Hardware clock fails to update. - NTP

This is a discussion on Hardware clock fails to update. - NTP ; Our embedded system syncs with the server via NTP. The kernel (2.6.9) has RTC set to UTC. When the system boots up, the hardware clock is off by 7200 seconds. When NTP syncs, the system time is set correctly and ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: Hardware clock fails to update.

  1. Hardware clock fails to update.

    Our embedded system syncs with the server via NTP. The kernel (2.6.9) has RTC
    set to UTC. When the system boots up, the hardware clock is off by 7200
    seconds. When NTP syncs, the system time is set correctly and the system runs
    synced for a decent while (2 hours). The system is then powered down and
    restarted. Upon restarting, the clock returns to its 7200 second offset.
    Various internet searches pointed to the fact that hardware clock syncing
    being placed under the responsibility of ntp. Is this really the case? If so,
    how can I check that NTP tries to update the hardware clock?

    Best regards,

    Frans

    _______________________________________________
    questions mailing list
    questions@lists.ntp.isc.org
    https://lists.ntp.isc.org/mailman/listinfo/questions


  2. Re: Hardware clock fails to update.

    Frans Grotepass wrote:

    > Our embedded system syncs with the server via NTP. The kernel (2.6.9) has RTC
    > set to UTC. When the system boots up, the hardware clock is off by 7200
    > seconds. When NTP syncs, the system time is set correctly and the system runs
    > synced for a decent while (2 hours). The system is then powered down and
    > restarted. Upon restarting, the clock returns to its 7200 second offset.
    > Various internet searches pointed to the fact that hardware clock syncing
    > being placed under the responsibility of ntp. Is this really the case? If so,
    > how can I check that NTP tries to update the hardware clock?
    >
    > Best regards,
    >
    > Frans


    I don't recall any NTP documentation that says the ntpd sets the
    hardware clock. AFAIK, this is something the operating system does at
    such times as the designers thought appropriate. I think that most do
    so when shutting down! If your O/S does not set the hardware clock,
    perhaps you should have your application do so.

  3. Re: Hardware clock fails to update.

    fmgrotepass@yahoo.co.uk (Frans Grotepass) writes:

    > Our embedded system syncs with the server via NTP. The kernel (2.6.9) has RTC
    > set to UTC. When the system boots up, the hardware clock is off by 7200
    > seconds. When NTP syncs, the system time is set correctly and the system runs
    > synced for a decent while (2 hours). The system is then powered down and
    > restarted. Upon restarting, the clock returns to its 7200 second offset.
    > Various internet searches pointed to the fact that hardware clock syncing
    > being placed under the responsibility of ntp. Is this really the case? If so,
    > how can I check that NTP tries to update the hardware clock?


    Please try to read "8.3.4.4.1. How can I read or write the CMOS clock?" in the
    NTP FAQ. The linux kernel does not write the full time back to the RTC AFAIK,
    and other programs during startup/shutdown may be involved when messing with
    the RTC chip.

    Regards,
    Ulrich

+ Reply to Thread