Reset to device, \Device\RaidPort0, was issued.

[MRuos1]

I am using a SATA RAID0 array as disk cache for a video production application. While the application is rendering (writing and reading from the array), there is a randomly recurring problem that freezes the video application, the IRST-application and prevents the computer from restarting / shutting down normally. When this happens, the system log shows the following error repeated every 30 seconds until a hard restart is done:

Reset to device, \Device\RaidPort0, was issued.

- System

- Provider

[ Name] iaStorA

- EventID 129

[ Qualifiers] 32772

Level 3

Task 0

Keywords 0x80000000000000

- TimeCreated

[ SystemTime] 2018-02-16T11:32:43.946354000Z

EventRecordID 9296

Channel System

Computer [MyComputerName]

Security

- EventData

\Device\RaidPort0

0F001800010000000000000081000480010000000000000000000000000000000000000000000000000000000000000006040000810004800000000000000000

My system info:

OS Windows 10 Pro N version 1709 build 16299.248

CPU Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz

Motherboard Gigabyte GA-Z270-HD3P rev.1

BIOS Z270-HD3P F9b (latest)

RAM 32GB G.Skill F4-2400C15D-32GVR

Drives connected to SATA ports:

SDD Kingston SV300S37A120G

HDD Hitachi HDP725050GLA360

HDD Toshiba MD04ACA50D

DVD-ROM Asus DRW-2014L1T

RAID0-array 2 x WD1002FAEX-00Z3A0

IRST version 15.9.0.1015 (latest from manufacturer)

RAID option ROM 15.5.0.2875

Disk data cache Enabled

Write-back cache Write back

Link power management Disabled

I have clean installed Windows and created the RAID0-array in the UEFI settings. The system boots UEFI without the legacy option ROM. IRST reports zero parity errors or blocks with media errors. I have also extensively tested all connected drives for errors with several other tools. IRST controls write-cache buffer flushing. All power options are on high performance – I even made the ACHI Link power management settings visible from the registry to confirm they are correct. The system temperature stays low enough.

The RAID array is fast, as it should, and sometimes works days on end before the error manifests itself again – it happens way too often to ignore, though. The system does not have any other issues. Can you please help me?

[idata]

Hello Suupala

Thank you for joining the community. Could you provide us with the following information:Intel® System Support Utility for Windows* report https://downloadcenter.intel.com/download/25293/Intel-System-Support-Utility-for-Windows- https://downloadcenter.intel.com/download/25293/Intel-System-Support-Utility-for-Windows- Intel® RST report following this steps: https://www.intel.com/content/www/us/en/support/articles/000006351/technologies.html https://www.intel.com/content/www/us/en/support/articles/000006351/technologies.html Picture of the error Please attach the information to the threadRegards,Leonardo C.

[MRuos1]

Hi Leonardo,

thank you for the quick answer! You'll find the reports and screenshots you asked for in the attachments. Let me know if you need anything more to help troubleshoot the issue!

Regards,

Mikko

[idata]

Hello Suupala

Thank you for the information provided What is the application that you are using for rendering? Does it happen with other applications? In what ports as the raid disk connected? Regards Leonardo C.

[MRuos1]

Hi Leonard,

I'm using Adobe After Effects.

Earlier I had my Windows swap file on the array also, which resulted in random BSOD's when the issue occurred - I now have it on another HDD to limit the risk of damage with hard restarts. However, since I use this system pretty much exclusively for doing After Effects work (which is what I need the fast RAID array for), I cannot say with absolute certainty whether the issue would persist with using only some other applications.

As indicated in my IRST report, my RAID disks are connected to ports 4 and 5.

Regards,

Mikko

[idata]

Hello Suupala

Thank you for the information provided Could you try the following steps to use your system? try using the system with de ASUS DRW-2014L1T unplug to see if the error happens, if yes could you try this set up: plug the RAID volume disk on port 0 and 1, and the ASUS DRW-2014L1T on port 5, use the other ports for the nonRAID disk Make sure the information is back up to try these steps Regards Leonardo C.

[MRuos1]

Hello Leonardo,

Thank you for the suggestion! I will reconnect the drives accordingly and post the results in a few days.

Regards, Mikko

[idata]

Hello Suupala

I was checking your case and would like to know if you need further help. If so, please do not hesitate in replying back. Regards,Leonardo C.

[MRuos1]

Hello Leonardo,

I got held up with a work project and had to delay testing the system a bit. I will post the results next week at the latest!

Regards, Mikko

[idata]

Hello Suupala

I was checking your case and would like to know if you need further help. If so, please do not hesitate in replying back. Regards,Leonardo C.

[MRuos1]

There was a further delay with my current project on the system, and I don't dare to change the system configuration until I've finished it, as the system works more or less OK without the write-back cache. Immediately when this project is out of the way, I'll test the new connections and post the results here!

[AData]

Greetings and salutations.

I had a similar problem. After days of testing I've discovered that both RAID-5 and RAID-0 arrays hang completely whenever "write back" cache option is used for array.

I'd like to point out that this makes RAID-5 COMPLETELY unusable, as with "write through" option write speed is 20 MB/s (and with other options it's even worse).

I'd also like to point out that internet is OVERFLOWING with reports of similar errors starting about year 2010. And during all this time you guys at Intel could not have figured this out?!

When I was truing to use RAID-5 your RST driver would post this message in my notification panel on every launch (attached). Well, guess what - it wasn't. It wasn't protected, and the reason was your own driver.

I can't help but think how bloody lucky I was that my system was on another drive. I pity anyone having OS on the same array, as in this case PC would hang completely, and it would be extremely difficult to troubleshoot the cause of it.

[idata]

Hello Suupala

I was checking your case and would like to know if you have news on the steps provided previously Regards,Leonardo C.

[idata]

Hello Suupala

I was checking your case and would like to know if you have news on the steps provided previously Regards,Leonardo C.

[MRuos1]

Hello Leonardo,

I tested reconnecting the drives according to your suggestion. Unfortunately the issue manifested itself again. The RAID array only worked stable with the write-back cache disabled.

The only new side note that might have to do with the problem is, that After Effects gives me a warning if I try to allocate a cache size bigger than would fit on just one of the HDDs of the array. It allows me to do it, though, and even staying under this limit the array keeps resetting. Still, there might be a compatibility issue worth looking at here.

Unfortunately, as either the reason for or consequence of the many hard restarts, I eventually got a SMART event notification (HARDWARE IMPENDING FAILURE GENERAL HARD DRIVE FAILURE [asc=5d, ascq=10]) on other one of the array HDDs. As I presently cannot afford to invest in new hardware and since the RAID array has been the only source of problems on my otherwise stable system, I've disassembled the array at least for now, and cannot continue troubleshooting.

Maybe you can give me some pointers about what the advantages or drawbacks of going to ACHI mode from BIOS would be?

Regards,

Mikko

[idata]

Hello Suupala. Thank you very much for your reply.

The only difference between AHCI and RAID mode on the BIOS will be the capability of your system of creating a RAID volume. Using AHCI the drives will be managed individually and can not be in an array.Antony S.