Daniel Kabs wrote:
> Harlan,
>
> thanks for pointing out that sntp does not work with an undisciplined
> clock.
>
> In this case, one has to resort to ntpdate or patch sntp::main.c:
>
> --- /tmp/ntp-4.2.0b-rc1/sntp/main.c 2005-12-21 00:33:36.000000000 +0100
> +++ main.c 2006-03-20 13:31:12.378670736 +0100
> @@ -457,7 +457,7 @@
> failed = (data->mode != NTP_SERVER && data->mode != NTP_PASSIVE);
> response = 1;
> }
> - if (failed || data->status == 3 || data->version < 1 ||
> + if (failed || data->version < 1 ||
> data->version > NTP_VERSION_MAX ||
> data->stratum > NTP_STRATUM_MAX) {
> if (verbose)
>
> I will update
> http://ntp.isc.org/bin/view/Dev/DeprecatingNtpdate
> accordingly
>
We need to look at the protocol requirements for sntp to see if this
patch conforms, if you want to create a bug report for it. I'm not sure
that it's the right thing to do to synchronize with an undisciplined
clock. Should we move this question to the Working group?
Danny
> Cheers
> Daniel
_______________________________________________
questions mailing list
[email protected]
https://lists.ntp.isc.org/mailman/listinfo/questions