[Ilugc] Re: BIOS and other timestamps

[Ramanraj K]

Chandrashekar Babu wrote:
 > Ramanraj K writes:
 >> A little OT but:
 >> I posted about the "Dreaded Deep Blue Screen" at about 8:40 AM but
 >> Chandrashekar Babu replied to that by 7:20 AM ? Did the monsoon blues
 >> eat up some clock ticks as well?
 >
 > The problem has actually got to do with my machine's time settings. When
 > I boot into FreeBSD and then reboot back onto Linux, I lose 5:30 hours.
 > Oh well, on both OS, I've set my timezone to Asia/Calcutta (Asia/India)
 > or IST, respectively. I'm a bit clueless about what might have gone
 > wrong. Infact, I'm facing this issue across different Linux distros
 > setup in multiboot configuration too, lately [Slackware 10.0 with
 > Mandrake 10.1 CE, for instance].
 > Maybe I must add 'ntpdate' in my system startup scripts.

FreeBSD may be the main reason, because it does not use  BIOS and may be
writing with a diff of 5:30 hrs while shutting down.  Using LinuxBIOS 
program, may solve the problem.

Sivasankar wrote about BIOS dependencies earlier:
http://www.ae.iitm.ac.in/pipermail/ilugc/2004-August/011650.html
in reply to
http://www.ae.iitm.ac.in/pipermail/ilugc/2004-August/011649.html


 >> BTW, should'nt GNU Mailman be stamping
 >> time on mails to prevent ante-dated postings?
 >
 >
 > I dont know if thats a good idea. Though in a typical postal mailing
 > system, the postal service would stamp date and time(?) seal _on_ the
 > cover of the letter, they cannot or should not change the date and time
 > stamp of the letter which is _in_ the cover (which is set by the
 > sender). I guess its a similar analogy in here too.
 > Another good reason that I could think of: It lets us figure the time
 > when someone sent an email from _his_ location. Now, if that person
 > resided in the US, and GNU mailman was setup here in India (following a
 > different timezone of course) and is configured to modify date/time
 > stamp of all email that it forwards, the whole purpose would be defeated.

GNU Mailman may carefully wrap and keep the date posters give, but it
should rely on its own timestamp to archive the message. My previous
mail posted this evening was dated 2004-10-28, and archived that way
taking it for granted. For many reasons, the poster may supply incorrect
dates without any serious consequences here, but mailman could keep the
records straight by using its own timestamp for reliable future referencing.

Regards,
Ramanraj.