Miroslav Lichvar wrote:
There is one already there, sort of. Leap seconds are included in the
"right" timezones from the tzdata package. If you use a distribution
that updates the tzdata package regularly and keep your system up to
date, the information about an upcoming leap second should be included
in the timezones before it actually happens.

chrony can be configured to use this by adding "leapsectz right/UTC"
to chrony.conf. Restart is not necessary when tzdata is updated, it
will be reloaded automatically.

However, this is not equivalent to having the NIST leapseconds file as
the expiration date is lost, the applications can't tell if there
really is no leap second in a given month or that the timezone just
wasn't updated yet.

There has been a discussion about this on the tzdist mailing list. See this thread:
http://www.ietf.org/mail-archive/web/tzdist/current/msg01132.html

I have proposed to use the IERS file as authoritative source, since the NIST file doesn't seem to be accessible from all over the world. See David Taylor's post on this list some hours ago.

The IERS file now also provides an expiration date, and the maintainer of the TZ data base, Paul Eggert, has mentioned that he has added some code which now provides the expiration date in the TZ DB leapsecond file:
http://www.ietf.org/mail-archive/web/tzdist/current/msg01135.html

However, I haven't tried this and I'm not sure which changes are required in some libraries or applications to *evaluate* the expiration date.

Current versions of ntpd should handle the expiration date, however the one provided by the NIST-style file.

Martin
--
Martin Burnicki

Senior Software Engineer

MEINBERG Funkuhren GmbH & Co. KG
Email: [email protected]
Phone: +49 (0)5281 9309-14
Fax: +49 (0)5281 9309-30

Lange Wand 9, 31812 Bad Pyrmont, Germany
Amtsgericht Hannover 17HRA 100322
Geschäftsführer/Managing Directors: Günter Meinberg, Werner Meinberg, Andre Hartmann, Heiko Gerstung
Web: http://www.meinberg.de
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to