CHKDSK found errors?

[Iggy Wolf]

So I was playing RE3 Remake when during a fight, one of the car textures became purple and I wasn't sure why. I attempted to alt-tab to desktop, at which point, my computer just froze and then proceeded to crash to a BSOD, which was stating that the computer has encountered an error and will restart.

The "error" in question was reading "System Service Exception". I decided to do an SFC scan after the restart, which found some errors and fixed them. But I also decided to do a CHKDSK, which ALSO found errors. I then scheduled to have the C drive, which had my OS files on it, to be checked after restart.

I restarted the PC and the CHKDSK ran and found and repaired the files. Should I still be concerned? I mean, I've had random unexpected shutdowns before, but usually it was simply the game freezing, and the PC instantly shut down and restarted without an actual blue screen. The Event Viewer would simply say a bug check occurred.

My specs are an i5 10400, RTX 3060 Ti, 16 GB of RAM, and Windows 10 Home if it matters. The C drive is an NVMe SSD. I used Crystak Disk Info and the S.M.A.R.T. scan shows the SSD to be "Good" Health, at 97%. I didn't see anything significant in the "raw" values but I couldn't say what's "good" or "bad" in the values though. I'm assuming the BSOD caused the disk errors, since it caused an unexpected, and therefore, "unsafe shutdown".

[PopinFRESH]

Which model SSD is it?

Also, CHKDSK is a filesystem level thing. Filesystem errors can occur for a plethora of reasons beyond hardware being an issue.

The BSOD was more likely a result of those filesystem errors than the other way around.

[Iggy Wolf]

Ahh, ok. And the SSD is an MSI M390 1 TB. I have also noted that whenever I exit RE2 or RE3 Remake, the event viewer will show that there was an "error". The "faulting module" in this case is KERNELBASE.dll. I don't know if that means my kernelbase.dll file is corrupt. I had done an SFC scan and it said it repaired the corrupted files it found. Not sure if that means my kernelbase.dll is still corrupt. The game shuts down. It doesn't seem like it crashes on exit or anything. In fact, Steam shows that it uploads the game data to the cloud after closing.

[PopinFRESH]

If you can post a screen shot of the SMART attributes table for the SSD

[Heretic]

Bad ram can cause all of this and if they're intermittent errors it may not appear obvious. I'd try and rule out the drive first, because that should be easy. For RAM you'll have to stress test, reboot and repeat.

[Bad 💀 Motha]

CHKDSK /F /R
Verify everything about the disk. Such as files, free space, marks off bad sectors if found so data can't go into a bad spot of the drive.

SFC /SCANNOW
Verify your Windows OS

[Illusion of Progress]

Origineel geplaatst door Iggy Wolf:

but usually it was simply the game freezing, and the PC instantly shut down and restarted without an actual blue screen. The Event Viewer would simply say a bug check occurred.

That is likely a BSOD. By default, Windows is set to automatically restart when encountering a STOP error (or BSOD) after making the log/dump of it, and it happens so fast you don't see it. If event viewer is saying that, then it's a STOP error/BSOD most likely that you've been having all along.

Regardless, if your PC is often restarting or turning off when doing demanding things, that's an issue that you shouldn't ignore. Unexpected shutdowns mean just that. The Windows system wasn't expecting to lose power or have its operation terminated, but it did.

They are often a PSU issue but they can be hard to troubleshoot. I'm (though not presently, unfortunately) dealing with that sort of nightmare only mine come and go when they please, and they never happen under load (if they did, I'd heavily lean towards PSU but in my case I already ruled that out).

Also, what PSU do you have?

[plat]

OP says he already ran sfc and it found and fixed some things.

Did Windows create a mini-dump?

C:\Windows\minidump

https://www.tomshardware.com/how-to/minidump-fix-blue-screen-of-death

[PopinFRESH]

Origineel geplaatst door plat:

OP says he already ran sfc and it found and fixed some things.

Did Windows create a mini-dump?

C:\Windows\minidump

https://www.tomshardware.com/how-to/minidump-fix-blue-screen-of-death

And a chkdsk on boot. Bad motha be like "hey do the two things you've already done".

Not to mention its an NVMe SSD, so the "marking off bad sectors" is transparent to filesystem level things.

[_I_]

normal chkdsk does not fix or find all problems

chkdsk /f /r
will force it to do a full scan and fix everything

the ssd should be marking bad sectors by itself, before the file system even sees they are bad unlike hdd
ssd has some overhead for replacing bad ones

which is why checking the drives smart table will show you what it actually is doing

[Iggy Wolf]

Well, I do also have a secondary drive which is an HDD. But I ran the Chkdsk on that one and it found no errors. My PSU is a Thermaltake SMART Series 700W 80+ Gold. MSI Afterburner shows my GPU rarely uses more than 200W, so I doubt the remaining 500W wouldn't be enough. These BSOD's happen rarely enough that it's hard to nail down who or why.

I mean, I've played entire sessions of games like Metro Exodus Enhanced Edition, without one BSOD, and that one is much more demanding than RE3 Remake. Ditto for Spiderman Remastered. Maybe it could be my PSU more so than my GPU. GPU rarely exceeds 65 C. Maybe SOMETIMES the VRAM might hit the ceiling limit (or get close with 7980 being used) but that would only cause the game to either crash or be forced to use the slower system RAM.

My RAM is 16 GB Olay 3000 Hz. Task Manager shows the PC uses 2900 Hz, so more or less the full amount. My RAM could be bad, but it could just as well be the PSU. It's hard to say whether it's long sessions that are responsible, as I'll sometimes leave the game to go eat or simply to change a mod.

I will say that with both RE2 and RE3, upon exiting the games, the Event Viewer will always show an Error specifically for the game's exe, indicating "Faulting Module: KERNELBASE.DLL ", which is odd. I had SCF repair corrupt files, so I don't know if that means my Kernelbase.dll file is corrupt.

[plat]

Not that it would necessarily help with the original problem, but it's routine now that if sfc /scannow comes up abnormal ( whether fixed or unable to fix) to run this directly afterward:

dism /online /cleanup-image /restorehealth

Then, when it completes, you run sfc again and it should find no "integrity violations."

[_I_]

post a cpuz validation link
http://www.cpuid.com/softwares/cpu-z.html
cpuz -> validate button -> submit button
it will open a browser, copy the url (address) and paste it here

[Azza ☠]

A CHKDSK finding errors and repairing them is fine... once in a blue moon. You should reboot and rerun the scan again. A drive has a small reverse area to even replace them. If that doesn't return more errors, then you can consider the drive is still fine.

It's not a concern until errors start repeating themselves.

If the drive is physically damaged and keeps creating new errors, back up your data and replace it quickly.

[Iggy Wolf]

Not sure if it matters but I'm still on the 516.24 drivers for my GPU. Hard to say whether the outdated drivers are necessarily causing the BSODs. I've had bad drivers in the past on my old PC, and when they did get corrupt, my MSI Afterburner no longer detected my GPU because Device Manager showed that the GPU was disabled and couldn't be reenabled until I updated the drivers.

I'm only more concerned about it on this PC because the CPU is an i5 10400F. So it doesn't have an "internal iGPU" to fall back on should my current GPU or drivers ever fail. Not sure how it would even load video for the desktop without an internal GPU. I'd literally have to plug in an old GPU just to access the desktop.

[Iggy Wolf]

Origineel geplaatst door plat:

Not that it would necessarily help with the original problem, but it's routine now that if sfc /scannow comes up abnormal ( whether fixed or unable to fix) to run this directly afterward:

dism /online /cleanup-image /restorehealth

Then, when it completes, you run sfc again and it should find no "integrity violations."

I actually decided to run the DISM Check Health and Scan Health. No component corruption was found.I did the SCF scan again and it found "no integrity violations". I honestly think it might be a hardware fault. The BSODs were actually happening ever since I bought the prebuilt PC. Within the first month, I got 2 BSODs. So I'm thinking they maybe either cheaped out on the PSU or RAM or said PSU or RAM might be faulty. I don't think it's the GPU.

To put it in perspective, I almost NEVER get a BSOD while simply browsing my Internet or being at the desktop. Only when either gaming or when exiting a game. So there's definitely an "under load" component that leads to the PC needing to shut down. Problem is, it happens rarely enough these days that the lack of consistency makes it hard to nail down. My last BSOD was probably months ago (last year in the summer). I think if anything, the corrupt system files with the SFC and CHKDSK occurred BECAUSE of the BSODs.