You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
On the Time Zone Database mailing list it was reported that Elixir's tzdata package by default checks on startup and then once per day whether the time zone database has been updated. It does so by consulting a URL at https://data.iana.org.
Please don't do that. data.iana.org is not intended to be hammered on by every computer on the planet. Instead, please arrange for the Elixir tzdata package to grab the data from somewhere else, so that Elixir programs don't overload the central TZDB server at iana.org. One place to do that might be https://github.com/lau/tzdata/somewhere; or perhaps you can think of a better place.
In addition to being a better network citizen, an advantage of copying from a downstream copy of TZDB that when there's a compatibility glitch in the latest TZDB release, such as occurred in TZDB 2024b which Elixir tzdata couldn't immediately read, the Elixir tzdata maintainers can fix the issue in the downstream copy of the TZDB database instead of waiting for the next TZDB release. This sort of thing is standard practice in GNU/Linux distributions of tzdata proper.
Thanks.
The text was updated successfully, but these errors were encountered:
And all other sources of tzdata do their own (re-)packaging and distribution of the upstream data and code changes: Android, *BSD, iOS/OSX, Windows Apps, etc. so they can decide to skip any changes or releases that may cause any issues, reduce the distributed tzdata to that most useful to their clients, and provide early custom updates for their clients, prior to official releases, which may be delayed by government bureaucracy.
On the Time Zone Database mailing list it was reported that Elixir's tzdata package by default checks on startup and then once per day whether the time zone database has been updated. It does so by consulting a URL at https://data.iana.org.
Please don't do that. data.iana.org is not intended to be hammered on by every computer on the planet. Instead, please arrange for the Elixir tzdata package to grab the data from somewhere else, so that Elixir programs don't overload the central TZDB server at iana.org. One place to do that might be https://github.com/lau/tzdata/somewhere; or perhaps you can think of a better place.
In addition to being a better network citizen, an advantage of copying from a downstream copy of TZDB that when there's a compatibility glitch in the latest TZDB release, such as occurred in TZDB 2024b which Elixir tzdata couldn't immediately read, the Elixir tzdata maintainers can fix the issue in the downstream copy of the TZDB database instead of waiting for the next TZDB release. This sort of thing is standard practice in GNU/Linux distributions of tzdata proper.
Thanks.
The text was updated successfully, but these errors were encountered: