I use Linux (Mandriva 2005LE) and It hangs "randomly". The screen "freezes" but I can move mouse on the screen (but cursor shape doesnt change). Keyboard is not working (I cannot switch to console ctrl+alt+F1 etc.) and I have to push reset button.
The question is how to find out what is wrong - some error log. (I am almost newbie in Linux)
I viewed /var/log/messages but found nothing (no error message related to the hang) then I looked at all other files and dirs in /var/log and also nothing.
How can I find out what happned and why the system hangs.


system configuration:
hw:
via a7v600
Athlon XP 1700+
geforce 3 (generic drivers)
WiFi ralink RT2500 (1.0.3b drivers - i tried 1.0 stable but system also hangs)
rtl 8139 network (integrated ethernet card disabled in bios)
hdd 40GB (ata100)

sw:
Linux Mandriva 2005LE
KDE 3.3


thanks for reply

Try "dmesg", which prints out critical messages.
Most likely this sounds like a hardware issue. Random freezes is very likely not at all related to the software. Maybe some bad RAM or overheating components.
Do you have another OS installed on the computer? Does it freeze up as well?

i tried dmesg but found nothing strange.
- I ran memtest86 (on boot CD Mandriva2005) and RAM is OK
- I have WindowsXP installed too. It shown bluescreen with fatal error but only when firewall (kerio or adorons) was installed.
I think it could be something with wifi card but its strange that without external firewall installed it never hangs (win XP). And in fact I dont know about settings of this win because that computer is not mine but my brothers and I didint install win XP and set it up (I know that brother has changed some things in win but I dont know what and how )
And in fact I have Mandriva 2005 on my notebook too and it hang the same way but only once or twice. And so I suppose that the problem is probably in Linux and I want to find whats wrong ...

Actually, I agree that it's most probably a hardware problem. memtest coming back OK doesn't really mean much. To find out if it really is a software or a hardware issue, try compiling a kernel.

gcc + linux kernel source is about the least buggy combination you can have. If it fails, it's almost certainly because the hardware has a problem and not because the software is faulty.

Compile a kernel (you don't have to keep or use it afterwards) once or twice, see what happens, and let us know.

Something to consider: Linux tends to use all memory pretty much 'at random, while Windows apps will often use the same portion of memory each time they run. This is why some apps always seem to crash a Windows that runs fine the rest of the time, while Linux crashes randomly: Linux will address the bad RAM at any time, while Windows is more likely to only address it at predictable intervals.

Same thing was happening to me with my rt2500 on Mandriva LE, however I installed the latest driver from http://rt2x00.serialmonkey.com/wiki/index.php/Main_Page and this solved the problem.

Now, however, the connection just randomly drops. Now I'm working at sorting that out.

This is in relation to your rt2500 card I mean.

It hanged (or ?hang? not sure with this irregular word) again (when setting up the kernel before compilation ). This time (when hanged) I connected it to other PC (via LAN) and I examined that at least some processes must had been running (SAMBA server - I conected and connection was working ...) so it seems that only screen (Xorg ? KDE ?) and keyboard stops (but mouse works).
I also have to mention that install process of madriva on my notebook hanged the same way several times but it had never happened with other distro (Federa2) or older Mandrake and these didint hang at all (never).
I still think that the problem could be somewhere in Mandriva2005 ... maybe. Or Wifi.

- I have the latest rt2500 driver already intstalled (rt2500 driver supplied with kernel does not support MAC changig and I need it).

thanks guys for your patience

I really am confused. It "hanged" again. I connected to the "hanged" computer with ssh (remote access). I did dmesg and here is result:



oom-killer: gfp_mask=0x1d2
DMA per-cpu:
cpu 0 hot: low 2, high 6, batch 1
cpu 0 cold: low 0, high 2, batch 1
Normal per-cpu:

cpu 0 hot: low 28, high 84, batch 14
cpu 0 cold: low 0, high 28, batch 14
HighMem per-cpu: empty

Free pages: 3512kB (0kB HighMem)
Active:56070 inactive:972 dirty:0 writeback:0 unstable:0 free:878 slab:3425 mapped:55262 pagetables:1058
DMA free:1080kB min:124kB low:152kB high:184kB active:10960kB inactive:20kB present:16384kB pages_scanned:11080 all_unreclaimable? yes
lowmem_reserve[]: 0 239 239
Normal free:2432kB min:1916kB low:2392kB high:2872kB active:213320kB inactive:3868kB present:245696kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0
DMA: 0*4kB 1*8kB 1*16kB 1*32kB 0*64kB 0*128kB 0*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1080kB
Normal: 124*4kB 6*8kB 4*16kB 1*32kB 0*64kB 0*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2432kB
HighMem: empty
Swap cache: add 0, delete 0, find 0/0, race 0+0
Free swap = 0kB
Total swap = 0kB
Out of Memory: Killed process 7135 (wine-preloader).



Then I killed wine and "frozen" screen melt up and computer is working again (without restart).
But wine was running the first time - so "hangs" before could not had been caused by wine.

Other thing is that /var/log/messages contains similar kernel error only for about two hangs of about 15 (other are without any error). So problem is elsewhere.

This is unfortunately a bit of a (un)known issue in Linux. It seems to be down to X locking up (which will also lock the keyboard) but no one has a satisfactory answer as yet.

There's a lengthy thread over at NvNews where it was thought to be down to Nvidia drivers. However, the same issue has been reported on ATI cards. There are a few workarounds knocking around (personally i have to reduce my AGP 8x down to 4x). Unfortunately it's mostly a mystery

Just got the same oom message from my Athlon system with a Matrox G450 video card, so it appears that the problem might affect most if not all video cards. I do have it enabled to AGP 4x, so maybe that's the problem. I've only had this happen once and I was running KDE. Haven't had it happen with GNOME.

I fixed this issue on my own box (with a GeForce FX 5200) by changing xorg.conf to use the nVidia AGP driver instead of the kernel driver (AGPGART). This was by changing/adding the line: to the video card "Device" section in xorg.conf. See the first entry of Appendix D of the nVidia README.txt file for details.

AGPGART must be either be disabled or an unloaded module for this to work.

I solved this problem by installing latest nvidia display drivers