Rapid Storage Technology
Intel® RST, RAID
2047 Discussions

Intel RST 12.5 Repeats Verification & Repairing RAID 5

TGold1
New Contributor II
3,139 Views

I have three Seagate Barracuda SATA drives in an Intel RST, version 12.5.0.1066, RAID 5 configuration. I recently had a disk failure and I replaced the disk and successfully rebuilt the array. However, the system recently initiated a verification and repair process and the process repeats on its own.

Indicated under the RST program "Manage" tab, Advanced, I can see the previous Verification and Repair results. Previously, it was 10 Verification errors found and 10 Verification errors repaired. Presently, it's 7 found and 7 repaired.

Is this unusual for the program to repeat after finding and repairing all indicated errors? I think this is the fourth iteration of the cycle of continuous verification and repair. Is this repeated operation an indication of a serious problem with the array that the program can't fix?

Regards

My system:

DX58SO, i7-930, 6GM RAM, 3 Seagate Barracuda SATA 500GB 7200rpm drives, Windows 7 Pro x64 with all updates

Also, I see that Intel RST 12.8 was released on 8/8/2013. Should I upgrade to the latest version?

Regards again.

0 Kudos
12 Replies
Silvia_L_Intel1
Employee
1,509 Views

Hello Tjg79,

The storage controller on this motherboard has another type of architecture and it was based on the ICH10R, newer motherboards are based on PCH.

This kind of error could be cause because the latest RST versions are based on PCH architecture. I recommend you to uninstall the Intel® Rapid Storage Technology 12.5 and then, install the latest version available for your motherboard. According to the downloadcenter.intel.com it is 10.1.0.1008, you can find it on the following link; https://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=20104&lang=eng&OSVersion=%0A%09%09%09%09%09%09%09%09%0A%09%09%09%09%09%09%09&DownloadType=Drivers

Regards.

0 Kudos
TGold1
New Contributor II
1,509 Views

Thank you Sylvia,

I'm unable to install RST 10.1. The system freezes every time and I've tried to install it with a clean install of the Windows 7 Pro x64 OS and just about everything else striped from the system to eliminate the unknown conflict.

In fact, I've attempted so many different installs of RST 10.1 on my system and failed, that I don't think anyone can install that driver on a DX58SO running Windows 7 Pro x64 and succeed.

I've posted about this surprising conflict before, but didn't get much of a response. It was during a support chat that an Intel tech rep suggested running RST 11.0 which installed without issues, but wouldn't create an array after the OS was installed by running the RST program. Another Intel tech support rep suggested RST 12.5 on this DX58SO. It worked and I haven't detected any issues until the drive failure, that wasn't due to RST.

RST 12.5 has worked flawlessly until I had a disk failure. It did run for about four cycles before it stopped. When I do a verification and repair, it does detect errors, but I think that's related to the failed drive. I've deleted my large windows back up files and system image, and I'm running the verification program now with 43% complete and no errors detected. I started this iteration after canceling a previous run with 20 errors found and repaired. So, I think the program is performing as advertised. When the current verification is complete, assuming no errors detected, I'll reboot and run it again to see if error occur after a reboot.

RST 12.5 works as advertised where RST 11.0 and 10.1 don't. That's why I considering trying RST 12.8 released a few weeks ago.

Here's the thread where I was seeking assistance with RST 10.1

/thread/28235 http://communities.intel.com/thread/28235

In several tech support chats I've suggested Intel remove RST 10.1 for the DX58SO and Win7 x64 OS. I see that nothing has been done yet.

I was also told by an Intel Tech Rep that the download center is not always current with the latest approved downloads which was surprising to hear.

0 Kudos
Silvia_L_Intel1
Employee
1,509 Views

Tjg79,

Before installing RS version 12.8 on the system, could you please check if you are running the latest BIOS version? According to downloadcenter.intel.com it supposes to be BIOS version 5600.

If you do not have the latest version installed, please let me know which is the current version before updating it.

Regards.

0 Kudos
TGold1
New Contributor II
1,509 Views

 

Sylvia,

Yes, I have the latest BIOS version installed; 5600.

