----- Original Message -----
> Hey Marko,
>
> Marko Myllynen <myllynen@xxxxxxxxxx> writes:
> > The easiest solution seems to work.
> >
> > ---
> > src/pmrep/TODO | 1 -
> > src/pmrep/pmrep.py | 15 ++-------------
> > 2 files changed, 2 insertions(+), 14 deletions(-)
>
> After looking through the buildbot results from last night[1] (which
> include this change). It appears that qa/1069 is still failing,
> specificially due to timestamp mismatches from the archive (using local
> time).
>
> The testcase does appear to use -z for most tests, however the tests are
> using my localtime (06:00ish) vs archive time(20:00ish). Is this a
> known issue? I've confirmed the relative timezones by manually
> checking the output of pmlogcheck sample-secs -l [-z].
>
Ah, thanks guys - yes, we need to be using the PCP-timezone-aware time APIs
in pmrep when reporting the sample time - I'll push a fix through shortly.
cheers.
--
Nathan
|