Re: [Ntp] Negative leap-second?

Sure.  And if a feature is rarely/never used, problems in the wild are 
rarely/never seen.

Hence my note that regular (at least once every 6 months) changes should 
be made to exercise the features.

This is common practice in other areas...

H

On 8/3/2022 7:45 AM, Doug Arnold wrote:
> When we tested both positive and negative leap seconds at PTP plugfest 
> events, we found implementations that handled the positive leap seconds 
> correctly, but not the negative ones.  The faulty list included 
> implementations that looked only at the leap second flag, and never 
> check whether it was forward or backward.  Forward was assumed.
> 
> Doug
> 
> *From: *ntp <ntp-bounces@ietf.org> on behalf of Warner Losh <imp@bsdimp.com>
> *Date: *Tuesday, August 2, 2022 at 10:00 PM
> *To: *Harlan Stenn <stenn@nwtime.org>
> *Cc: *Danny Mayer <mayer@pdmconsulting.net>, NTP WG <ntp@ietf.org>
> *Subject: *Re: [Ntp] Negative leap-second?
> 
> On Tue, Aug 2, 2022, 7:55 PM Harlan Stenn <stenn@nwtime.org 
> <mailto:stenn@nwtime.org>> wrote:
> 
>     I don't think this really matters.
> 
>     The fastest way to make sure that systems will handle leap seconds
>     is to
>     decide to implement a forward leap second in, say, January, and then a
>     backward leap second in, say, March.  Then add one in May and subtract
>     one in July.  Repeat.  When an actual leap second is really needed,
>     just
>     skip one of these, as appropriate.
> 
>     Doing this will give folks plenty of opportunity to make sure that leap
>     second processing code works, on an ongoing basis.
> 
>     If folks prefer, do these once every third month instead of every other
>     month.  The trick is to do them often enough that vendors have a clear
>     reason to get this working correctly.
> 
> Yea, do them all the time, or never. Either are good, though I'd prefer 
> the latter. Also a way to make sure it works...
> 
> Warner
> 
>     On 8/2/2022 4:05 PM, Danny Mayer wrote:
>     > Apparently the earth's rotation is speeding up: 
>     > https://www.timeanddate.com/news/astronomy/shortest-day-2022
>     <https://www.timeanddate.com/news/astronomy/shortest-day-2022>
>     > 
>     > If we have to subtract a leap-second I think we will see failures, since 
>     > this is rarely if ever tested.
>     > 
>     > Danny
>     > 
>     > _______________________________________________
>     > ntp mailing list
>     > ntp@ietf.org <mailto:ntp@ietf.org>
>     > https://www.ietf.org/mailman/listinfo/ntp
>     <https://www.ietf.org/mailman/listinfo/ntp>
>     > 
> 
>     -- 
>     Harlan Stenn <stenn@nwtime.org <mailto:stenn@nwtime.org>>
>     http://networktimefoundation.org <http://networktimefoundation.org>
>     - be a member!
> 
>     _______________________________________________
>     ntp mailing list
>     ntp@ietf.org <mailto:ntp@ietf.org>
>     https://www.ietf.org/mailman/listinfo/ntp
>     <https://www.ietf.org/mailman/listinfo/ntp>
> 
> 
> _______________________________________________
> ntp mailing list
> ntp@ietf.org
> https://www.ietf.org/mailman/listinfo/ntp

-- 
Harlan Stenn <stenn@nwtime.org>
http://networktimefoundation.org - be a member!