Repeated \Device\RaidPort0 reset events during standby

arkoenig · Nov 24, 2023

Dell Precision 3660, Windows 11 versioin 23H2 (build 22631.2715). Intel Optane controller being used for RAID (no cache) with two arrays, one 2xSSD, the other 2xHDD.

The machine runs fine--however, while it is asleep (in "modern standby" mode), the Event Viewer shows a steady trickle of warning events of the form "Reset to device, \Device\RaidPort0, was issued." Source: iaStorVD, event ID 129. Several events per minute the whole time the machine is asleep.

I have tried the suggestions in this article, to no avail:

https://www.drivereasy.com/knowledge/reset-to-device-device-raidport0-was-issued-fixed/#a

Can anyone help me figure out what is going on? Is this even a problem? If so, what is causing it? Is it possible that the HD drives I'm using don't handle "modern standby" power management properly (they're Western Digital Red Plus, which are supposedly intended for RAID applications).

iko22 · Nov 24, 2023

arkoenig said:
Can anyone help me figure out what is going on?

Guidance for troubleshooting data corruption and disk errors - Windows Server

Provides guidance to help troubleshoot data corruption and disk errors in Windows.

learn.microsoft.com

arkoenig said:
Is it possible that the HD drives I'm using don't handle "modern standby" power management properly (they're Western Digital Red Plus, which are supposedly intended for RAID applications).

Several users with similar problem reported the problem went away after disabling Link State Power Management. Even temporarily disabling this would help indicate whether you have the same problem or not. See:

Turn Link State Power Management On or Off in Windows 11

Learn more about PCI Express Link State Power Management and find you whether you need it and how to turn it On or Off in Windows 11/10.

www.thewindowsclub.com

arkoenig · Nov 24, 2023

I've tried every combination of Link State Power Management settings that I have been able to think of--nothing seems to work.

The last paragraph of the "Turn Link State Power Management On or Off" article says:

In order to disable the Power Management link, you need to open the System Configuration and click on the BIOS/Platform Configuration option. Then, select the Power Management option, and Advanced Power Options menu, respectively.

When I open the System Configuration app, I cannot find a BIOS/Platform Confirmation option--so I'm stuck. Suggestions?

iko22 · Nov 24, 2023

It is in the BIOS. For the Dell Precision 3660, enter UEFI BIOS > System Setup Options > Power, then select ASPM (Active State Power Management) and configure option as Enable/Disable (Default is Auto).

See Table 9. System Setup Options - Power Menu on page 86 of the Service Manual, here:

https://dl.dell.com/content/manual59803069-precision-3660-tower-service-manual.pdf?language=en-us

arkoenig · Nov 24, 2023

Thanks for the pointer! It looks like the options available are Auto, Off, and F1 only. I have no idea what the latter means; so I'm setting it to Off to see what happens.

arkoenig · Nov 24, 2023

The answer is that it doesn't help -- I'm still getting "Reset to device, \Device\RaidPort0, was issued." events while the machine is asleep but not while it's awake.

iko22 · Nov 25, 2023

arkoenig said:
I have tried the suggestions in this article, to no avail:

https://www.drivereasy.com/knowledge/reset-to-device-device-raidport0-was-issued-fixed/#a

Just checking. Did you try update the device drivers that was also mentioned in the link?

You could also try and obtain a Power Efficiency Diagnostics Report, and check the Report for Errors and Warnings. Here is how to obtain the Power Efficiency Diagnostics Report:

Create Power Efficiency Diagnostics Report in Windows 10

How to Create a Power Efficiency Diagnostics Report in Windows 10

www.tenforums.com

arkoenig · Nov 25, 2023

I've checked specifically that the driver for the RAID controller is the latest available. I've also verified, using Dell's automated tool, that all drivers are the latest versions available from Dell. Intel says that they do not recommend installing their own drivers when OEM versions are available, because the OEM drivers may contain manufacturer-specific customizations.

arkoenig · Nov 25, 2023

When I tried generating the report as suggested, the following messages appeared:

Enabling tracing for 60 seconds...
Observing system behavior...

and then nothing. No report. No suspicious events in Event Viewer, either.

Other powercfg options, such as "powercfg /sleepstudy", seem to work just fine.

The help output from powercfg says

The system sleep diagnostics report has been deprecated and replaced with the system power report. Please use the command "powercfg /systempowerreport" instead.

When I run that command, I get a report, a copy of which is attached. I believe that the \Driver\RDID1110 mentioned in the report is associated with my audio interface (Roland Octa-Capture).

iko22 · Nov 25, 2023

arkoenig said:
The help output from powercfg says

The system sleep diagnostics report has been deprecated and replaced with the system power report. Please use the command "powercfg /systempowerreport" instead.

Oh!? Works on my Windows 11 .. Did you run the command in an Elevated Command Prompt Window (also known as an Administrators Command Prompt Window)?

arkoenig · Nov 25, 2023

Yes, I did run it in an elevated command window.

Just for fun, I tried "sfc /scannow". It said that it "did not find any integrity violations."

arkoenig · Nov 27, 2023

More information...

