Hero Z690 having issues with getting windows errors and nvme drive not working

[Medwynd]

EDIT:
ASUS has recognized this as an official issue:

Hi all,
For WHEA 17 error, our tech team have reported it and co-work with Intel.
If there is any news, I'll keep updated.
Thank you.

Possible solutions for this issue can be accomplished in one of two ways, both involve turning off PCI Express power management.
1. In windows go to Power Options and change the plan setting to High Performance . You may also be able to just go to the advanced settings and change Link State Power Management to "off"
2. In the BIOS set PCI Express Native Power Management to disabled instead of auto


Trying to track down is this is an issue with the motherboard, the nvme drive, or drivers.

Using windows 11 on a Hero Z690 with a WD sn850 nvme drive.

Occasionally the drive will appear, sometimes it will take a reboot, sometimes it will appear then disappear. Getting these types of error an warning events:

After rebooting I see
Source: Microsoft-Windows-WHEA-Logger
Description: A corrected hardware error has occurred.
Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)
Primary Device Name:PCI\VEN_8086&DEV_7AB4&SUBSYS_86941043&REV_11

Looking in device manager I see that Intel PCI Express Root Port #13 - 7AB4
Device PCI\VEN_8086&DEV_7AB4&SUBSYS_86941043&REV_11\3&11583659&0&EC had a problem starting.

Then later
Source: stornvme
Reset to device, \Device\RaidPort3, was issued.

Source: stornvme
The driver detected a controller error on \Device\RaidPort3.

Followed by some warning errors on the source disk such as ntfs unable to flush data etc. presumably because the drive no longer has connection or is in some otherwise bad state.

I'm using the latest drivers from the support page for the chipset and everything else.

Any ideas where to start with this? I'm not sure if I have a bad motherboard, a bad nvme drive or possibly just some bad drivers.

[STARRAIN_ROG]

Hi jorimt,
Is only WD_BLACK SN850 becoming unreadable and disappearing or any SSD on M.2_3 would been unreadable and disappearing in windows?
Does it also occur on bios 2103 with bios all default settings and only disable PCIE Express Native Power Management?
Does it happen after run the specific apps or any settings changed?
Thank you.

[jorimt]

STARRAIN@ROG wrote:

Is only WD_BLACK SN850 becoming unreadable and disappearing or any SSD on M.2_3 would been unreadable and disappearing in windows?

Only the SN850 thus far.

For further background, when I initially built the PC, I had the 980 in M.2_1, the SN850 in M.2_2, and the 970 in M.2_3, and from December 2021 all the way up to July 2022 (around the time I upgraded the BIOS to 1601), zero issues.

It's only after I moved the 980 to M.2_2, the SN850 to M.2_3, and the 970 to the DIMM card (and yes, I tried removing it to see if it was the cause; it wasn't) and upgraded to 1601 that this issue started occurring. Again, I was able to fiddle around with some of the BIOS settings (can't remember for the life of me the exact combo; the issue was so random that my "solution" matched it) and made it stop for a few months again until I upgraded to BIOS 2103.

STARRAIN@ROG wrote:

Does it also occur on bios 2103 with bios all default settings and only disable PCIE Express Native Power Management?

Interesting you should ask that, because after I posted my message last night, I reset the BIOS settings to default, and put back all my settings overrides (shared in my previous post) but revised/left the below:

PCIE Express Native Power Management: Disabled > Enabled
ASPM: Disabled > Auto
Next Boot After AC Power Loss: Normal Boot > Fast Boot

And as of last night and this morning, the drive hasn't disconnected in the time it usually does and is still working...for now. Why that is is currently beyond me. This is the most random issue I've experienced, and I've now built multiple PCs across many Intel platforms and generations.

I kept disabling PCIE Express Native Power Management for this platform because everything I've read suggests it. Whether that ends up being the cause in my case remains to be seen (and still makes no sense). Do note however, I always have PCI Express Link State Power Management disabled whether I'm using the Balanced or High Performance power plan Windows-side.

Anyway, after I give this round of settings more time to run, I will report back here on whether the "fix" sticks. It would be great to know exactly why these issues are happening, regardless.

STARRAIN@ROG wrote:

Does it happen after run the specific apps or any settings changed?

No, entirely random. All I have to do is start the computer up or reboot. The drive initially appears, but after anywhere from a few minutes to an hour, the drive slowly becomes unreadable until the errors appear in the Event Viewer, and clicking on it throws "drive doesn't exist" type errors. At that point, if I refresh the file explorer in "This PC" the drive disappears fully.

As far as I can tell, there's no way to accelerate the process, and it happens whether I turn the PC on and immediately game from the problem drive, or leave the drive inactive all day while I work.

[jorimt]

STARRAIN@ROG wrote:
Hi jorimt,
Does it also occur on bios 2103 with bios all default settings and only disable PCIE Express Native Power Management?

Quick update...

PCIE Express Native Power Management: Disabled > Enabled
ASPM: Disabled > Auto
Next Boot After AC Power Loss: Normal Boot > Fast Boot

After two days with the above, I've not experienced the M.2_3 errors or disconnects.

Wanting to narrow it down further yesterday, I disabled both PCIE Express Native Power Management and ASPM again and re-enabled Normal Boot in the BIOS, and the issue returned.

I then re-enabled PCIE Express Native Power Management, and the issue resolved.

Thus the issue of my M.2_3 drive controller connection slowly degrading to the point of disconnection shortly after each startup or reboot appears to occur on 2103 only if I disable PCIE Express Native Power Management. Apparently there's an issue with allowing the BIOS to control PCIE power management instead of the OS then, or?

Z690 seems to be really sensitive to power-related factors, be it XMP profiles, PCIE, etc.

[STARRAIN_ROG]

Hi jorimt,
The WHEA Error on ASUS motherboards is a Gen4 signal, M.2 SSD and related driver compatibility message, not a hardware design issue on the motherboard itself, and does not affect PCIE and M.2 related functions.
The PCIe slot and M.2 slot related signals on our motherboards are also far superior to the design specifications.

If you use Windows 11 but still experiencing this issue with the latest BIOS and have concerns, we recommend that you consider disabling the PCI Express Native Power Management feature in your BIOS to avoid a WHEA error. Disabling this option will have little impact on normal usage, so please don't worry.

If you have other usage anomalies, we may suggest you to check your peripherals.
Sorry for any inconvenience it may cause.

[jorimt]

STARRAIN@ROG wrote:
Hi jorimt,
The WHEA Error on ASUS motherboards is a Gen4 signal, M.2 SSD and related driver compatibility message, not a hardware design issue on the motherboard itself, and does not affect PCIE and M.2 related functions.
The PCIe slot and M.2 slot related signals on our motherboards are also far superior to the design specifications.

If you use Windows 11 but still experiencing this issue with the latest BIOS and have concerns, we recommend that you consider disabling the PCI Express Native Power Management feature in your BIOS to avoid a WHEA error. Disabling this option will have little impact on normal usage, so please don't worry.

If you have other usage anomalies, we may suggest you to check your peripherals.
Sorry for any inconvenience it may cause.

Thanks for the reply, but that's a copy & paste response from as recently as here in this very thread, and doesn't apply to my situation, since "disabling the PCI Express Native Power Management feature in your BIOS" causes my M.2_3 drive to slowly lose connection until a reboot, after several of which it thinks my drive is corrupted (even though it's not) and tries to repair it only to ultimately put it into read-only mode and disappear again anyway.

As stated in my previous response, enabling PCI Express Native Power Management in the 2103 BIOS is the only thing preventing my issue currently. Apparently no one knows why, including (official?) forum support and moderators, so I'll take your response as an "I don't know" and move on...