NTP GPS Problem



  • Hi,

    i've a Problem with the NTP Daemon. On the COM-IF I use a Trimble Acutime 2000 with GPS Time. Over the WAN-Side i got the Timeservers of the PTB in Germany (ntp1.ptb.de, ntp2.ptb.de, ntp3.ptb.de) as fallbacl. After a long time of tuning i found a config (Fudge time 2 - 0.400) i got the the GPS working (Screenshot attached). But after some time all Pears go to "False Ticker" and the NTP-Daemon is restarting. Does anybody had the same Problem? Or any ideas to got this working correctly?
    On the 2.1.5 there was no way to make the GPS working.

    Thanks,

    MasterD



  • Here the Screenshot




  • I might be talking out of my ass here, but from what I know, you need PPS to get accurate and reliable time from GPS. This is what I've noted with my RaspberryPi + GPS (with PPS).

    Snippet from my RaspberryPi with Ublox MAX-7Q GPS module attached.

    
         remote           refid      st t when poll reach   delay   offset  jitter
    ==============================================================================
    oPPS(0)          .PPS.            0 l    3   16  377    0.000   -0.003   0.002
    *SHM(0)          .GPS.            0 l    2   16  377    0.000   -9.308   2.441
    
    


  • @fragged:

    I might be talking out of my ass here, but from what I know, you need PPS to get accurate and reliable time from GPS

    ^^ This.  Without PPS, your gps is worse than most ntp time servers on the net, unless you have a very bad & variable WAN connection.

    @masterd01:

    Over the WAN-Side i got the Timeservers of the PTB in Germany (ntp1.ptb.de, ntp2.ptb.de, ntp3.ptb.de) as fallbacl.

    Your screenshot shows them with 'reach' of '1', which means either you are not getting valid responses from them, or you are looking right after you (re)started ntpd.

    But after some time all Pears go to "False Ticker" and the NTP-Daemon is restarting. Does anybody had the same Problem? Or any ideas to got this working correctly?

    Bad clocks, poor peer response, etc. should not cause ntpd to restart.  More likely the restart is due to a change you made in the gui?

    Your best bet is to get PPS working properly; it does work with pfSense:

    ntpq> pee
         remote           refid      st t when poll reach   delay   offset  jitter
    ==============================================================================
    oGPS_NMEA(0)     .pps.            0 l    -   16  377    0.000   -0.003   0.001
    +wtf.roflcopter. 195.83.222.27    2 u   50   64  377  111.666    2.243   2.230
    +108.61.194.85 ( 132.163.4.101    2 u   32   64  377  107.992  -13.892   2.647
    *repos.mia.lax-n 200.98.196.212   2 u   46   64  377   55.956   -1.887   1.289
    ntpq>
    
    

Log in to reply