Event ID 129 - stornvme - Reset to device, \Device\RaidPortx - WHEA_UNCORRECTABLE_ERROR

Tryndabeard · 7 Aug 2023 at 22:21

I recently upgraded my desktop and since doing it stability has gone down the drain. New build consists of

5800X3D
ROG STRIX B550-F
RX 7900 XT
2x16gb Corsair DDR4 Vengeance RGB PRO SL Black
SK hynix Plat P41 2TB m.2
be quiet! Straight Power 11 Platinum 850 Watt 80+ Platinum

And carried some old drives over

Samsung SSD 860 EVO 1TB
ADATA SX8200PNP

Since first boot it has randomly started stuttering intensely and eventually BSOD's with a WHEA_UNCORRECTABLE_ERROR but fails to produce a minidump, BSOD just sat there on x% until I had to forcibly shut down. Seems to be more frequent when playing games. Everytime it does this there's stornvme entries in the event viewer.
Initially I thought the new m.2 which I had installed in the m.2_1 slot was to blame. I tried running every diagnostic I could find and everything came back saying it was healthy but I removed it to be sure. Since removing it seems to last a bit longer without throwing the WHEA_UNCORRECTABLE_ERROR. Additionally since removing it the log shows Reset to device, \Device\RaidPort1, was issued. whereas before it was Reset to device, \Device\RaidPort2, was issued.. You can see here there's a whole load of entries, I was playing BG3 when these started and new entries became significantly less frequent shortly after I closed the game around 21:24ish but didn't stop completely:

For the times where it did BSOD while I was running with both m.2's installed the event log notes an unexpected shutdown and reports about the failed to create the dump:

I've also done RAM diagnostics and CPU stress tests which all seemed fine, and numerous DISM / SFC / chkdsk runs which all come back good.

I'm running out of ideas, I was going to try resetting windows but if that doesn't fix it I'll be back to replacing bits of hardware starting with the CPU followed by the mobo.

Has anyone else got any other suggestions?

Tryndabeard · 7 Aug 2023 at 22:54

Tell a lie, got one minidump

073023-18234-01.dmp

DMP File

1drv.ms

Murphy · 7 Aug 2023 at 23:22

WHEA is a Windows Hardware Error Architecture and stornvme is a NVME storage so it's going to be something wrong with you NVME drive.

If you go into device manager > Storage controllers > Samsung NVME controller > Properties > Drivers tab. What driver provider does it say it's using?

Tryndabeard · 7 Aug 2023 at 23:39

Murphy said:
WHEA is a Windows Hardware Error Architecture and stornvme is a NVME storage so it's going to be something wrong with you NVME drive.

If you go into device manager > Storage controllers > Samsung NVME controller > Properties > Drivers tab. What driver provider does it say it's using?

All the information seems to point to that but there's only 1 NVME drive currently installed in the desktop which I have had and used for years without issue making me think root cause is the mobo, but i'm out of my depth here.

Interestingly opening the properties of the Disk Drives entry (last image) caused a little freeze and has kicked off a trickle of the stornvme entries

For clarity, the Samsung is a 2.5

Murphy · 8 Aug 2023 at 06:20

Ah, sorry i missed that the NVME is the P41 last night, it was late.

Thanks for posting the images as that rules out a driver/software issue.

I'm going to guess you're not overclocking anything?

It may sound silly but have you tried reseating (taking out and putting back in) the RAM and NVME drive, there's a slim possibility they may not be making good connections.

And i assume you've not installed the SATA (AMD RAIDXpert2) drivers? Only the LAN, Chipset, and Audio drivers from the ASUS site?

Tryndabeard · 8 Aug 2023 at 09:53

Murphy said:
Ah, sorry i missed that the NVME is the P41 last night, it was late. Thanks for posting the images as that rules out a driver/software issue.

I'm going to guess you're not overclocking anything?

It may sound silly but have you tried reseating (taking out and putting back in) the RAM and NVME drive, there's a slim possibility they may not be making good connections.

And i assume you've not installed the SATA (AMD RAIDXpert2) drivers? Only the LAN, Chipset, and Audio drivers from the ASUS site?

Everything is at bios defaults and everything has been reseated at one point or another

Yes only installed those drivers.

I reset windows last night to no avail
There's a lot online about unhiding and disabling AHCI power saving settings but that hasn't helped.

Murphy · 8 Aug 2023 at 12:11

I was afraid that may have been the case as having pretty much ruled out a software/driver being the cause it kind of only leaves an issue with the hardware, probably the P41.

I'm not sure if you can be bothered with the faff but it maybe worth sticking Linux on a USB drive and live booting that just to see if the kernel logs (dmesg) provide you with a bit more info, I've always found Linux to be a bit more revealing when it comes to faulty hardware.

e: The only reason I'd do that is to potentially save time returning the P41 only to find out it may not be the cause, although that does seem like the most likely candidate TBH.

2nd e: I assume you've tried temporarily disconnecting all the other drives except the P41, just to try and isolate if it's that drive that's causing it.

Tryndabeard · 8 Aug 2023 at 15:56

The P41 hasn't been installed in a while, just the adata, so I don't think there's any issues with it. The adata I had been using for years previously without issue so unlikely to be that either.

Murphy · 8 Aug 2023 at 17:08

Then I'm at a bit of a loss TBH, I'm +90% sure it's being caused by a drive and seeing as the adata is the only drive attached to the system that's the most likely culprit.

It's worth saying the sort of errors you're getting won't be detected by any disk scanning software like chkdsk as it's not a filesystem error, it's a hardware error. I had similar warnings when an old SATA SSD started disappearing from Windows but that didn't cause BSOD as it was a secondary drive and not the one Windows was installed on.

I've got what drives you're using pretty wrong every time so far but is it safe to assume the adata drive is the one Windows is installed on? If so it maybe worth cloning it to another drive and booting off that other drive just for a few days to see you get the same problem when using a drive other than the adata one to boot from.

Tryndabeard · 8 Aug 2023 at 17:14

Murphy said:
Then I'm at a bit of a loss TBH, I'm +90% sure it's being caused by a drive and seeing as the adata is the only drive attached to the system that's the most likely culprit.

It's worth saying the sort of errors you're getting won't be detected by any disk scanning software like chkdsk as it's not a filesystem error, it's a hardware error. I had similar warnings when an old SATA SSD started disappearing from Windows but that didn't cause BSOD as it was a secondary drive and not the one Windows was installed on.

I've got what drives you're using pretty wrong every time so far but is it safe to assume the adata drive is the one Windows is installed on? If so it maybe worth cloning it to another drive and booting off that other drive just for a few days to see you get the same problem when using a drive other than the adata one to boot from.

It is my boot drive, I'll give cloning a go

Donnie Fisher · 8 Aug 2023 at 23:10

How old is the psu ?

Could be a cpu / psu issue as I had similar instability issues with a 5800x I installed in my machine with a slightly older psu.

I think the cpu jumps so aggressively between its boost clocks and back to idle that the spikes and drops in power demands on the PSU trip something in the power delivery circuit throwing the whea errors.

Mine would trip on medium / lighter loads as that is when it was boosting / idling more than if you were running a stress test which would draw sustained power from the PSU.

In the end I turned down the pbo settings and similar to get a more stable setup.

I,ve since learned that I think I might have done better by limiting the power on the cpu in PBO to something sensible (105w or even less etc )… that way it would let individual cores boost, but not draw unlimited loads from which the PSU might not like dropping back from.

Might be worth a look at that side of thing.

Tryndabeard · 9 Aug 2023 at 00:17

Tryndabeard said:
It is my boot drive, I'll give cloning a go

I opted for a fresh install on the P41 as I'd already reset Windows so still needed to sort out my settings and whatnot, and that's fixed it!

Means the older nvme is defective. I was convinced it couldn't be the problem as I didn't have this issue prior to the upgrade and every diagnostic tool said the drive was fine but now I'm questioning if some of the occasional strange behaviours I saw prior to the upgrade existed because this drive was defective. Or maybe it got damaged when I moved things about. Either way I should have had more faith in what the event log was telling me I guess. Thanks for the help