Re: [chrony-dev] Bug -- interaction between ntpdate and chronyd at bootup -- will never sync up

[Miroslav Lichvar <mlichvar@xxxxxxxxxx>]

On Thu, Feb 21, 2013 at 10:17:39AM -0600, ray vantassle wrote:
> > Can you please post your config and if possible also the tracking,
> > measurement and statistics logs (configured by the log and logdir
> > directives in chrony.conf) when that happens?
> 
> Here is another run.  This time, it never synced up.  In fact, it
> seems to have stopped making queries altogether, because the next
> morning the REACH fields were all 0.

Thanks for the data.

It seems the problem is that chronyd can't recover because all
incoming NTP packets are ignored due to failed test4. It tests if
the absolute value of peer delay is shorter than 16 seconds.

The peer delay is calculated like this:
delta = local_interval - remote_interval / (1.0 - source_freq_lo);

The source_freq_lo value is from the already collected samples and it
seems to be very close to 1.0, which amplifies the remote_interval to
thousands of seconds.

I think the fix is to discard the old data when the frequency
estimate reaches a specified limit, similarly to the case when a
backward jump is detected.

-- 
Miroslav Lichvar

-- 
To unsubscribe email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "unsubscribe" in the subject.
For help email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.