I read the "Read Me" file for RST 12.8.0.1016 and I didn't see any issues with installing it.

I installed it yesterday and it works fine.

I upgraded from RST 12.5.0.1066

Although, I'm still getting parity errors detected and repaired by RST after each reboot. The number ranges from 8 to 16 errors. I've never had any Blocks with media errors.

RST 10.1.0.1008, RST 11.0.0.1032, & RST 11.2.0.1006 do not work on my DX58SO/i7-930 running Win 7 Pro x64. An Intel Tech Rep suggested RST 11.0, 11.2, & 12.5 to resolve my issues with RST 10.1 as the download center suggests. RST 10 freezes the system on install and the install fails completely. RST 11 installs, but won't create a RAID volume using the RST create function after the OS is installed.

What's Intel's position of running Windows ChkDsk or Win Defrag on an Intel RST RAID 5 array?

Thank you for your assistance.

Regards

Below is a current Intel RST System Report:

System Report

System Information

 

OS name: Microsoft Windows 7 Professional

 

OS version: 6.1.7601 Service Pack 1 7601

 

System name: INTEL-I7

 

System manufacturer: INTEL_

 

System model: DX58SO__

 

Processor: GenuineIntel Intel64 Family 6 Model 26 Stepping 5 2.794 GHz

 

BIOS: Intel Corp., SOX5810J.86A.5600.2013.0729.2250

Intel® Rapid Storage Technology enterprise Information

 

Kit installed: 12.8.0.1016

 

User interface version: 12.8.0.1016

 

Language: English (United States)

 

RAID option ROM version: 11.1.0.1413

 

Driver version: 12.8.0.1016

 

ISDI version: 12.8.0.1016

Storage System Information

 

RAID Configuration

Array Name: SATA_Array_0000

 

Size: 1,430,819 MB

 

Available space: 1 MB

 

Number of volumes: 1

 

Volume member: Volume_0000

 

Number of array disks: 3

 

Array disk: 9VMLK2SP

 

Array disk: 9VMLNN47

 

Array disk: 5QM41C18

 

Disk data cache: Enabled

Volume name: Volume_0000

 

Status: Verifying and repairing 11% complete

 

Type: RAID 5

 

Size: 953,878 MB

 

System volume: Yes

 

Data stripe size: 128 KB

 

Write-back cache: Write through

 

Initialized: Yes

 

Parity errors: 1

 

Blocks with media errors: 0

 

Physical sector size: 512 Bytes

 

Logical sector size: 512 Bytes

Hardware Information

Controller name: Intel(R) Desktop/Workstation/Server Express Chipset SATA RAID Controller

 

Type: SATA

 

Mode: RAID

 

Number of SATA ports: 6

 

Number of volumes: 1

 

Volume: Volume_0000

 

Number of spares: 0

 

Number of available disks: 0

 

Rebuild on Hot Plug: Disabled

 

Manufacturer: 32902

 

Model number: 10274

 

Product revision: 0

 

Direct attached disk: 9VMLK2SP

 

Direct attached disk: 9VMLNN47

 

Direct attached disk: 5QM41C18

Disk on Controller 0, Port 0

 

Status: Normal

 

Type: SATA disk

 

Location type: Internal

 

Usage: Array disk

 

Size: 466 GB

 

System disk: No

 

Disk data cache: Enabled

 

Command queuing: NCQ

 

SATA transfer rate: 3 Gb/s

 

Model: ST3500418AS

 

Serial number: 9VMLK2SP

 

SCSI device ID: 0

 

Firmware: CC38

 

Physical sector size: 512 Bytes

 

Logical sector size: 512 Bytes

Disk on Controller 0, Port 1

 

Status: Normal

 

Type: SATA disk

 

Location type: Internal

 

Usage: Array disk

 

Size: 466 GB

 

System disk: No

 

Disk data cache: Enabled

 

Command queuing: NCQ

 

SATA transfer rate: 3 Gb/s

 

Model: ST500DM002

 

Serial number: 9VMLNN47

 

SCSI device ID: 1

 

Firmware: JC4B

 

