> > I'm a bit confused about the state of DST handling; there was a patch > > applied in Debian revision -53, but the patch is large (and combined > > with other changes) so the (intended) effect of the changes aren't > > entirely clear to me. As I recall, that patch also seemed to have > > been applied in the ISC release. > > As I can see here's the openbsd patch included into debian version -53: > http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin/cron/cron.c.diff?r1=1.3&r2=1.4&f=h
> Thanks; my current understanding is that the original patch worked > when the time was manually changed by 1-3 hours, but didn't work for > real DST changes (the details aren't clear to me). I guess you are right. The core change is in set_time routine which didn't detect DST change before because time() always returns seconds since epoch. Now they use localtime(), save tm_isdst flag and check it. When the flag changes -- here comes the DST change. > BTW, I think there's a new bug here: > http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin/cron/misc.c.diff?r1=1.8&r2=1.11&f=h We might use the following code to calculate GMT offset which seems cleaner to me than the original sendmail code they are using: int get_gmtoff(time_t *clock) { struct tm *ps; time_t tmp1, tmp2; ps = localtime(clock); tmp1 = mktime(ps); tmp2 = timegm(ps); return(tmp2 - tmp1); } Petya.
