Question WHEA Uncorrectable Error and BootMGR is missing

[TLindhard]

Hi, I have over the last 6 months had trouble with constant crashes and being unable to boot, and I have so far failed to identify the cause. My setup is:

GPU: AMD Radeon RX 5700
CPU: AMD Ryzen 5 3600 3.6 GHz with stock cooler
Boot drive: Samsung EVO Plus 970 500GB (previously Samsung EVO 970 500GB)
Mobo: MSI Tomahawk B450
RAM: Corsair Vengeance lpx 2x 8GB DDR4 3000MHz and Corsair Vengeance lpx 1x 16 GB DDR4 3000 MHz
PSU: Corsair CX600 600 W
OS: Windows 10

Aside from the SSD (2 months old) and the PSU (~10 years old) everything is 2 years old.

I have one older SSD and several ancient HDDS connected (but not the boot drives) and a CD drive. During some of my troubleshooting I have disconnected all but the boot drive (only with the old SSD, I have not tried after replacing it, but the error still occurred).

The issues:
Constant BSODs (about 20-30 min intervals, but sometimes as soon as Windows is booted. It seems to be consistently triggered faster under heavy load such as gaming, but not sure how correlated it is). This is often followed by a "BootMGR is missing", which after a few restarts goes away and Windows boots normally again until the next BSOD. However sometimes the BootMGR is harder to get rid of, taking multiple days before magically booting normally again. The BSODs are typically very quick, but I have seen the error code WHEA_Uncorrectable_Error. Moreover, in the log it shows "Windows kernel power error 41, bugcheckcode 292" (sometimes 0, but that seems to be when I restarted it myself)
Moreover, I have noticed fairly high temperatures on my CPU: about 60-70 idle and 80-90+ under heavy load.

Issues I've only encountered on my old SSD:
"Windows has encountered a problem communicating with a device connected to your computer", error code 0xc00000e9
Cleaning the disk manually using diskpart resulted in "a device which does not exist was specified"
BSOD with error: "winload.efi-file missing" (error code: 0x000000f)
And not being able to see the SSD in BIOS at times or the disk being inaccessible by Windows Repair

After replacing SSD everything was fine the first month, but now the above issues (the ones not related to the old SSD) have resurfaced.


Attempted solutions:
1) Update all drivers and all updates from Windows. My GPU driver gave me a lot of trouble, as AMD Adrenaline disappears from my computer as soon as I update it. However, with other tools such as Driver Booster I have checked that it indeed has installed the newest driver (currently Adrenalin 23.7.1, but I have attempted other versions as well to fix some other problems).
2) Checked boot order
3) Ran Windows Startup Repair (with my old SSD this did not complete)
4) Reinstalled Windows
5) Replaced my boot SSD from 970 EVO to 970 EVO Plus
6) Ran tests on RAM with no issues found
7) Checked components are seated properly and reseated everything and checked all cables.
8) Disabled precision boost on CPU -- this lowered temperature of CPU significantly
9) Tried the following commands in CMD: bootrec /fixmbr, bootrec /fixboot, bootrec /rebuildbcd

Currently my computer is stuck with the BootMGR problem, so I cannot retrieve a dump of the error codes. I will update if I get access to it again.

I have no idea how these different issues are related, but hope you can help me out!

 

[Aeacus]

It seems to be consistently triggered faster under heavy load such as gaming, but not sure how correlated it is).

Very much so.

You look at your issue as being software, while it very much can be hardware. Namely your CX-series PSU is both old and not enough build quality for your PC.
Corasir CX series, at best, is only good enough for office build, without dedicated GPU. But for any build with dedicated GPU, especially gaming builds, far better build quality PSU should be used.

So, i suggest that you replace your PSU with good quality unit. E.g Seasonic Focus/PRIME or Corsair RMx/RMi/HX/HXi/AX/AXi.

Though, it is possible that your old CX damaged your hardware (hence why you see your issues) and new PSU does not "fix" the damage already done. But at least, when you go forward with component replacement, new hardware won't be damaged anymore.

 

[TLindhard]

Very much so.

You look at your issue as being software, while it very much can be hardware. Namely your CX-series PSU is both old and not enough build quality for your PC.
Corasir CX series, at best, is only good enough for office build, without dedicated GPU. But for any build with dedicated GPU, especially gaming builds, far better build quality PSU should be used.

So, i suggest that you replace your PSU with good quality unit. E.g Seasonic Focus/PRIME or Corsair RMx/RMi/HX/HXi/AX/AXi.

Though, it is possible that your old CX damaged your hardware (hence why you see your issues) and new PSU does not "fix" the damage already done. But at least, when you go forward with component replacement, new hardware won't be damaged anymore.

Thank you very much for the reply! Would these issues appear at all if only my PSU was bad, or does this very much indicate it has done damage on the other components? Also, I happen to have a completely identical PSU lying around. Would I be able to test anything using that? Seeing as my PSU have served me well for 10 years and in this specific build for 2 years, a properly functioning model should be able to run properly for a while, assuming no components have been damaged?

 

[Aeacus]

Would these issues appear at all if only my PSU was bad

BSoD/shut down during heavy output (gaming) is tall tale sign of either of the two:
* high CPU/GPU temps
* poor PSU

Your temps, while high, aren't quite at the level of thermal throttle (for CPU ~100C, for GPU ~80C), after which PC would shut itself down, to keep components burning up. But since you have a poor and old PSU in use, i lean more towards PSU issue.