Physical sector size: 512 Bytes

 

Logical sector size: 512 Bytes

Disk on Controller 0, Port 2

 

Status: Normal

 

Type: SATA disk

 

Location type: Internal

 

Usage: Array disk

 

Size: 466 GB

 

System disk: No

 

Disk data cache: Enabled

 

Command queuing: NCQ

 

SATA transfer rate: 3 Gb/s

 

Model: ST3500320AS

 

Serial number: 5QM41C18

 

SCSI device ID: 2

 

Firmware: SD1A

 

Physical sector size: 512 Bytes

 

Logical sector size: 512 Bytes

ATAPI device on Controller 0, Port 4

 

Location type: Internal

 

SATA transfer rate: 1.5 Gb/s

 

Model: PIONEER DVD-RW DVR-216R

 

Serial number: Not ...
0 Kudos
Silvia_L_Intel1
Employee
1,509 Views

There is a possibility that the reason why you are getting is inconsistency is because the three drives have different specifications such as the "Sustained data transfer rate OD", the "Internal data transfer rate" etc.

Check these links for additional information:

HDD ST3500320AS PAG 8 http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.11/100452348g.pdf http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.11/100452348g.pdf

HDD ST500DM002 PAG 11

http://static.highspeedbackbone.net/pdf/Seagate%20Barracuda%20Internal%20Hard%20Drive%20Manual.pdf http://static.highspeedbackbone.net/pdf/Seagate%20Barracuda%20Internal%20Hard%20Drive%20Manual.pdf

HDD ST3500320AS PAG 6

http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.12/100529369b.pdf http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.12/100529369b.pdf

It will be a good idea to get in contact with Seagate, and verify if these 3 drives are compatible for a RAID5 since there are some differences between them.

0 Kudos
TGold1
New Contributor II
1,509 Views

Thank you Sylvia,

That may be an issue with the different drives. However, I wasn't experiencing the issue when I created the Intel RST RAID 5 with two ST3500320AS drives and one ST3500418AS drive.

Originally, I had four ST3500320AS drives with 5-year Seagate Limited Warranty. Three of the four drives have failed and Seagate sends a remanufactured replacement. You don't have any control over what they send. That's were I got the ST3500418AS and the ST500DM002.

One interesting point is that the original RAID 5 drives, ST3500320AS & ST3500418AS, were rated at SATA 3Gb/s. When one of the ST3500320AS drives in the array failed, I replaced it with the ST500DM002 which is rated at SATA 6Gb/s.

I haven't received my replacement for the recently failed ST3500320AS from Seagate yet, but they will likely send a newer remanufactured SATA 6Gb/s drive.

My DX58SO can't take advantage of the SATA 6Gb/s feature.

I don't know if that is what's causing the issue. I haven't read anything in the technical documentation from Intel about RST that indicates that would be a problem. Although the drives are rated for a certain speed of data transfer, both internally and externally, they don't always transfer data at the max rate. I'm only getting indications of parity errors after a reboot. The RST verification feature finds and repairs the errors. If I run the verification feature twice with out rebooting, no errors are detected on the second run. That indicates to me that the drives are compatible in Intel RST RAID 5. Something is happening during reboot that is causing parity errors to be detected.

Please let me know if there is any technical information available to further investigate this issue. I was looking for some software utilities that would give more information about the errors that are showing up.

Regards

0 Kudos
TGold1
New Contributor II
1,509 Views

Sylvia,

I contacted Seagate as you suggested. They indicated that the Barracuda family of drives aren't tested for RAID compatibility, because the Barracuda's are intended for use as a Desktop drive. Their Enterprise level drives are tested for RAID compatibility. Seagate's main concern seems to be companies that use consumer level desktop drives in a business level/enterprise type server application. And, that the Barracuda family drives weren't intended to be in a 24/7 business or server environment.

Does Intel consider the desktop RST to be an enterprise application?

From what I recall of my study of Intel RST, Intel doesn't indicate that you must use three identical drives for a Intel RST RAID 5. The only point I can recall is that if you replace a drive, that replacement drive must be at or above the capacity of the replaced drive. That's a clear indication that the drives don't need to be identical.

