A leap second may only be introduced at the end of a UTC month; therefor
the leap indicator may be set at any time within the UTC month with an
impending leap second.
2:57am [EMAIL PROTECTED] [~] ntpdc -c sysinfo | grep leap
leap indicator: 01
On Sun, 18 Dec 2005, DamZ wrote:
Simon Arlott a écrit :
RFC 1305 doesn't say that it must not be set more than 24 hours early, it
just states that it must
be set more at least 1 minute before it happens:
Leap Indicator (sys.leap, peer.leap, pkt.leap): This is a two-bit code
warning of an impending leap second to be inserted in the NTP timescale.
The bits are set before 23:59 on the day of insertion and reset after
00:00 on the following day. This causes the number of seconds (rollover
interval) in the day of insertion to be increased or decreased by one.
(Later in the RFC it does say it would be set on the day before.)
Perhaps I'm missing something here, but this means that the Leap Indicator is
set on and only on the day of insertion. The Leap Indicator is about telling
other servers that there will be a leap second on a given day (so that they
don't have to slowly drift to get back in sync), it only has a meaning the
day of the insertion.
So, it's very clear to me that *no* leap indicator should be set by now. Are
27,7% of the servers wrong? (or simply assuming that leap seconds only occurs
at the end of December or June?)
-- damz
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers