Hello guys, I’m quite new to Linux so this thread might lack information you’d like to see, so I’d be quite happy if you just ask.
I’m running openSuse 10.3 on a 500MHZ 128MB PC.
The sole purpose of this computer is to control Network Traffic with NTOP.
The machine has run fine for 3 months without any problems, nor any reboots or something.
Well but since one week, the computer reboots 1 to 8 hours after NTOP is started and I really have no clue why. My limited Linux knowledge isnt a great help there either.
I checked the System Log but I cannot see anything suspicious there. 10 of 12GB HDD space is free.
The Computer runs fine, just to see if it could be some hardware issue I used it for several hours and there was no reboot at all, its just when I startup NTOP.
Is anyone able to help me? All I know is checking SystemLog but I(may be wrong) cannot see anything suspicious there.
>
> Hello guys, I’m quite new to Linux so this thread might lack
> information you’d like to see, so I’d be quite happy if you just ask.
>
> I’m running openSuse 10.3 on a 500MHZ 128MB PC.
> The sole purpose of this computer is to control Network Traffic with
> NTOP.
>
> The machine has run fine for 3 months without any problems, nor any
> reboots or something.
> Well but since one week, the computer reboots 1 to 8 hours after NTOP
> is started and I really have no clue why. My limited Linux knowledge
> isnt a great help there either.
>
> I checked the System Log but I cannot see anything suspicious there. 10
> of 12GB HDD space is free.
>
> The Computer runs fine, just to see if it could be some hardware issue
> I used it for several hours and there was no reboot at all, its just
> when I startup NTOP.
>
> Is anyone able to help me? All I know is checking SystemLog but I(may
> be wrong) cannot see anything suspicious there.
>
> thanks in advance
>
>
Initial question would be: Is it rebooting properly? A controlled
shutdown, logged and everything? Or it spontaneously dying, and restarting
on it’s own?
If it’s been running nicely for 3 months or more, does it now have a heat
problem? Dust bunnies reproduce at an amazing rate, and can clog cpu
cooling fans and heat sink fins, power supply fans… open it up and clean
it out. (be careful, static electricity can ruin your day)
ntop can use quite large amounts of ram, and over time its needs grow as it
caches much of the data it collects. Is it possible to upgrade the ram
from 128MB to somethine more? I know it shouldn’t happen, but out-of-ram
issues can cause the kernel to (almost) randomly kill processes… which
could eventually reboot or lock a machine.
ntop is a pretty cool application… used it a few times in an office
environment… it can show you things you really didn’t want to know.
Thanks for your reply and oh well, stupid me, could’ve thought myself about adding that info, sorry.
To answer your question:
It’s simply rebooting, no shutdown or anything at all!
Heat shouldnt be a problem, cause the environment is really cool. Gonna try it though!
Ram should be no problem(well it ran for 3 months, should be enough for >8 hours then :P) and upgrading would be no option, as I’ve choosen that “bad” config to save electricity.
I got to reverse my statement too. I’ve just been trying around reading logs etc when it suddenly went POOF reboot.
So it’s not NTOP at all. Could be Hardware aswell(most likely it seems)
>
> Hi Loni,
>
> Thanks for your reply and oh well, stupid me, could’ve thought myself
> about adding that info, sorry.
>
> To answer your question:
> It’s simply rebooting, no shutdown or anything at all!
>
> Heat shouldnt be a problem, cause the environment is really cool. Gonna
> try it though!
>
> Ram should be no problem(well it ran for 3 months, should be enough for
>>8 hours then :P) and upgrading would be no option, as I’ve choosen that
> “bad” config to save electricity.
>
> I got to reverse my statement too. I’ve just been trying around reading
> logs etc when it suddenly went POOF reboot.
>
> So it’s not NTOP at all. Could be Hardware aswell(most likely it seems)
>
>
>
Well, with it 'poof’ing on you… it’s not ntop!! {Grin}
Still leaning towards heat issues. But could also be bad ram. If you have
the time and ability to take machine down for a little while, you might try
running memtest from an install cd (various sources online, I use ‘INSERT’
rescue cd quite often)
If the machine’s bios give you access to the temperatures and voltages, I’d
examine them as quick as possible after a reboot… see what the system and
CPU temps are, you can quickly rule out heat issues that way.
Kinda seems like bad ram the more I think about what you’ve posted. Heat
issues tend to cause a machine to die, and stay dead until manually
restarted… ram issues can cause random reboots easily.
And (hopefully), reading logs isn’t something that would raise cpu temps to
a dangerous level… (but you never know!! {Grin})
> So it’s not NTOP at all. Could be Hardware aswell(most likely it seems)
>
well, don’t throw it out yet…blow out the crud, check the cpu fan…and the
heat sink…is it clean?? might need to remove, clean and then reinstall the cpu
heat sink with NEW thermal paste…
may have a failing power supply, leaking capacitors on the MB…an overheating
hard drive, etc etc etc…hard to tell from here…
might be best (easier, cheaper, faster, longer lasting) to find a newer model
some ‘power user’ is junking to ‘upgrade’ to vista…
–
DenverD (Linux Counter 282315)
A Texan in Denmark
500mhz… i’d say this system is pretty old. It would definatly be a hardware problem, something is overheating and causing the shutdown. you really will want to consider the age of the motherboard and power supply, ram maybe too. depends on the age of any or all of these components.
Just to add a tad to the posts above . . . re power regulation, the voltage regulators are the MOSFETs, inline capacitors adjacent to the cpu. Take a close look with a flashlight and magnifying glass, for any sign of bulging or any material that looks like crystallized liquid. If you see such, it’s about to blow. On the power supply, look for excessive heat or you can put a meter on it (they’re pretty cheap). If you suspect it’s what is failing, please don’t try to open it up; you could get hurt. On some systems, you can monitor voltage to the cpu, ram, and northbridge in the bios relative to what it should be; in Windows you can get that with cpu-z or Cbid, in Linux lmsensors may give you that as well.
Finally re heat - always the first suspect - ambient temp is not an indicator of heat in the cpu socket or on the ram sticks. You can get readings on the former, but not the latter - you check ram by manually switching out sticks or by stressing it with memtest, more so with Prime or in Windows with Orthos blend.