BootMRG missing is either drive issue or MoBo issue. Since you have good quality Samsung drives in use and you've also replaced it, it isn't drive issue. That leaves MoBo issue. And as far as why MoBo produces this issue, can be down to the out of spec voltages fed to it, over the past years, by inferior PSU.

In similar example: if you eat contaminated food, you won't drop dead on an instant. Instead you get food poisoning, resulting in stomach aches and vomiting. Eat contaminated food on daily basis and you'll end up 6 feet under. Essentially the same is with your MoBo and other hardware, which has been fed the "bad" power by your poor PSU.
How far the damage has developed to - hard to tell. MoBo is usually 1st one to go when used with iffy PSU. GPU is next in line. CPUs are robust but not invulnerable either.

Also, I happen to have a completely identical PSU lying around.

How old?
Then again, it doesn't matter.

a properly functioning model

Like i said, Corsair CX, at best, is mediocre quality PSU and unless you test with top-of-the-line PSU (e.g Seasonic PRIME TX, Corsair AXi or Super Flower Leadex Titanium), you can not rule out PSU issue.

Moreover, how can you tell that the 2nd CX also functions properly? Do you have high-end, very expensive PSU testing equipment, which you can use, to make sure that 2nd CX works within the specifications of ATX PSU standard?

Seeing as my PSU have served me well for 10 years and in this specific build for 2 years

I wouldn't call a PSU being good (or served well), which in the past 6 months, has damaged/degraded your hardware up to the point of actual hardware failure, that makes PC inoperable. Sure, it powers it on, but what good is this power, when it damages your hardware?

For example, in my main build, Skylake (full specs in my sig), i have Seasonic PRIME TX-650, which i bought back in 2016 and which now, 7+ years later, has produced 0 issues. No hardware degradation, damage or outright death. Nada. Nothing. <- This is to be expected when buying the best 650W PSU money can buy. Which also has the best efficiency there is (80+ Titanium), mythical levels of voltage regulation, very tight ripple. Double + then some of hold up time, over what ATX PSU standard specifies (standard is at least 16ms, PRIME TX-650 has it at 35ms), very quiet operation, 12 years of warranty, etc.

Your CX600 unit is 80+ Bronze with 3 years of warranty. For a 10 year old unit, your CX is way past it's time. And it wasn't good enough even when it was brand new.

So, you trying with another CX unit, regardless the age (CX600 platform was released in 2013) wouldn't be a good idea. Better get new, up-to-the-date, good quality PSU and test with that, while crossing your fingers that the rest of your hardware isn't damaged (i won't get the hopes up though).

 

[TLindhard]

BSoD/shut down during heavy output (gaming) is tall tale sign of either of the two:
* high CPU/GPU temps
* poor PSU

Your temps, while high, aren't quite at the level of thermal throttle (for CPU ~100C, for GPU ~80C), after which PC would shut itself down, to keep components burning up. But since you have a poor and old PSU in use, i lean more towards PSU issue.

BootMRG missing is either drive issue or MoBo issue. Since you have good quality Samsung drives in use and you've also replaced it, it isn't drive issue. That leaves MoBo issue. And as far as why MoBo produces this issue, can be down to the out of spec voltages fed to it, over the past years, by inferior PSU.

In similar example: if you eat contaminated food, you won't drop dead on an instant. Instead you get food poisoning, resulting in stomach aches and vomiting. Eat contaminated food on daily basis and you'll end up 6 feet under. Essentially the same is with your MoBo and other hardware, which has been fed the "bad" power by your poor PSU.
How far the damage has developed to - hard to tell. MoBo is usually 1st one to go when used with iffy PSU. GPU is next in line. CPUs are robust but not invulnerable either.


How old?
Then again, it doesn't matter.



Like i said, Corsair CX, at best, is mediocre quality PSU and unless you test with top-of-the-line PSU (e.g Seasonic PRIME TX, Corsair AXi or Super Flower Leadex Titanium), you can not rule out PSU issue.

Moreover, how can you tell that the 2nd CX also functions properly? Do you have high-end, very expensive PSU testing equipment, which you can use, to make sure that 2nd CX works within the specifications of ATX PSU standard?


I wouldn't call a PSU being good (or served well), which in the past 6 months, has damaged/degraded your hardware up to the point of actual hardware failure, that makes PC inoperable. Sure, it powers it on, but what good is this power, when it damages your hardware?

For example, in my main build, Skylake (full specs in my sig), i have Seasonic PRIME TX-650, which i bought back in 2016 and which now, 7+ years later, has produced 0 issues. No hardware degradation, damage or outright death. Nada. Nothing. <- This is to be expected when buying the best 650W PSU money can buy. Which also has the best efficiency there is (80+ Titanium), mythical levels of voltage regulation, very tight ripple. Double + then some of hold up time, over what ATX PSU standard specifies (standard is at least 16ms, PRIME TX-650 has it at 35ms), very quiet operation, 12 years of warranty, etc.

Your CX600 unit is 80+ Bronze with 3 years of warranty. For a 10 year old unit, your CX is way past it's time. And it wasn't good enough even when it was brand new.

So, you trying with another CX unit, regardless the age (CX600 platform was released in 2013) wouldn't be a good idea. Better get new, up-to-the-date, good quality PSU and test with that, while crossing your fingers that the rest of your hardware isn't damaged (i won't get the hopes up though).

Again, thank you very much.
The other PSU I own I got from a friend, no clue how old it is. I decided to try it out anyhow and it has been running smooth for a few hours now with no issues. Will continue tomorrow, while looking for a more permanent PSU.