[ntp:questions] NTP - orphan mode with SAME stratum ("tos orphan 6") FAIL to sync.

Unruh unruh-spam at physics.ubc.ca
Mon Apr 13 21:19:21 UTC 2009


"Richard B. Gilbert" <rgilbert88 at comcast.net> writes:

>David Woolley wrote:
>> sheikdawoodrajali at gmail.com wrote:
>> 
>>>
>>> Package Version:
>>> NTP v4.2.4.p5
>> 
>> I'm not sure that is new enough to stay locked.  Unless you use a very 
>> recent version, which may have to be a development version, orphan mode 
>> will fall apart a day or two after the system is started.
>> 
>> 
>>>
>>> 4) tweak the time on ALL the nodes.
>>> i.e
>>> 1st machine with <systemtime>
>>> 2nd machine with <systemtime + 10 minutes>
>>> 3rd machine with <systemtime + 20 minutes>
>>> 4th machine with <systemtime + 30 minutes>
>>> 5th machine with <systemtime + 15 minutes>
>>> 6th machine with <systemtime + 5 minutes>
>> 
>>  From what Dave Mills has said, this will not synchronise because the 
>> error bands or all sources are disjoint.  You will definitely need the 
>> latest development version, which may not have been issued.
>> 
>> (The ability to cope with this may be subject to further refinement, as 
>> I think simply enabling it unconditionally could result in a the 
>> addition of a new machine pulling all the others to its time.

>Asking NTPD to cope with this does not seem, to me, to be the correct 
>solution.  It is not asking too much for the owner/administrator/user of 
>the machines in question to set the clock with nptdate or ntpd -g or 
>even with "date" (or equivalent on your O/S) to a reasonable 
>approximation of the correct time.

>Further, I think the time and effort to code such a solution could be 
>better spent!

While he has absurd differences, the default error bound is something like
5ms. So change all those "minutes" in his list with "milliseconds" and you
will get the same behaviour. 





More information about the questions mailing list