Motherboard (at the very least) is likely dead. Looking for any insights and if it could be the PSU or CPU as well.

Cyber Akuma

Limp Gawd
Joined
Jan 3, 2009
Messages
447
This is my system when I built it back in 2012, there have been a few upgrades (mostly the GPUs being replaced with a single GTX 1070, the SSDs being upgraded, and a scratch drive being added) but most of the major hardware is still the same:

https://pcpartpicker.com/user/Cyber_Akuma/saved/BBvV3C

So I shut down my PC that is normally on 24/7 to swap out a drive. I didn't even open it up to do so, the drive was in an external bay, but when I tried to power it back on I would get lights and fans, but most USB devices (such as keyboard and mouse) were not getting power, there was no display on the screen, and no POST beep. Also, half the time it would power on for about 10 seconds, then shut off, then power on again, sometimes stay on, sometimes keep looping.

I stripped it down to it's essentials, removed the CMOS battery for a while, and when I tried again now USB devices were getting power, but that was the only change. I did notice however that I was getting a post beep and the POST code was "A0" which apparently means that POST finished and it sent the boot over to the OS... or would have if any drives were connected. But still no display.

I decided to try plugging in my GTX just to see, and now I got a display. But it was acting very funny after that. Sometimes the system would screen at the POST logo, not even Ctrl+Alt+Del would respond. Many times when it froze like that the CMOS settings would reset to default. This would happen almost every time I had any sort of SATA drive connected, even just the optical drive. I put memtest86+ on a usb drive and ran it for 24 hours, no errors detected.

The next day I tried booting off some spare HDDs I had, even off an optical disk, but it would hang on post every time with an optical drive connected, usually giving me Post code A2.... which apparently means either a Boot or GPU error. I tried running a Linux off a USB drive to test a few things but was ultimately unsuccessful as I could not get Linux to recognize the wiFi and I was too far to connect it by Ethernet, I just powered it off for the day.

The next day when I turned it on again the BIOS gave me the warnings "CPU has changed" "Case intrusion detected" and "Critical Error - System Halted"...... which considering I never touched the CPU at all and there is no form of case intrusion detection installed on the system whatsoever was not a good sign.....

At this point, I am guessing at LEAST the motherboard is badly damaged, if not also the PSU, and I have no idea about the CPU (about the only thing I can confirm is good is the RAM). But I have no idea how to further test what parts are broken or where to go from here. Any help on if any other parts could be broken, if somehow it's not the motherboard, or what else I can try or test would be helpful. I tried shopping around to see if I can find that exact same motherboard used, but the cheapest I could find it was for about $250...... from China, or $175 from Russia with $500 shipping. Needless to say, that's not an option.

Additionally, the reason I was trying so much to see if I can get it to boot properly and perform stress tests on the psu/ram was to see if there was a risk of failure if I tried to get a backup of my drives off of it. I originally wanted to try to get it in a booting properly from usb or optical drive state WITH the drives connected to try to create a backup image of them. Since the OS drive is a RAID0 that was run by the motherboard, AFAIK I can't just plug them into any sort of external device and dump an image. I do have a backup from July, which would probably be about 98% of my data, 99% with some cloud data backed up, but it would be nice if I could image the current state of the drive when this all went south. My backup of the HDDs which are in RAID5 is much older and more important.... but since those were controlled by a discrete LSI card that is pretty easy and cheap to get right now, my chances of being able to recover that when I have a system working again is pretty high..... provided two or more of the the HDDs didn't get damaged somehow. I was told that I would need to have at least the same or better Intel Chipset to try to recover the RAID0, but I don't know how true that is, and that even if it is true, how far I can deviate before it's not compatible anymore. (for the prices some of these used boards are going for, I could probably build an i7-9000 era system.... but then I worry the chipset that new can't remount the RAID).

But mostly I am looking for advice on what else I should try, check for, or what else might be broken so I can start to create a plan on what parts to buy. At this point the motherboard at the very least seems beyond recovery.
 

bwang

Gawd
Joined
Aug 6, 2011
Messages
1,006
With a decade-old system and such strange errors my first point of suspicion would be either the PSU or the CMOS battery. It's almost never the CPU unless you've had a *really* bad overclocking accident.
I feel like the fact that it boots Linux successfully means the board is not dead.
 

Cyber Akuma

Limp Gawd
Joined
Jan 3, 2009
Messages
447
The CMOS battery is definitely fine, I removed it to reset the cmos and used a battery tester while it was out, was still at nearly full power.
 

Cyber Akuma

Limp Gawd
Joined
Jan 3, 2009
Messages
447
Yeah, if I had the money I would have bought a single larger SSD and just imaged them to it, it was from 2012 and the setup just stayed.

I am considering it. I can't find my exact same board for anything even close to reasonable, and even the one below which would be a downgrade for me is pretty pricy. Might be best to get a cheap but still same chipset/series board just to try to restore the raid and then get something more modern... but that would also likely not let me migrate my Windows install and likely invalidate my OEM license.
 

Cyber Akuma

Limp Gawd
Joined
Jan 3, 2009
Messages
447
May just import into a new one fine. That’s mostly software
Yeah, I had someone else tell me that too./ Worth a shot, if not, I was also recommended dmraid but I am very unfamiliar with Linux. I have used it before many times, but not that good with it.
 

lopoetve

Fully [H]
Joined
Oct 11, 2001
Messages
30,956
Yeah, I had someone else tell me that too./ Worth a shot, if not, I was also recommended dmraid but I am very unfamiliar with Linux. I have used it before many times, but not that good with it.
Intel raid would have done it's own "thing" to it - but if it was just a RAID1, you should be able to get it back at the very least. May not boot, but you can get the data off.
 

GiGaBiTe

[H]ard|Gawd
Joined
Apr 26, 2013
Messages
1,500
Yeah, if I had the money I would have bought a single larger SSD and just imaged them to it, it was from 2012 and the setup just stayed.

I am considering it. I can't find my exact same board for anything even close to reasonable, and even the one below which would be a downgrade for me is pretty pricy. Might be best to get a cheap but still same chipset/series board just to try to restore the raid and then get something more modern... but that would also likely not let me migrate my Windows install and likely invalidate my OEM license.

If you still need a board to revive your rig, I have a P8Z77-V LK. It's a few ticks down from yours, but it very close. The only major missing things I can see are the second LAN, some USB ports and two less SATA ports. I think it has an i3-3220 in it, it's yours for $60. I'll test it before I ship it out since it has been sitting awhile unused. If you need an extra heatsink or RAM, let me know and I can toss something in with it.
 
Top