Brian Inglis via tz wrote:
On 2023-12-15 02:31, Martin Burnicki via tz wrote: [...]
Has anyone ever suggested a formulation that meets the requirement of the TZDB project, and told them why this was a requirement? I have no doubt that the folks at IERS would pick it up and integrate it.
I was in touch with them when they initially provided the files, about their often inconsistent validity and expiry dates and timestamps, often expiring before the next Bulletin C would be issued, until they standardized on 0Z on the Bulletin C issue date and the 28th of the last month of the half year following the expected date the next Bulletin C is issued.
Yes, I remember the first publications had some 'glitches', and I had several discussions with the guys. The hardest part was to explain why it is best practice (under the given conditions) to set the expiration date almost 12 months in the future. The problem was that they didn't know how the NTP leap second file is being used. They were assuming that it would be better to let the currently installed file expire, then generate some alarm that the file needed to be updated and then update the file. However, this would mean that the NTP daemon runs for some time without leap second information. BTW, the leap second file is also a good way to provide a device with the TAI/UTC offset, which is important if you run for example a PTP grandmaster. Martin -- Martin Burnicki Senior Software Engineer MEINBERG Funkuhren GmbH & Co. KG Email: martin.burnicki@meinberg.de Phone: +49 5281 9309-414 Linkedin: https://www.linkedin.com/in/martinburnicki/ Lange Wand 9, 31812 Bad Pyrmont, Germany Amtsgericht Hannover 17HRA 100322 Geschäftsführer/Managing Directors: Günter Meinberg, Werner Meinberg, Andre Hartmann, Heiko Gerstung Websites: https://www.meinberg.de https://www.meinbergglobal.com