[ntp:hackers] Protocol nits

Serge Bets serge.bets at laposte.net
Mon Jun 25 09:36:31 UTC 2007


Hello David,

 On Friday, June 22, 2007 at 18:45:42 +0000, David L. Mills wrote:

> The code counts the number of leap bits set in the survivor list. If
> the number is greater than MINSANE, which is the tinkerable minimum
> number of survivors (default 1), a system leap event is declared.

What's the status of the NIST leap-seconds file in this scheme? The
leapfile can force a leap event, regardless of servers and refclocks.
But does the leapfile force the *absence* of a leap event, against any
other source(s) mistakenly sending leap bits set?


Serge.
-- 
Serge point Bets arobase laposte point net



More information about the hackers mailing list