hi,
GoodCrypto Support wrote (05 Jan 2016 20:30:45 GMT) :
> > ... should be in the htpdate service ... systemd
For the record, on our devel branch (based on Debian Jessie), htpdate
is managed by systemd:
https://git-tails.immerda.ch/tails/tree/config/chroot_local-includes/lib/systemd/system/htpdate.service?h=devel
The service will enter active (running) state if, and only if, htpdate
has succeeded.
> There are always better ways. We haven't worked with perl for many years, and are
> just starting to tame systemd. Something that works now is usually better
> than waiting.
nn
Yes, most of the time (and depending on how other problems "something
that works" might introduce, such as maintainability issues). In this
case, however, we're not quite there yet, as important bits are
missing before we have "something that works now": specifically,
integration with how we start htpdate in Tails (via a NetworkManager
hook).
> If persistent time sync failure isn't important enough to work
> around, why include htpdate?
Indeed. We want the same thing, it seems:
https://labs.riseup.net/code/issues/10494
Cheers,
--
intrigeri
