 80x25 Linux console. My 80x25 Linux console is UTF-8. > This is a hex 3F and renders as a question-mark everywhere. Euro should be hex E2 82 AC > This is a hex B5 Greek mu should be hex C2 B5 > and renders properly in the xterm and the 25 line > console. It looks like a squashed upper-case A with an acute accent on > the 60 line console. > This is a hex 3F and renders as a question-mark everywhere. z-caron should be hex C5 BE I assume you are using ISO-8859-1 as your default locale. ISO-8859-1 does not contain a Euro or z-caron. Regards, Rob -- Anglo-Saxon management is a memetic virus ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there David J Taylor wrote: > Thanks, Uwe. If there's a setting I have wrong, I'll gladly alter it. Google; Outlook Express, missing "charset="; http://newsgroups.derkeiler.com/Archive/Comp/comp.sys.hp.hardware/2005-09/msg00167.html Your posting uses local language characters like u+umlaut, but it does not contain the required MIME headers which specify the correct character set (probably ISO-8859/1). You can fix this as follows: Tools -> Options... -> Send -> News Sending Format -> Plain Text Settings... You probably have "Message format" set to "Uuencode" (which is some stupid OE default). If so, set it to "MIME", set "Encode text using:" to "None" and do *not* set (i.e. no tic-mark) "Allow 8-bit characters in headers". You may want to set the "Mail Sending Format" the same. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there David J Taylor wrote: > Those characters display correctly on my NNTP feed from my ISP, using > Microsoft Outlook Express for news-reading. You convert the text without stating the charset used, which makes non ascii unreadable. This is probably an Outlook bug. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there Ryan Malayter wrote: > Interesting. The non-ASCII symbols in your message appeared just fine > to me, But your post doesn't; The Content-type is text/plain; charset="windows-1252". The content encoding is 7bit. Something is seriously broken. > but I assume that is because we are both using the email list. I'm not. I posted via news.xs4all.nl > I received your message encoded as: > Content-Type: text/plain; charset="utf-8" > Content-Transfer-Encoding: base64 The original Content-Transfer-Encoding is 8 bit. > I am in total agreement, even though I am an "ugly American." It's a standard. I would have to look up which one. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there Ryan Malayter wrote: > On Sun, Oct 26, 2008 at 2:48 PM, Unruh <[EMAIL PROTECTED]> wrote: > >> That may be what you expect, but you can get it 1usec (1 micro second). > > Is there something wrong with the mail gateway and Unicode? I posted > my message as text/plain with charset=UTF-8, which has been an IETF > standard for more than a decade. And my message does, in fact, appear > correct with UTF-8 characters such as "l" I see a 'l' (006C) here. > (Greek Small Letter Mu, > Unicode 03BC) in the list archives at: > https://lists.ntp.org/pipermail/questions/2008-October/020235.html Charsets contradict; 'META http-equiv="Content-Type" content="text/html; charset=us-ascii"' The transport charset is ISO-8859-1. Transport overrules content. However, non ascii is encoded as 'ddd;', where 'ddd' is de decimal Unicode value, so non ascii is legible anyway. > However, all replies to my message were in 7-bit charset="us-ascii", > which of course mangles the non-ASCII chasracters. > > So is it the pipermail gateway that is not Unicode compliant, Does it just convert mail to news? On news to mail as well? So what does this look like? Euro: € mu: µ z-caron: ž > or is it > the MUAs of the respondents that is at fault? If it is the list > itself... well, isn't it absurd to restrict content of a mailing list > to 7-bit us-ascii? It is 2008, not 1988. Post 1999 software should support UTF-8. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] list posts in UTF-8
Hi there David Woolley wrote: > Strictly speaking, there is no standard that permits non-ASCII material > on USENET, although the de facto position is that MIME is permitted. > There are still some important USENET user agents that are not MIME > aware and USENET can get transported over non-TCP channels. From RFC 3977; Although the protocol specification in this document is largely compatible with the version specified in RFC 977 [RFC977], a number of changes are summarised in Appendix D. In particular: o the default character set is changed from US-ASCII [ANSI1986] to UTF-8 [RFC3629] (note that US-ASCII is a subset of UTF-8); That's transport, not content. For content the RFC refers to MIME. Regards, Rob ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Problems distibuting time from GPSD programto NTPD
Hi there Johan Swenker wrote: > Be sure to upgrade to gpsd 2.36, or backport the leapyear patch which > was distributed on januari 1st. You also need a patch; http://www.nabble.com/Bug---Switching-Drivers-from-Generic-NMEA-to-Garmin-serial-creates-2-gpsd_ppsmonitor-threads-td14770083.html Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] First attempt GPSD/PPS ->NTP time server
Hi there Unruh wrote: > This is confusing. You first say that one NMEA sentence pers second is too > much data, and then that youarranged that it sent 6 sentences per second. > Note that only one sentence ( which should take about 140ms at 4800Bd) is > allo you need. GPRMC, GPGGA, GPGSA and three GPGSV lines (default) are max 367 chars. All sentences enabled are max 575 chars, which takes longer then one second to send. If you just want time, GPRMC is all you need. If you want to see where the sats are, you need more. See http://www.garmin.com/manuals/425_TechnicalSpecification.pdf Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] First attempt GPSD/PPS ->NTP time server
Hi there Steve Kostecke wrote: > There is no benefit to sending all of those NMEA sentences. > > Select one and turn the rest off. For just time GPRMC will do. Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] First attempt GPSD/PPS ->NTP time server
Hi there Rob van der Putten wrote: > My Garmin was sending to much data, sending a NMEA sentence once per > second. Sorry, once per two seconds. Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] First attempt GPSD/PPS ->NTP time server
Hi there root wrote: > No GPS NMEA should not do that. The length of the sentence is almost fixed > length, so the timing on it should vary by perhaps a few msec, as you found, > certainly not by seconds. It sounds like you have troubles. > > You could try using minicom ( assuming you are on Linux) to look at the output > on the serial port. PPS is tied to DCD. So online is off 90 % of the time. With the following circuit you can configure the GPS receiver; Garmin Minicom MaleFemale 9P 25P RXD-RXD 23 TXD-TXD 32 GND-GND 57 +-DCD18 | *-DTR4 20 | +-DSR66 +-CTS75 | +-RTS84 SH--SH > The GPS sentence should certainly be able to pull you into the second range > very easily. > Anyway use some serial port terminal program to read what the gps is sending > and how often. > (Remember at 4800Bd, it takes about 2 msec to send each character. My Garmin was sending to much data, sending a NMEA sentence once per second. So I put '$PGRMO,,4' in a file and send it to the Garmin. Now it's at six lines per second; GPRMC, GPGGA, GPGSA and three GPGSV lines (plus one PGRMT per minute). Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Harlan Stenn wrote: > I would hope you sent email to their webmaster... No. Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Hal Murray wrote: > I got mine with a Mastercard. Is that restriction for out of US > delivery? Apparently. Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Dennis Hilberg, Jr. wrote: > It works fine for me using Firefox 2.0.0.9. Have you tried using a > different browser? Mozilla and Konqueror. Complains about not being able to calculate the shipping charges. This seems to work though; http://www.flash-memory-store.com/garmin--gps-18-lvc-oem-system-waas12.html I hope it actually does work. Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Dennis Hilberg, Jr. wrote: > Garmin sells it from their website for 74.50 USD: > https://buy.garmin.com/shop/shop.do?pID=223&tab=gps18oem They don't ship outside the USA. > I bought mine from MegaGPS.com for 64.99 USD + 5.95 USD shipping: > http://www.megagps.com/index.asp?PageAction=VIEWPROD&ProdID=121 > > MegaGPS.com ships internationally. I don't know about Garmin. Their webform is broken. Any suggestions? Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Hal Murray wrote: > I got my GPS 18 LVC from Provantage (www.provantage.com). > Their web page says they ship internationally, but I don't > know how their total cost compares to any other place. They do American express only (no other credit cards) and bank transfers. Transfers in de Euro zone [1] are free of charge. Outside the EU they are very expensive. [1] Also called Euro land. These are countries which use the euro (not the entire EU). It doesn't include the UK. Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there Hal Murray wrote: > Be sure you get the bare/OEM version, not one with mapping software. With software it's € 280.- BTW, the USD is € 0.68 right now. So EU companies are charging 150 % extra. Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Hi there John Ioannidis wrote: > Out of curiosity: what is wrong with the Garmin GPS 18LVC that someone > would like to look at an alternative? At < $70, it's practically free. Where can get one in Europe? USD 68.50 is ca € 50.- Some EU webshops sell it for € 114.-, which is ca USD 156.- Or can I order one from the Garmin site? Regards, Rob -- Avoid alphabet soup. Include the charset in your HTML header; META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=Your_Charset" ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Home Time Server
Hi there David L. Mills wrote: > Not to get into a pissing contest, but both WWVB and the European LW > services operate at similar frequencies and ERPs European services > operate at higher latitudes where noise levels are lower, while WWVB > operates at lower latitudes and generally longer ranges. I've lived in > Europe and can attest the differences. > > WWVB has claimed carrier accuracy of 1e-11 and has carrier phase > ambiguity of 16 microseconds. Electronics-wise, a narrowband crystal > filter ahead of a phase-lock loop with a time constant of a couple > minutes pins the zero crossings well below one microsecond. While the > modulation bandwidth is quite narrow, finding the right on-time cycle > crossing with 5 ms risetime (30 cycles) and a comb filter isn't hard. > Thats how the WWV audio demodulator works. > > The Spectracom WWVB radios we have used to do an excellent job; however, > the claimed accuracy was only 0.1 ms, mostly because the computer used > for demodulation was so chintzy. Sadly, WWVB is dead here because of EMI > generated by hundres of UPS systems all over campus. I'd be curious to > know if this has also affected DCF-77 users. AFAIK DCF-77 runs at 77.5 kHz and a lot of receivers mix this down to 625 Hz; This means a mirror at 76.875 kHz. Interference sources will therefore be at 38.75 and 38.44 kHz (2nd harmonic). Switching stuff often runs at 25, 30 or 33 kHz. So a frequency of 60 kHz (WWVB) will have more interference. Regards, Rob -- Lifestyle sux ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP MRTG graphs?
Hi there Dennis Hilberg Jr wrote: > I have the same issue on my apollo machine. I believe the problem is > the positive frequency value, as it seems to have no trouble graphing > saturn's negative values properly. > > At any rate, adding the option '--alt-y-grid' should solve the problem. > I put the script I use for apollo's plots in my scripts/ntp directory > for an example: > > http://saturn.dennishilberg.com/scripts/ntp/graph_ntp_values.apollo > > Hope that helps, It does! Thanks! I have got some parent graphs, if anyone is interested; http://www.sput.nl/ntpstats/remote/ The scripts generate meta files, so each image has its own expiry date and refresh time. Regards, Rob -- +------------------+ | Rob van der Putten, [EMAIL PROTECTED]| | http://www.sput.nl/spam/spam-policy.html | +--+ ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP MRTG graphs?
Hi there Dennis Hilberg Jr wrote: > Like Roger said, you should try RRDtool. It will do what you are asking > for. > > I have my scripts available that I use to create my RRDtool plots. > Check out http://saturn.dennishilberg.com/ntp.html for the data plots, > and http://saturn.dennishilberg.com/scripts/ntp for the actual scripts. > > Also, be sure to check out Steven Bjork's website > http://www.wraith.sf.ca.us/ntp/index.html#monitoring for help with > creating RRDtool plots with ntp. It helped me out a lot with figuring > out RRDtool, as documentation alone is not much help. > > The documentation is somewhat helpful though, > http://oss.oetiker.ch/rrdtool/doc/index.en.html . But it's much more > helpful to see some scripts in action rather than rely completely on the > documentation, IMHO. I have got a weird problem with the frequency. On my Sid box it won't do the PPM scale properly; http://www.sput.nl/ntpstats/pc5/pc5-ntp-freq.png It just says 74, 74, 74 etc.. Instead of 73.9, 74.0, 74.1, etc. It's OK on my Sarge box though; http://www.sput.nl/ntpstats/rrdtool/sput-ntp-freq.png Regards, Rob -- +--------------+ | Rob van der Putten, [EMAIL PROTECTED]| | http://www.sput.nl/spam/spam-policy.html | +--+ ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions