|
|
| Author |
Message |
ea
|
|
Forcing a crash
| 
Oct 9 18:54 UTC 2000 |
What would you do to try to force a computer to crash? I'm trying to
repair a supposedly damaged computer. The description of the problem is
"Freezes and sometimes restarts itself randomly" First I tried a virus
scan and scandisk scan to see if that was the problem. Negative. So,
then I tried to recreate the problem. No matter what I did, I could not
crash it. I tried opening a bunch of windows, using lots of
badly-written java, opening really large numbers of documents, and
playing an MP3 all at once. I managed to get processor usage at 100%
for over 5 minutes, and memory usage was above 90%. Still no crash.
I've tried everything I can think of. What do you suggest?
Limitations: I cannot do anything with the hardware, cannot change the
OS (currently Win95), cannot install any new software.
|
| 30 responses total. |
gull
|
|
response 1 of 30:
|
Oct 9 19:24 UTC 2000 |
The "cannot install any new software" limitation is a tough one. My first
thought would be to download a copy of MEMTEST-86 and run it. It's something
you write out to a floppy, which you then reboot the machine with. It does
a really good job finding bad SIMMs in my experience, though some machines
refuse to run it. If you made the floppy on a different machine you
wouldn't technically be installing any new software.
|
tod
|
|
response 2 of 30:
|
Oct 9 19:25 UTC 2000 |
I'd load it with NT 4.0 SP3 and a Tivoli product. *snort*
|
ea
|
|
response 3 of 30:
|
Oct 9 19:33 UTC 2000 |
Floppy I can do, as long as it doesn't write to the HD.
|
rcurl
|
|
response 4 of 30:
|
Oct 9 19:53 UTC 2000 |
Find out what the user was doing when it crashed. Most of my system's
crashes are due to running obsolete software simultaneously with
newer software, or trying to print from obsolete software. I can almost
guarantee some crashes for my system.
|
scott
|
|
response 5 of 30:
|
Oct 9 19:58 UTC 2000 |
Use desk junk to block the air vents, which is what the user might have been
doing.
|
n8nxf
|
|
response 6 of 30:
|
Oct 9 20:02 UTC 2000 |
Perhaps the user is tapping on the power strip with his foot while using his
computer. Perhaps the plug he has it plugged into does not hold on to the
prongs tightly?
|
ea
|
|
response 7 of 30:
|
Oct 9 20:26 UTC 2000 |
It's a laptop. The power thing shouldn't matter too much unless the
battery's totally drained.
The only thing I've found so far that could be causing it - the laptop
has a fan. The fan is making a grinding noise, so I'm thinking that it
could be excessive heat.
|
aruba
|
|
response 8 of 30:
|
Oct 9 20:32 UTC 2000 |
I think you need some more information, Eric.
|
ea
|
|
response 9 of 30:
|
Oct 9 20:43 UTC 2000 |
I've left a message asking for the owner to call back and describe
what they're trying to do when it crashes. I haven't been able to find
any problems.
|
abndaddy
|
|
response 10 of 30:
|
Oct 10 01:16 UTC 2000 |
Hello my name is Chad. I have been on Grex for about 2 weeks. I am 30 years
old and I am a Special Forces Communications Sergeant in the US Army. I am
learning to Speak Serbian here in the states right now, and next year I will
move to Germany. I have little experience with UNIX, but am trying to learn
it. Back in the mid 80's I had my own BBS in Louisana. Boy, have things
changed!
|
tod
|
|
response 11 of 30:
|
Oct 10 02:02 UTC 2000 |
Semper Fi, Chad!
|
n8nxf
|
|
response 12 of 30:
|
Oct 10 03:40 UTC 2000 |
Perhaps he's pounding on it when he gets upset? I had a laptop crash
whenever the case was twisted a little. It turned out to be a loose
connector inside. Is the battery good? i.e. Dose it have its rated
capacity? Is there an intermittent connection in the battery or between]the
battery and the computer? Are the contacts in good condition?
|
bdh3
|
|
response 13 of 30:
|
Oct 10 03:49 UTC 2000 |
There are LCD temperature strips that you can get. Mount one over the
fan and one on the front and have the user record both temperatures
after a crash. It sounds a little like a temp. thing to me.
|
johnnie
|
|
response 14 of 30:
|
Oct 10 16:57 UTC 2000 |
I had a PC that did the "freeze and sometimes restart randomly" thing.
I popped the top and made sure everything was connected tightly.
Evidently something wasn't, 'cuz that was the end of the problem.
|
tod
|
|
response 15 of 30:
|
Oct 10 17:04 UTC 2000 |
Often, leaving the top off fixes it. ;)
|
ashke
|
|
response 16 of 30:
|
Oct 10 20:32 UTC 2000 |
I had this problem, when opening certian programs, and when using the internet
on my computer. I reloaded it a few times, still did it. Went back to Win98
with IE4 instead of 5 and it solved my problems.
Microsquish sucks
|
danr
|
|
response 17 of 30:
|
Oct 10 22:10 UTC 2000 |
Back in the olden days, I was a test engineer for a floppy-disk drive
manufacturer. We had one piece of test equipment that acted up regularly. When
it did that, we popped the top, reseated the ICs, and most times it worked
like a charm after that.
|
sno
|
|
response 18 of 30:
|
Oct 11 14:13 UTC 2000 |
My cpu fan started making noises, and my system suddenly became very
unstable. Either the cpu is not cooling sufficiently, or the fan is
causing radio interference. I've removed the fan and put in some
turbo fans that plug into the ISA slot. Only one crash since then,
and that's after I put the cover back on the frame.
I suppose that once a cpu gets injured by heat, that it will remain
injured. If my problem continues, I may consider replacement.
|
scott
|
|
response 19 of 30:
|
Oct 11 14:40 UTC 2000 |
You definitely want the fan on the CPU running. Replacements are under $10
at most computer stores.
|
mcnally
|
|
response 20 of 30:
|
Oct 12 00:22 UTC 2000 |
Cooling becomes more important and more difficult the faster your
processor..
|
mdw
|
|
response 21 of 30:
|
Oct 12 02:08 UTC 2000 |
Oh, I don't know about that. The ES/9000 watercooled systems needed
some pretty impressive cooling systems to avoid melting down, but were
hardly impressive in terms of speed.
|
n8nxf
|
|
response 22 of 30:
|
Oct 12 12:35 UTC 2000 |
A resistor can dissipate impressive amounts of heat too and those are
even less impressive in terms of speed ;-)
|
gull
|
|
response 23 of 30:
|
Oct 14 18:46 UTC 2000 |
I've seen CPU's made unstable by heat that were fine once their cooling
systems were fixed. I'm not sure why this is, though I think some CPUs
automatically shut down for a while if they get too hot, to protect
themselves.
|
n8nxf
|
|
response 24 of 30:
|
Oct 16 01:53 UTC 2000 |
Transistors and such stop working right if they get too hot. If they get
way too hot, they stop working forever. Transistors and diodes can even
be calibrated to be used as temperature probes.
|