Page 1 of 2 12 LastLast
Results 1 to 10 of 14

Thread: System Monitor

  1. #1

    Default System Monitor

    Morning All,

    I've searched the forums (and the internet) for "System Monitor" and "log files" and am not getting anywhere, so maybe I'm the first person in the world to have this problem - though I doubt it! I'm having some system stability issues - random kernel panics with no log messages at all - and I was hoping to log some system parameters to a file to try and work out what the system was doing next time it freezes.

    The System Monitor seems to have a "log to file" option when you drag a sensor on to a worksheet, but there is no way of creating a new file when it comes up with the file browser. (Obviously?) if I create an empty text file for it, it doesn't use it. Does this feature work? The Help doesn't help. Am I being a bit dim?

    Ta,

    Nick

  2. #2
    goldie NNTP User

    Default Re: System Monitor

    my personal opinion is you are more likely to find the cause of your
    freezes by searching these fora for the terms

    freeze "kernel panic"

    or similar..

    then without dragging a sensor to a worksheet or digging through logs
    you would learn that the most often cause of a kernel panic is faulty
    ram..

    if you search and read enough you would probably find where someone
    would suggest you carefully remove the ram and gently clean the metal
    contacts with a pencil eraser...then boot from an openSUSE CD or DVD
    and run memtest....over night, at least..

    if all your ram passes, then it is likely some other hardware
    problem...i'd start by . . .

    well, search around..

    --
    goldie
    Give a hacker a fish and you feed him for a day.
    Teach man and you feed him for a lifetime.

  3. #3

    Default Re: System Monitor

    Well, actually some of the searches resulted in people suggesting that logging information (usually by installing some logging software) might lead to some clues as to what the computer was up to when it fell over. I'm trying to get as much information as possible - for example, if it always crashes when the PC is overloaded and the temperature is high, then that would indicate a cooling issue. Other than that, there seems to be as many solutions to system freezes as there are people who've had the problem - I'm struggling to work out which ones are worth looking into and which aren't. Examples include Beagle, Firefox, Thunderbird, compiz, graphics drivers, some network services, etc., etc.

    Beside all that, I would still like to know if the sensor logger works on System Monitor or whether it's a known bug. I can't find anything useful in Google / forums. And I would like to be able to log information to file regardless of my system stability issues.

    Was going to run memtest tonight, so great minds thing alike!

    Ta for reply.

    Nick

  4. #4
    goldie NNTP User

    Default Re: System Monitor

    > Beside all that, I would still like to know if the sensor logger works
    > on System Monitor or whether it's a known bug.


    i do not know anything about that...

    as for system logging, i my opinion, you get get much better than atop..

    that and the usual in /var/logs is pretty complete...(so complete it
    is not 'easy' to use if you don't also know about grep..)

    oh, and search/find how to install lm_sensors and running
    'sensors-detect' as root...if your bios/mb works well with those it
    will shove some heat info into some logs, somewhere (maybe 'messeges'
    i don't recall this second..

    my bottom line: i personally wouldn't be trying to solve this "what is
    causing it" problem only via GUI things... computers have been shoving
    health info into log files before the mouse was invented..

    --
    goldie
    Give a hacker a fish and you feed him for a day.
    Teach man and you feed him for a lifetime.

  5. #5

    Default Re: System Monitor

    Still, it would be nice if the logger worked, seeing it is there!
    That aside, the memory checks out ok, but I am getting a lot of SMART messages (many to do with ECC correction) on both of my disks, and the computer seems less stable during massive disc usage (for example md5deep -rl /home/nick - which contains 40GB of data). So it could be my hard drives are knackered.

  6. #6
    goldie NNTP User

    Default Re: System Monitor

    > a lot of SMART messages (many to do with ECC correction) on both of
    > my disks


    i think you should (quickly) back up your data...and THEN deal with
    the stability issues..

    it could be a heat issue..
    you might try cleaning out the cat hair and chicken bones...

    this might help:
    http://www.endpcnoise.com/cgi-bin/e/...ercooling.html

    or you might be having an overtaxed or failing power supply..

    did atop show anything?

