On 2010-04-05 at 23:39 +0100, Always Learning wrote:
> CEDT (Central European Daytime Time) is a American invention which
[snip]
The common timezone names on the Internet come from the Olson timezone
database. You can retrieve the current code and definitions from
ftp://elsie.nci.nih.gov/pub/ -- yes, it's American, so what? At least
*someone* is sponsoring a common standard. The effort is international
and, AFAIK, volunteer. In fact, I believe the most prominent
contributor, besides Mr Olson himself, is Dutch (Paul Eggert).
This is the source of names such as Europe/Amsterdam, which maps to
CET/CEST. The file doesn't claim to be authoritative, but it is the
de-facto standard which almost everyone uses these days. They rush out
the updates when countries decide to change their mind about switching
to summer time in the middle of the week leading up to the cut-over, or
when Brazil fragments apart in terms of summer-time, and generally do a
*phenomenally* good job of tracking the chaos and confusion of this
intensely political mess.
It's bad enough already, without casual slurs being thrown around.
(Specific and accurate slurs are another matter. A certain large
computer OS manufacturer not managing to get "GMT" right, for
instance; that just makes things worse for everyone)
For what it's worth, this is the current definition for The Netherlands,
with the current rule it maps to.
----------------------------8< cut here >8------------------------------
# Zone NAME GMTOFF RULES FORMAT [UNTIL]
Zone Europe/Amsterdam 0:19:32 - LMT 1835
0:19:32 Neth %s 1937 Jul 1
0:20 Neth NE%sT 1940 May 16 0:00 # Dutch Time
1:00 C-Eur CE%sT 1945 Apr 2 2:00
1:00 Neth CE%sT 1977
1:00 EU CE%sT
# Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S
Rule EU 1981 max - Mar lastSun 1:00u 1:00 S
Rule EU 1996 max - Oct lastSun 1:00u 0 -
----------------------------8< cut here >8------------------------------
(The EU has had solid rules for the past 14 years and I believe most of
the EU has actually standardised on using the EU change dates, which is
a good job).
Downloading the Olson DB commands and DB is useful if you ever need to
professionally keep up-to-date on the insanity, since OS distributors
tend to prioritise updates only if major market areas are affected by
changes. Which is a reasonable cost/benefit trade-off on their part.
As to the original:
* check to see if mail(1) is inserting a Date: header;
* echo $TZ; date
* if the listening server has the correct timezone, then it's highly
likely that the submitting server, using the same config, will be
using the same configuration. The exception would be if the server
config explicitly inherits the timezone from the environment.
* exim -d is your friend
* for sanity, I *highly* recommend this exim.conf setting:
timezone = UTC
as this means your log-files never repeat a time-window and the trace
headers you insert are going to be understood by any system
administrator, anywhere in the world, without having to contrast
CST/CDT against CET/CEST and going bat#@$% crazy as they debug mail
delays.
* Telling Exim to use UTC will of course put UTC timestamps into any
Date: headers which have to be added by the MTA. If you care, make
the client add the Date: header. Most decent MUAs will translate
times into the local timezone though, so it shouldn't matter for the
end-user.
Regards,
-Phil
