Brian Hirt <bhirt at me.com> writes:
>> 
>> Actually, that's backwards.  We have always used timestamp *without*
>> timezone, and the proposal is to change to use timestamp *with*
>> timezone.
>> 
>> There's a small issue that "with timezone" consumes a little more space;
>> I wouldn't consider that terribly important.
>> 
>> In my view, to use a timezone which has a known incontinuity (e.g. -
>> where you can expect a 1 hour "leap" every so often) *is* the footgun.
>
> Christopher,
>
> You are right, i did get it backwards and say it the other way around.
> Right now the timestamp in the sl_* tables are timestamp without time
> zone.  If they were all timestamp with time zone, couldn't the slon
> clients just issue a "set timezone TO 'UTC'" command when they start
> up and everything would work nicely because postgres would be able to
> translate to a time zone that slony would like to see the data in.

Ah, and you're probably right there.

We have places where we force other aspects of environment.

Notably, the logtrigger functions force DateStyle to "ISO".

It doesn't seem "out there" to have slon processes expressly operate in
UTC, which should address this issue.  I'll put in a bug that indicates
this as one of the options, at any rate.
-- 
output = reverse("ofni.sailifa.ac" "@" "enworbbc")
Christopher Browne
"Bother,"  said Pooh,  "Eeyore, ready  two photon  torpedoes  and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"