[ntp:questions] NTP on LAN

David L. Mills mills at udel.edu
Mon Sep 29 03:04:05 UTC 2008


Steve,

May I remind folks that the single definitive documentation source is 
indeed me and the reason for that is very simple. I've taken a great 
deal of trouble, years in fact, to make sure the documentation actually 
does describe how the animal actually works. The fact that the 
definitive documentation is essentially under lock and key is absolutely 
intentional, mainly to avoid inaccuracies that creep in when other OS 
suppilers twitch it for their local customers and get it wrong.

Informal support documentation is a good thing, especially when it saves 
folks from reading detailed technical material. However, when it comes 
down to real chapter and verse, the definitive documentation comes from 
me. As it says in the documentation, I am happy to receive bug reports, 
typos and even complaints about the lock and key policy. I am also jappy 
to receive and integrate new pages in the document set; I have done that 
several times, but those pages get a careful read and sanity check first.

Dave

Steve Kostecke wrote:

> On 2008-09-28, David Woolley <david at ex.djwhome.demon.co.uk> wrote:
> 
> 
>>Steve Kostecke wrote:
>>
>>
>>
>>>Nonsense.
>>>
>>>A single ntpd can be configured in Orphan Mode to serve as the
>>>"server" for a time island.
>>
>>I hadn't considered the degenerate case. However, if one looks at the
>>key selling points for orphan mode, in the documentaion, one finds
>>that they all depend on having multiple orphans at the same stratum:
>>
>>
>>>| There are many disadvantages using the local clock driver:
>>>| multiple source redundancy is not possible and the subnet is
>>>| vulnerable to single-point failures. Orphan mode is intended to
>>>| replace the need for
>>
>>If it really has benefits in contexts where there is only one orphan
>>mode server, the documentation needs amending to mention them.
> 
> 
> The Distribution Documentation is maintained by one person; you'll have
> to contact him about this.
> 
> Anyone who is interested is welcome to contribute to the Community
> Supported Documentation at http://support.ntp.org/support
> 
> 
>>Note two orphans at the same stratum aren't possible here, as the 
>>requirement included a strict precedence and two at different strata is 
>>another degenerate case.
> 
> 
> It is possible to set the orphan strata to enforce precedence.
> 
> There is no requirement for members of an orphan mode group to operate
> at the same stratum. 
> 




More information about the questions mailing list