    --
    goldie

  7. #7

    Default Re: System Monitor

    Well, the temperatures reported appear to be in Fahrenheit - the drives are cool to touch, so I don't think it's a heat issue, sadly.

    I've re-installed on a completely different hard drive on an IDE cable (as opposed to the original installation on two SATA drives), and strangely, this drive (brand new) is also reporting lots of ECC errors and heat messages. I'm pretty sure all three of my hard drives can't be broken in exactly the same way! Predictably, the reinstalled system is also horrendously unstable. It's almost impossible to run the updates without it hanging. Again, no messages at all.

    This system has been running SuSE 10.3 and Windows XP (almost - it crashed maybe once every three or four months) flawlessly for two years, so I'm finding it hard to believe it's a hardware issue. Windows XP still runs flawlessly, in fact, but it's not my cup of tea really.

    Haven't tried atop yet - trying to get the system to install anything is a bit of a nightmare at the moment!

  8. #8
    goldie NNTP User

    Default Re: System Monitor

    sounds more and more like a hardware problem..

    everyone has there own set of hardware diagnosis tricks..
    i've given mine several different places here and don't wanna retype
    them ...

    HINT: use google's site specific trick to find stuff....like this:

    site:forums.opensuse.org "hardware problem" heat "power supply" cables

    turns up a fairly good rundown of hardware problem sleuthing (though
    the symptoms [uncommanded reboot] is different from yours [kernel panic]:

    http://forums.opensuse.org/hardware/...websearch.html

    change around, add or subtract terms [freeze stall "kernel panic" hang
    PSU "power supply" heat cable shorts cracked {many more}] in the
    search string and find many other such discussions in these
    forums....leave out the "site:forums.opensuse.org" and you will find
    millions of hits on the net--including some good step-by-step decision
    matrices..

    by the way, you wrote "the memory checks out ok" and i wonder how long
    you ran memtest....overnight is just barely enough to get
    started....if you have LOTS of RAM you should let it run 24 hours, at
    least...

    many say (google it) that it is best to pull RAM and leave only one
    stick to test at a time....and, once each stick has proven itself (at
    least "overnight") then move the sticks around until you find which
    slot is messed up..

    it just takes one TINY electrical fault to mess up the whole
    machine...frayed cable and vibration, tiny crack in mother board and
    heat changes enough to open a circuit...

    on and on and on....if i were you i'd spend time tracking down the
    hardware problem..

    btw, kernel panic is most often (but not exclusively) caused by bad
    RAM (which might also mean bad RAM slot, bad circuits to/from RAM, etc
    etc etc etc)

    --
    goldie

  9. #9

    Default Re: System Monitor

    I'm pretty sure it's not a heat issue - I barely see temperatures above 30C in any of the components, and everything is cool to touch. The RAM was tested for three or four days, sadly both together as they're dual channel sticks, so taking one out isn't an option. It ran through all 2GB at least a few hundred times, so I'm suspecting that the RAM isn't the issue! And, as I said before, Windows XP runs flawlessly.

    I've just got a stable install going on a separate hard drive on IDE, and completely removed the RAID drives from the system. So far, it has crashed once before I did a system update (seems normal for SuSE - seems to do this on almost any machine I install on!). Since then it's been running fine without problems. But then it used to run for four or five days sometimes. I might try cleaning all the dust out of the machine too. Oh, and the new hard drive also reports ECC SMART messages. My other machine (which runs SuSE flawlessly) doesn't. I'm hoping the ECC messages are red herrings, or else I have a large stack of duff hard drives!

  10. #10

    Default Re: System Monitor

    Well, the power supply exploded last week. Since I've bought a new one, everything seems ok even with the RAID array plugged back in, so maybe it was just on the way out. Will try the RAID presently. However, I still haven't had anyone answer original question - does the Log to File work on System Monitor? It doesn't for me.

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •