From owner-freebsd-questions Mon Apr 5 11:58: 3 1999 Delivered-To: freebsd-questions@freebsd.org Received: from revolution.3-cities.com (revolution.3-cities.com [204.203.224.155]) by hub.freebsd.org (Postfix) with ESMTP id 1E09914D3E for ; Mon, 5 Apr 1999 11:57:56 -0700 (PDT) (envelope-from kstewart@3-cities.com) Received: from 3-cities.com (kenn1227.bossig.com [208.26.241.227]) by revolution.3-cities.com (8.8.5/8.7.3) with ESMTP id LAA02278; Mon, 5 Apr 1999 11:55:24 -0700 (PDT) Message-ID: <37090724.E59C2A08@3-cities.com> Date: Mon, 05 Apr 1999 11:55:32 -0700 From: Kent Stewart Organization: BOSSig X-Mailer: Mozilla 4.51 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 To: Mark Ovens Cc: Greg Lehey , flygt@sr.se, FreeBSD Questions Subject: Re: Timezone question References: <19990404044642.A60884@sr.se> <19990404132026.T2142@lemis.com> <19990404125723.A61426@sr.se> <19990405010416.L299@marder-1.localhost> <19990405095153.I2142@lemis.com> <19990405015204.A257@marder-1.localhost> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Mark Ovens wrote: > > On Mon, Apr 05, 1999 at 09:51:53AM +0930, Greg Lehey wrote: > > On Monday, 5 April 1999 at 1:04:17 +0100, Mark Ovens wrote: > > > > > > Does FreeBSD update the RTC for Daylight Saving? > > > > You mean the clock? No. > > > > > I only ask because my machine was booted to FreeBSD when the clocks > > > changed and the next day FreeBSD was showing the correct time. When > > > I rebooted the RTC was showing the correct time as well. > > > > I'm not sure what you mean by RTC. If you mean the CMOS clock, > > Yes, I do. Am I getting my terminology mixed up here? RTC == "Real > Time Clock". I've always known RTC to be a synonym for CMOS clock. I think your usage is more correct. The CMOS is really memory and doesn't have a clock. There is a real time clock on the mother board, which has part of it's information displayed in the CMOS bios display. On my mother boards, the RTC is a Dallas Semiconductor Real Time Clock. The Dallas RTC is contained in an IC looking package that now has a battery built in. It has always been a standalone chip and not part of the CMOS. When the battery dies, you replace the entire RTC. > > > it *should* be showing UTC. UNIX doesn't use local time, it converts the > > time representation to wherever you happen to be. That's why you can > > use the TZ environment variable to display time anywhere in the world. > > > > Having said that, FreeBSD has a kludge for coexisting with Microsoft > > on the same machine: it's called adjkerntz, and it does just about > > what you're describing, so you probably have it running. You can > > check if it's running with ps(1). Unless you happen to run Microsoft > > on the same machine, you can safely remove it. > > > > adjkerntz is running (I have M$ on this machine). I've just rebooted > and gone into BIOS setup to confirm the CMOS clock time and it is > showing the correct time. Since you say FreeBSD doesn't change the > CMOS clock I think the BIOS must take care of the change to DST > and since 95 & NT are both set to *not* adjust for DST everything > is fine. > > I don't have a problem, everything is working, I was just curious > *why* it was working. > > > > FWIW, both W95 & NT also both showed the correct time (probably > > > because the RTC was correct & both Winblows had "Automatically > > > adjust for DST" *unchecked*) although 95 correctly thinks the TZ > > > is GMT, Daylight Saving but NT thinks it is just GMT. > > > > You'd think they'd get these things right, wouldn't you? > > > > Now don't be silly Greg, this is M$ we're talking about ;-) > > BTW, In a previous post of yours in this thread you mentioned about > Windows not knowing about the TZ in Northern Territories, which > from then on you referred to as NT. The last sentence (I've deleted > the message so I can't quote it verbatim) read something like > "...obviously Windows doesn't know about NT". I don't know if the > double entendre was intentional but it was highly amusing either > way :-) This whole conversation made me curious and I went to see what MS had in their knowledge database. It turns out that several countries were not handled properly by W95 and NT4-. The two that I saw were Australia and Finland. In the kb article Q183907 (18 Mar 1998) they provide a URL at MS.au where you can download a program that will fix the time zone problems for Australia. You had to add a timezone for Finland by editing the registry. MS has changed the way the display the timezones and no longer highlights the region. The discontinuities were given as the reason. When you change timezones you look up the city within your timezone and that sets up your DT/ST parameters. What I was really curious about was their abreviations for the timezones but I couldn't find anything on that. A couple of the DOS based programs used by customers required that they set the tz environment variable and I was concerned about side effects. I don't have coexistant problems on a particular computer but there is a mix of computers on the networks and they all exchange messages. Time and the conversion to GMT is an important QA parameter. An hour shift would/could represent an entirely different product and is something that an auditor with an agenda would catch. -- Kent Stewart Richland, WA mailto:kstewart@3-cities.com http://www.3-cities.com/~kstewart/index.html To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message