Am Sun, 29 Mar 2015 16:49:21 -0600
schrieb Canek Peláez Valdés <[email protected]>:

> On Sun, Mar 29, 2015 at 4:32 PM, Marc Joliet <[email protected]> wrote:
> >
> > Am Sun, 29 Mar 2015 12:48:18 -0600
> > schrieb Canek Peláez Valdés <[email protected]>:
> >
> > [...]
> > > Are you really sure "0/2:00" means "every 2 hours"? I don't see an
> explicit
> > > mention in man 7 systemd.time that 0 means "*-*-* 00:00:00". It really
> > > worked bi-hourly before?
> >
> > Yes, it definitely worked before (I've been running this and other timers
> for
> > about a month). I don't remember how I inferred that rule, but I think it
> was
> > this bit from systemd.time(7):
> >
> >     "Either time or date specification may be omitted, in which case the
> >     current day and 00:00:00 is implied, respectively. If the second
> component
> >     is not specified, ":00" is assumed."
> >
> > But I don't see any definition for these components, so maybe I'm wrong
> and my
> > timer only works by accident.
> >
> > > Either way, it cretainly could be a bug.
> >
> > Perhaps, since it's back to the way it was before:
> >
> >     # systemctl list-timers
> >     NEXT                         LEFT          LAST
>   PASSED             UNIT                           ACTIVATES
> >     Mo 2015-03-30 02:00:00 CEST  1h 44min left Mo 2015-03-30 00:00:00
> CEST  15min ago          backup-hourly.timer
>  [email protected]
> >     [...]
> 
> Perhaps the bug is only in how systemctl presents the information.

If I understand you correctly, then no: [email protected] did not run
in spite of the contradictory next run time.  The only times it ran were when I
manually started it intermittently.

I suppose I'll have to file a bug report.  At the very least I'll find out if
this was a systemd bug or not.

Greetings
-- 
Marc Joliet
--
"People who think they know everything really annoy those of us who know we
don't" - Bjarne Stroustrup

Attachment: pgp0KI5QEnPRb.pgp
Description: Digitale Signatur von OpenPGP

Reply via email to