The following applies to Vera V1@1.1.1186 (should be the same for Vera V2@1.1.1234):
Last night (2011-03-27) there was a clock shift from 02:00 CET to 03:00 CEST due to DST.
DAD updates the data for sun and moon once a day using a Luup timer (between 00:00:01 and 00:01:00).
2011-03-26 23:00:54 DAD[16872]: Timer: 2011-03-26 04:22:21 SunTwilightAstronomicalBegin, result: -1
2011-03-26 23:00:54 DAD[16872]: Timer: 2011-03-26 05:05:00 SunTwilightNauticalBegin, result: -1
2011-03-26 23:00:54 DAD[16872]: Timer: 2011-03-26 05:45:25 SunTwilightCivilBegin, result: -1
2011-03-26 23:00:54 DAD[16872]: Timer: 2011-03-26 06:19:18 SunRise, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 18:52:19 SunSet, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 19:26:14 SunTwilightCivilEnd, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 20:06:42 SunTwilightNauticalEnd, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 20:49:29 SunTwilightAstronomicalEnd, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 02:30:00 MoonRise, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2011-03-26 10:22:00 MoonSet, result: -1
2011-03-26 23:00:55 DAD[16872]: Timer: 2038-01-19 02:14:07 PossibleImpact2007VK184, result: 0
2011-03-26 23:00:55 DAD[16872]: Timer: 2038-01-19 02:14:07 PossibleImpact2011AG5, result: 0
Lua (first timestamp) gets it right, LuaUPnP gets it wrong …
In addition, a scene scheduled for 2011-03-27 01:00 CET was executed at 2011-03-27 00:00 CET.
I cannot believe that such a basic functionality doesn’t work! I reported the bug a year ago!