New Member - Random Reboots - Help?????

[deniro0311]

I am having the dreaded random reboot problem. I am an IT guy, so I am not a complete noob when it comes to computers, just partial noob. We were in a pinch at my restaurant and needed a computer, fast and cheap. I went to ebay and found what I thought was a good one. I know, I am a jackass for going to ebay for a computer, it won't happen again. The computer is an older dell dimension 4600. Pretty standard setup, nothing fancy, it has an older nvidia card, 512 system mem, one hard drive, two opticals, and one floppy. I would be best served to just get another computer, but I have to fix it. I wouldn't be able to live with myself if I quit.

The reboots happen very randomly. Sometimes its days without one, other times it happens over and over. Example, I was transfering files from an external hard drive to the computers hard drive, and I had two reboots in a short time span. I get no BSOD and nothing in the logs.

Now I will explain what I have done to try and fix this problem (no particular order).
1. Reinstalled windows xp pro sp2
2. Ran all virus, adware, etc. scans
3. Ran memtest over night - No errors
4. Used some program to check hard drive temp. Not sure if it was any good, but the temps it showed were fine.
5. Got a new PSU
6. Downloaded and installed new drivers
7. Checked all plugs and reseated all cards, modules, etc.
8. I made sure "automatically restart" in the startup and recovery was not checked
9. Reduced the agp memory in bios
10. Disabled hyperthreading in bios
11. Made sure the cpu fan is running
I am sure I have done more, but I can't recall at this time. I am out of ideas. Is it possible that I need more memory. It doesn't make sense that 512 isn't enough, but I can't think of anything else. Correction, the only other thing I can think of is overheating. My next step will be to remove the cpu/heatsink and apply some arctic silver I have laying around. Is there any programs that will test temps on an older Dell. Every program I have tried says it is not compatible with the mobo. Any help would be much appreciated.

 

[jonnyGURU]

Two possibilities:

1. Bad stick of RAM. And Memtest won't expse this. The RAM has to be hot and the faulty address hit just at the right time.

2. Week rail on the PSU. All it takes is for a voltage to drop out of spec for a split second and it reboots.

Of course, worst case scenario is just a bad board, but we won't go there.

 

[deniro0311]

 

[deniro0311]

So what I am understanding is that memtest is not the end all be all for checking ram. I hope this is true, because I have extra ram sitting around. As far as the PSU suggestion, I have tried two PSUs with the same results. And no, we won't go there. As a matter of fact, I have been praying that its not a bad mobo. That is the absolute worst case. Good to see I am on the same page with somebody. In a few hours I will try the following....change the ram, apply arctic silver to cpu, find a utility to test my temps. Thanks for the help johnny. Almost forgot, a little while ago I looked in the logs in the Bios and had a few messages (i think it said warning) about memory size decreased or changed. I can't remember exactly what it said. I will have to wait to get back to the office to see what it was.

 

[oynaz]

You have a piece of faulty hardware, IMO, the order of probability is:

RAM
Mobo
PSU
CPU
GPU

 

[Seekermeister]

I realize that you have completely different hardware, but the behavior that you described suggested to me that this might be related to a problem that I described here:

http://forums.anandtech.com/messageview...atid=27&threadid=1971428&enterthread=y

I might be totally wrong, but it wouldn't hurt to check.

 

[Jeepfreek]

I just went through this same scenario and mine ended up being the processor. I had swapped everything except the processor and it was still happening. Including a new motherboard and new ram (smacks head). I was literally thinking of changing the case when it hit me I hadn't changed the processor. Surething that was the culprit. It sure sounds like a hardware issue. Just have to narrow it down.

 

[deniro0311]

 

[deniro0311]

A little update. Its not the PSU. I replaced it like I said earlier. It's not the ram. I pulled some ram out of another system, and it still rebooted randomly again. I am running chkdsk right now to rule that out. I won't know the results until tomorrow, when I get back to the office. There is one difference now with the reboots. After I put in new ram I actually got feedback from the computer.

During the boot I got a message telling me that the computer just shut down due to a problem, etc. etc.. Of course it was vague. It gave me multiple options for what I could do. I chose to boot into windows as usual, and it sent me back to the message screen. Then I chose "last known good.." and it booted fine. I wonder why I am getting that now, and I am still not getting anything in the logs. Maybe I am not looking in the right area. I believe I am, but I have been living mostly in a Linux world for a while. Its almost time to scrap this Dell and salvage the few parts I know are good. Then I will use them for a project that I have always wanted try....pegboard computer.