Philip Warner <pjw@rhyme.com.au> writes:
> At 00:59 3/04/01 -0400, Tom Lane wrote:
>> However, the horology diffs are not, and I can't reproduce them here.
>> Did anyone else see that?
> I've just started seeing both...
What is the date of the nearest daylight-savings transition in your
timezone?
Wait a minute ... considering that the regress tests run in PST8PDT,
your local timezone shouldn't make a difference. Maybe a platform-
specific issue? What platform (esp. which C library) do you use?
FWIW, as of this morning I'm back to no failure on timestamp test
(as expected), and still no horology failure either.
regards, tom lane