I tried removing the HDD drives from the system, leaving only the three NVMe drives (two in a RAID array, the third one separate), and the reset events went away.

I'm wondering whether perhaps there is a compatibility problem between how the HDD drives (Western Digital Red Plus) interact with the computer's RAID controller. Because the Red Plus drives are intended for use in NAS, perhaps they do something odd when they get standby requests because (for example) they want to be ready for access at a moment's notice.

So I am taking advantage of Western Digitals Cyber Monday sale to buy some WD Black drives, and we'll see if using those instead of the Red Plus drives improves the situation. If not, I can return them.

iko22 · Nov 27, 2023

arkoenig said:
I tried removing the HDD drives from the system, leaving only the three NVMe drives (two in a RAID array, the third one separate), and the reset events went away.

In that case, are the Western Digital Red's very old, or something? You can check their health with Hard Disk Sentinel:

Hard Disk Sentinel - HDD health and temperature monitoring

Monitoring hard disk health and temperature. Test and repair HDD problems, predict disk failure.

www.hdsentinel.com

Or, use Western Digital's own software:

Download, Install, Test Drive and Update Firmware Using Western Digital Dashboard

Find detailed answers to your support questions for your Western Digital, SanDisk, WD_BLACK, or WD storage product.

support-en.wd.com

arkoenig · Nov 27, 2023

The drives are brand new. And I can't really check much of anything on them because they're in a RAID array and the controller shields them from direct access.

In principle I could take them off RAID, but if I do, rebuilding the RAID array afterwards will take about two days because the drives are so large. So I would rather wait for the replacement drives to show up and see whether they behave differently.

Also, remember that the problem is showing up ONLY when the machine is in standby mode--that is, when the drives are NOT being accessed. That's why I think the problem has to be in the power-cycling handshake between the drives and the controller.

iko22 · Nov 27, 2023

arkoenig said:
And I can't really check much of anything on them because they're in a RAID array and the controller shields them from direct access.

Did you use the Intel Memory and Storage Management software to create the RAID array? You might be able to use that software to "see" the individual disks and check individual status, although I am not familiar if this is possible or not.

arkoenig · Nov 27, 2023

The

iko22 said:
Did you use the Intel Memory and Storage Management software to create the RAID array? You might be able to use that software to "see" the individual disks and check individual status, although I am not familiar if this is possible or not.

Yes, I did (and do) use the Intel Memory and Storage Management software. And it shows me vey little about the individual disks--see attached image. I can find no way of drilling down beyond this little display.

iko22 · Nov 27, 2023

arkoenig said:
The

Yes, I did (and do) use the Intel Memory and Storage Management software. And it shows me vey little about the individual disks--see attached image. I can find no way of drilling down beyond this little display.

Well, if Windows cannot see the detail, then it is hardly able to generate an error Event as a response.

arkoenig · Nov 27, 2023

iko22 said:
Well, if Windows cannot see the detail, then it is hardly able to generate an error Event as a response.

The "device" that Windows generates the error event for is the RAID array, not the individual drive. And the error event that it is generating is that the RAID array is reporting that it was reset.

Somewhere I saw the claim that this particular event can happen when the RAID controller is waiting for one of its disks to do something that the disk fails to do within a time limit. Which is why I think the problem is that the disk itself is not intended to be compatible with the kind of hybrid hardware/software RAID on this maching, but rather the kind of standalone hardware RAID devices one finds in larger storage facilities or data centers.

iko22 · Nov 27, 2023

arkoenig said:
Somewhere I saw the claim that this particular event can happen when the RAID controller is waiting for one of its disks to do something that the disk fails to do within a time limit.

Here, https://www.intel.com/content/dam/s...tel_raid_basic_troubleshooting_guide_v2_0.pdf
on page 43, under Data bus or device timeout error, Intel recommend checking the SATA cable connection and cable type. Make sure the cable is properly seated in the motherboard connector. It also recommends NOT to dismantle the disk array until the fault has been diagnosed.

arkoenig · Nov 27, 2023

iko22 said:
Here, https://www.intel.com/content/dam/s...tel_raid_basic_troubleshooting_guide_v2_0.pdf
on page 43, under Data bus or device timeout error, Intel recommend checking the SATA cable connection and cable type. Make sure the cable is properly seated in the motherboard connector. It also recommends NOT to dismantle the disk array until the fault has been diagnosed.

I do not understand what kind of cable problem would result in timeouts only while the machine is asleep and trouble-free behavior when the machine is running. In any event, we'll see what happens with the drives I have on order.

Repeated \Device\RaidPort0 reset events during standby

Member

My Computer

Well-known member

My Computers

Member

My Computer

Well-known member

My Computers

Member

My Computer

Member

My Computer

Well-known member

My Computers

Member

My Computer

Member

Attachments

My Computer

Well-known member

My Computers

Member

My Computer

Member

My Computer

Well-known member

My Computers

Member

My Computer

Well-known member

My Computers

Member

Attachments

My Computer

Well-known member

My Computers

Member

My Computer

Well-known member

My Computers

Member

My Computer