How can I determine what is causing my system to hang?
Linux - HardwareThis forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
How can I determine what is causing my system to hang?
I have system (Intel Celeron 600MHz, 256MB RAM, 40GB WD EIDE HD, Asus CUV4X-V motherboard, ATI Radeon 9000 AGP video (motherboard onboard video disabled), Lite-On LTR-12101B CD-RW, SMC EZ Card 1211TX 10/100 ethernet).
I've installed FC2 and Mandrake 10. Both of these installed fine and detected all hardware with no problems. However, shortly after bringing up Gnome and opening Open Office, or a game, or surfing the web, or dragging a window, the system freezes and I'm forced to power off to recover. I've tested the harddrive, the memory and the CPU. How can I determine what is causing this? I've ran a burn-in diagnostic for over 8 hours and the system did not hang. I have not tried any other O/S to see if it would behave differently.
Well, sounds like you've already eliminated a few possibilities. It's not the CPU, RAM or hard disk. I'd say it's probably safe to eliminate the CD-RW too since you used it to install and weren't using it when the machine locked up. That leaves us with the motherboard chipset, the radeon and the SMC NIC. Fedora Core 2 and Mandrake 10 both use the 2.6 kernel, so reverting to a non-dev kernel version like 2.4.22 would be a good way to eliminate the possibility of a chipset incompatibility. Transfering a bunch of files from the CLI to another machine on the LAN would be a good way to eliminate the NIC. At which point I'd have to say it's likely an X driver problem with your Radeon.
edited to add: now that I think about it some more, the first step should probably be to switch to runlevel 3 and try to isolate the problem to X (the most likely suspect.) If the problem goes away there's no reason to try a different kernel. If the problem persists, then continue eliminating the possibilities. Testing the NIC before changing kernels is probably a good idea too, just because it's easier/faster to do. I'm pretty sure Mandrake 10 comes with both kernels, not sure about FC2. GL
Distribution: Linux Redhat 9.0, Fedora Core 2,Debian 3.0, Win 2K, Win95, Win98, WinXp Pro
You didn't say in your post whether you have checked any of the logs (messages, secure, etc....). I had a very similar problem but could not find anything going wrong. No log entries stating anything went awry. But my machine (running RH9 and acting as a file server) would just freeze. I had even gone so far as to low level format all the scsi drives (six of 'em) and reloaded. Still didn't do it. What I have ended up finding was that the cpu cooling fan was going out and as the machine would heat up, the cpu would just lock up from heat. So you might try checking the other things out also. It seemed rather intermittent so before I pulled the case apart and found it, I was pulling out my hair. Your problem definitely sounds as if it is hardware related. You might even check to see if there is an IRQ conflict (possibly shared) or something of that magnitude. Good luck and re-post when you find something.
As I mentioned in the HCL, my ASUS P3B-F board did the same thing. The cause was a mismatched voltage spike of some sort to the CPU. I solved the problem by soldering a capacitor between a1-a3 pins on the underside of the board. The CUBX had the same issue. perhaps that's it for you. or, it could be in need of a BIOS update. I have a suspicion that its something on your mobo that's causing it.
Originally posted by vectordrake As I mentioned in the HCL, my ASUS P3B-F board did the same thing. The cause was a mismatched voltage spike of some sort to the CPU. I solved the problem by soldering a capacitor between a1-a3 pins on the underside of the board. The CUBX had the same issue. perhaps that's it for you. or, it could be in need of a BIOS update. I have a suspicion that its something on your mobo that's causing it.
This may have been it. I did some searching and found a BIOS update for my motherboard (beta, which scared me, but figured it was worth a shot). I installed the BIOS update, and so far, the computer is running without a lockup on Mandrake 10 (I'm posting this from it). I'm going to let it run all night and see how it goes.
BIOS ipdates are always a cause for a scare. Note to those who have never done one......read the instructions and do them exactly! You can fry your chip if you don't do it right. If its done right, it only takes a minute. if it screws up, NEVER TURN OFF YOUR MACHINE!
Glad that it seems stable. Good luck with the overnighter!
Wondering what program you used for the burn-in diag. Most are pretty good at testing the CPU anyway. If the BIOS upgrade fixed the problem it was probably a bus timing issue with your AGP video or PCI NIC card.
Yeah, it seems that ASUS has a habit of sometimes rushing a board out the door to catch up to Abit's latest offering. perhaps they should do what Mandrake is trying to do and have a longer QA cycle on the BIOS