onno@titan.hts.hsa.nl (Onno Witvliet) (03/29/90)
I have some discussion with HP about the definition of the timezone notation
in the sendmail headerlines. I have read the RFC 822, but I can't find the
"real" definition.
I have the following configuration:
- a HP9000-825 with HP-UX 7.0 sendmail
- the local timezone is MET (Middle European Time), which is the time 1 hour
east of Greenwich.
The problem is as follows:
When daylight savings time is NOT in effect, HP sendmail gives the date in the
following notation:
Received: by titan.hts.hsa.nl (HP) Mon, 05 Mar 90 18:11:00 -0100
I have ported sendmail 5.61++ which gives the notation just as I think it
should be:
Received: by titan.hts.hsa.nl (5.61++) Mon, 05 Mar 90 18:11:00 +0100
According to our local HP response-center the first notation is correct, but
I think it should be +0100, since all American timezones are negative according
to the RFC 822 page 26
When daylight savings time is in effect, HP sendmail gives the date in the form:
Received: by titan.hts.hsa.nl (HP) Mon, 26 Mar 90 18:11:00 met
but sendmail 5.61++ gives:
Received: by titan.hts.hsa.nl (5.61++) Mon, 26 Mar 90 18:11:00 +0200
which is correct in my opinion
My questions are:
- what is the real definition for date en time specification
Is the definition:
local-time = GMT + timezone
or
local-time + timezone = GMT
(Are the timezones east of Greenwich positive or negative, so
is MET = +0100 or is MET = -0100 right, Am I right or is HP right)
- is met (lower-case) legal
- on what BSD version of sendmail is HP sendmail based
Thanks in advance
Onno
--
Onno Witvliet, Hogeschool Alkmaar | Email: onno@titan.hts.hsa.nl
Bergerweg 200, 1817 MN Alkmaar |
Phone: (31)(0)72-183419 | Uucp: {backbones}!hp4nl!halka!onno