uptime 136 year

When my nto 6.2 just started it displays a very nice uptime:

uptime

localhost x86 49710d, 05:43 since Wed Feb 25 10:21:40 2004

date

Wed Feb 25 09:36:58 CET 2004

date -u

Wed Feb 25 08:37:00 UTC 2004

spin too gives the optimistic uptime.
Note the start time is in the future. After one hour the uptime is given
much lower values.
My question is: is this an error in my configuration or is it a bug in
uptime/spin/library ?
My rtc is in localtime, TZ is set in the sysinit file:

echo $TZ

CET-01CEST-02,M3.5.0/2,M10.5.0/2

Pim

Reality is bad enough, why should I tell the truth?
– Patrick Sky

My guess would be that the syspage has bogus boot time. What hardware is it?

– igor

“Pim Bollen” <pbollenNOSPAM@vimec.nl> wrote in message
news:c1hn26$kjd$1@inn.qnx.com

When my nto 6.2 just started it displays a very nice uptime:

uptime

localhost x86 49710d, 05:43 since Wed Feb 25 10:21:40 2004

date

Wed Feb 25 09:36:58 CET 2004

date -u

Wed Feb 25 08:37:00 UTC 2004

spin too gives the optimistic uptime.
Note the start time is in the future. After one hour the uptime is given
much lower values.
My question is: is this an error in my configuration or is it a bug in
uptime/spin/library ?
My rtc is in localtime, TZ is set in the sysinit file:

echo $TZ

CET-01CEST-02,M3.5.0/2,M10.5.0/2

Pim

Reality is bad enough, why should I tell the truth?
– Patrick Sky

Igor Kovalenko wrote:

My guess would be that the syspage has bogus boot time. What hardware is
it?

– igor

It’s an AAEON SBC-675B.

However when I put the UTC time in the BIOS it works normal.
Of course this is the preferred situation.

Pim

Immigration is the sincerest form of flattery.
– Jack Paar

kovalenko@attbi.com sed in <c1ju86$io3$1@inn.qnx.com>:

My guess would be that the syspage has bogus boot time. What hardware is it?

When using localtime in the RTC, uptime (pidin info) output
doesn’t look correct;
use “TZ=GMT pidin info”, read the output in localtime and you’ll get
something close.

As kernel won’t know about timezone until /etc/TIMEZONE is available but
_SYSPAGE_ENTRY(sysinfo,qtime)->boot_time is written before that,
hence whatever RTC time is written for the boot time.

Using UTC for RTC should be the “correct” solution.
It won’t be a problem unless dualbooting with Microsoft.

kabe