Server crash

[mattdcknsn]

Well my server crashed tonight. I am still trying to figure out what the heck happened.

Here's the background. Everything has been working fine for 9 months now. No major problems, recording fine, playback fine. Today my wife says that she cant browse the videos (Running SageMC). So after the kids go to bed I take a downtime and all I do is initiate a reboot. I let it sit for a good 10 minutes and then try the HD-100. No server out there. Do a ping test, responds fine. Try TS (RDP) and cant connect. Go downstairs and check the server. It looks like it hung during shutdown. I didnt think much of it, so I power it off at the power button, wait a good 5 minutes then power it back on. This is where things go wrong.

The server POSTs, I hear the single beep and then the screen goes to the POST readout. It hangs at the processor. I try to power off the server with the power switch on the back of the computer. Wait a while, turn it back on, same thing. I tried to remove a couple of PCI cards, both 150s, Firewire card, all with the same results. I tried removing and moving the memory DIMMs, same results.

At this time I am thinking something happened with the BIOS. So I clear the BIOS (with the computer unplugged) and power it back on. Here is where it gets odd. Now the computer POSTs like before but I get one step further. Now the computer hangs at the memory stage. It reads the correct amount of memory, it just doesn't go past there.

I then try to unplug all the HDs and give it a shot, same thing.

Hardware specs:
Mobo: Asus M2N-E
Proc: AMD 64 X2 5000 Black edition. I overclocked it a small bit (no voltage increase just multipliers).
RAM: Munchkin 1 Gig sticks (x4) They are matched, 2 2 GB kits
PCI slots: 2 PVR 150s, Firewire card (don't know the name off the top of my head)
Video: EVGA 7200 outputting to basic monitor
I do have USB devices attached, the ones I remember are HD-PVR, UIRT, a couple of USB Hubs, 2 USB hard drives, 1 4 bay USB enclosure and a memory stick.
Power Supply: 700 Watt, I don't remember the name brand.


I have come to the conclusion that its probably the Mobo. I have not tried replacing the power supply, worth a shot. I'll try that tommorow.

I am hoping that somebody here can help me out and figure out what I am missing. Sorry for the long post, trying to organize my thoughts on this. I'm just hoping my backups worked last night. Thanks for all the help.


edit: one other thing I tried, I took out the proc and reseated it. I cleaned it off and reapplied thermal grease. Same results.

[davephan]

You might try powering down the server, and unplug all power for a couple minutes, then plug in the power and power up the server. Sometimes removing power in addition to powering down the server can make a difference - I've seen it many times in our server enterprise at work.

How far does it get in the POST? If you get up to 'no operating system', do you have an image of your boot drive?

Dave

[Conejo]

Sounds like he can't even get past POST and into the BIOS setup.
I'd try doing a 30 second clear of the BIOS, and check the battery voltage.

...The lengths some people will go to in order to justify buying a faster server.

[autoboy]

Take everything to the bare minimum. One stick, no drives, processor at stock or below multiplier and graphics. If you have PCI graphics backup try that too. If nothing, then it should be a bad mobo.

[mattdcknsn]

Thanks for the suggestions. I'll unplug everything today before I head out to my parents house. That should give it a good 5 hours of no juice.

I can't get into the BIOS, it hangs at the memory reading now. If I hit <del> fast enough it says that it's going into BIOS but never gets there.

Again thanks for the suggestions, I'll try them today.

[mattdcknsn]

Ok, figured out what was wrong. The main problem was between the chair and the keyboard.

I "thought" I unplugged all my devices except my keyboard. Until I looked closer (and with a night's sleep) and noticed a USB hub was still attached. So I unplugged that and presto everything worked. Apparently the USB hub went bad and was really messing up the system. Now I just have to reconfigure everything. But at least its working.

Thanks for everybody's suggestions.

[Conejo]

Oh yea. Been there. I've torn a couple of machines apart, only to find that nothing's wrong and that it was a USB device holding the 5V line low and preventing the motherboard from booting. I guess you were lucky it wasn't a USB keyboard doing it.