Tim Shoppa wrote:
> Eugene writes:
>> You need more commonality between the two servers
>> configs.
> 
> Isn't this also called "putting all your eggs in one basket"?

For testing purposes you need a baseline. The best baseline FOR TESTING is a 
nearby ST1 server common in both configs.

> I know that a lot of folks here really really desire a homogenous
> set of optimally-performing stuff, but in reality aren't we
> better served by a wide variety and distribution of not
> just NTP time servers, but also basic refclocks (e.g. not just GPS,
> but also shortwave, longwave, international clocks?)

I NEVER said homogeneous or that both should be exactly the same. To figure out 
the WHY of what's happening both configs need COMMONALITIES to test against.
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers

Reply via email to