Why is my system rebooted?
rober...@hotmail.com
I have a DELL PowerEdge 6300 with SCO Openserver 5.0.4 and RS504C
and OSS469B(OSS469C can not be installed) installed, and sometimes it
will reboot automatically without any information , any reason.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
John Gray
Can't be sure but turn on the kernel debugger scodb and relink and
reboot.
If the reboot can be caught in software the debuuger will catch it.
edit /etc/conf/sdevice.d/scodb and change the N to Y. cd /etc/conf/cf.d
and ./link_unix reboot.
^X breaks into the debugger and q will quit out.
If you are in the debuger the reboot() command will force the system
to reboot.
good luck
-john
Michael A. Gaul
at random times during the day for no apparent reason: it would behave as
though
someone had hit the "reset" button. This started happening all of a sudden,
after many
years of successful operation with the same system, os, etc.
The symptom persisted after he changed his multi-port serial board, cabling,
motherboard,
power-supply, etc.etc. After many months the puzzle was revealed:
(1) His company was built over a subway line. The system would tend to
reboot
when a train ran underneath.
(2) Shortly before the symptom originally appeared, "the electricians" had
made some
changes to his building's wiring.
It turns out that in doing so they had disrupted his "earth" ground. This,
combined with the
electrical fields from the train. managed to couple to his reset signal.
rober...@hotmail.com wrote in message <7oo3gv$g6s$1...@nnrp1.deja.com>...
Kenneth McCormick
Cool story. If Robert doesn't have a subway running below,
nor a shorting power cord nor loose cable, etc.,
what he's experiencing is a catastrophic hardware failure due
most likely to his master interrupt controller. Software failure
would be caught by the kernel, kernel failures would yield a kernel trap,
often seen on NT as the blue screen of death, but certain hardware failures
can't be caught and cause the system to spontaneously reboot. These failures
corrupt the bus and the system reboots without any notice. This used to
be caused, more often than not, by video cards and loose chips on the mobo,
but as system MHz has gone up, the interrupt controller and the cmos now tend
to conflict. This conflict happens when the keyboard or mouse trys to raise
their interrupt. The poorly constructed Winbond W83977xx master interrupt
controller, that is used on almost every modern motherboard, is responsible.
Look for it near the ISA slots towards the back of the mobo. There are
many many posts to the comp.periphs.mainboard.abit ng about these
exact kinds of problems. I have seen it many times over the years, and
the problem goes away with a new mobo, preferable an Asus for me.
Of course you are welcome to swap out every single piece of hardware
one at a time with identical working prieces as I did many times, but
it always seems to be the mobo when I see this problem.
Kenneth
Bill Vermillion
In article <MPG.121bb1fe8...@news.earthlink.net>,
>|One of my company's customers had such a problem: His system would
>|reset at random times during the day for no apparent reason: it
>|would behave as though someone had hit the "reset" button. This
>|started happening all of a sudden, after many years of successful
>|operation with the same system, os, etc.
>|It turns out that in doing so they had disrupted his "earth"
>|ground. This, combined with the electrical fields from the train.
>|managed to couple to his reset signal.
>Cool story. If Robert doesn't have a subway running below,
>nor a shorting power cord nor loose cable, etc.,
>what he's experiencing is a catastrophic hardware failure due
>most likely to his master interrupt controller.
Not neccesarily. But again it is not to be ruled out.
I've seen spontaneous reboots in offices that were stable.
On investigation I found that there was a building addition being
done and some of the power equipment was putting surges on.
In another - where there was a dedicated computer line run
directly from the main breakers, I discovered the voltage to the
computer was showing a 5 VOLT drop anytime someone used
the big copy machine.
It seems that the ONLY thing plugged into the dedicated line
was a small portable radio.
Other legendary reports from the past have also shown a pattern
of resets - eg certain times of day. Those have been traced
to such things as a new microwave in the lunch room, and the
computer always seemed to reboot about noon. Brand new
RF towers - either cell phone - or local/private radio systems.
In some instances you really need a serious UPS - true UPS not
standby - with severe filtering such as BEST (or others in that
line). APC makes some really BIG suckers too, but I've not seen
specs on theirs. (I did see a full size air-condited equipment
rack and built-in power supply and air-conditioner. I think that
was APC too)
One other thing - the problem could be occuring in connected
peripheral equipment. Anthing that has a hardwire connection to
the server is a potential candidate for problems.
The original poster said it acted like someone hit the reset
button.
It could be a memory problem IF the system was not using parity
RAM. In that case you'd get a PANIC message, and a shutdown.
If using standard RAM or a motherboard not set up for PARITY,
it will just stop.
I fought that one once when a vendor did not put in the specified
motherboard and put in something cheaper. This started happening
after about 2 years, and was immediately after rewiring to add new
terminals. So that was immediately suspect. The questionable wire
in place was replaced with properly installed cable. Still the
problems.
Since all the machines I had specified would note memory errors and
generate a PANIC message, and that had been specified for that
machine, memory was the last place we looked.
(Of course it would be the last place we looked. Looking further
after finding that would be counter-productive :-0 )
weeks of power-monitoring with really good equipment,
Bill
--
Bill Vermillion bv @ wjv.com
Bill Vermillion
Bill Vermillion
Colin Woodford
this is not SCO..... Our old ICL Mainframe
used to hang at 11am (ish) every day.......
All the cream of ICL took the system apart
and rebuilt it but it still hung at 11am.
Guess What....... The cleaner used to hoover the
Computer room around 11 each day and
the odd dodgey spike from the motor used
to crash the system!!
Kindest Regards to all
Colin Woodford. S.C.C.
Michael A. Gaul wrote in message <7orurp$d2e$1...@nntp4.atl.mindspring.net>...
>One of my company's customers had such a problem: His system would reset
>at random times during the day for no apparent reason: it would behave as
>though
>someone had hit the "reset" button. This started happening all of a sudden,
>after many
>years of successful operation with the same system, os, etc.
>
>The symptom persisted after he changed his multi-port serial board,
cabling,
>motherboard,
>power-supply, etc.etc. After many months the puzzle was revealed:
>
>(1) His company was built over a subway line. The system would tend to
>reboot
>when a train ran underneath.
>
>(2) Shortly before the symptom originally appeared, "the electricians" had
>made some
>changes to his building's wiring.
>
>It turns out that in doing so they had disrupted his "earth" ground. This,
>combined with the
>electrical fields from the train. managed to couple to his reset signal.
>
>