It would be helpful if a more advanced and robust utility was available that could yield more information about the nature of the parity errors. Why do they only occur after a reboot? What type of repair is the RST verification feature performing? Intel may have more technical information that isn't available to consumers.

There's nothing published from Seagate or Intel that indicates my present configuration is susceptible to issues or problematic.

Regards

0 Kudos
Silvia_L_Intel1
Employee
1,509 Views

You won't find anything stating about the type of Hard Disk Drives required to use the Intel® Rapid Storage Technology because this is a tool created to manage RAID arrays within Windows.

It is an industry standard to use Hard Disk Drives optimized for RAID configurations, this is not just because you are using an Intel storage controller, and it is well known practice since RAID technology was developed.

Intel also has an Intel® Rapid Storage Technology enterprise version used for server workstations.

0 Kudos
TGold1
New Contributor II
1,509 Views

The Intel RST for my Intel desktop board using desktop Seagate Barracuda drives was working without issues until I had a drive failure. The good news and bad news is that the replacement drive I used to rebuild the array has just failed a Seatools for DOS long test. It passed a short test. I had tested the three drives with Seatools for DOS when I first began investigating my parity errors. All the drives had passed all the tests.

I suspect the Intel RST verification program detects the early indications of drive problems, before they arise to the level that they can be detected by the operating system or the threshold for test failure in Seatools. I noticed an Event 11, Disk error in my Event Viewer Windows Log for the System yesterday. The description indicated the driver detected a controller error on \Device\Harddisk1\DR1. The replacement disk is on SATA port 1. I ran the Seatools for DOS again and the drive failed.

I'll replace the drive when the warranty replacement arrives. The RST program hasn't kicked it out the array yet. If the parity error issue is resolved with the second replacement drive, then everything was and is working as intended.

0 Kudos
Silvia_L_Intel1
Employee
1,509 Views

Perfect.. Let me know the results.

0 Kudos
TGold1
New Contributor II
1,509 Views

Hi Sylvia,

I'll post the results as soon as available.

Do you have any information on what is normal behavior for the RST program and parity errors in RAID5? Is it normal to have parity errors when you run the verification program daily? If so, how many parity errors are considered normal and how many should be cause for concern?

As I've mentioned, before my recent drive failure, parity errors were very few and infrequent.

I've had several telephone discussions with Seagate tech support. Per their Desktop HDD Data Sheet, the SATA 500GB ST500DM002, Model Name formerly Barracuda, lists Desktop RAID as a Best-Fit Application. I don't think the different drives are an issue, but I am going to change the drive models in my present RAID5 array to a ST500DM002 on SATA port 0, ST500DM002 on port 1, and a ST3500320AS on port 2.

Before I insert a new drive into the array, I'm going to prep the drive by running Seatools for DOS long test and full erase. The full erase will identify and remove from service any bad physical sectors and reassign replacement physical sectors if available. I'll then complete a full format in Windows and run Windows Check Disk with both options checked. After the array rebuilds, I'll run the verification program until I get a good run with no parity errors detected and then pull another disk out of the array and repeat the new drive prep until all drives have been checked and serviced for bad physical sector issues.

If that procedure doesn't resolve the parity error issue. I'll full erase all the drives with Seatools for DOS and do a clean install of the Win 7operating system. If a clean install of the OS doesn't resolve the issue, I'll replace the drive on SATA port 2 with a new ST500DM002 so all the drives will be identical and hopefully running the same firmware.

If I do a clean install of the Win 7 OS, I'll again try to install the Download Center recommended RST 10.1.0.1008. It wouldn't install without freezing my system when I upgraded to Win 7 x64, with a clean install from Win XP x64. I have the final version AA number DX58SO now, that I didn't have when I attempted to install that driver the first time on Win 7.

Regards

0 Kudos
TGold1
New Contributor II
1,509 Views

Hi Sylvia,

I'm not completely sure about all the sources or causes of the parity errors that appeared with regularity and varying quantity, but there was a problem with the physical sectors on at least two drives.

I removed the drives with physical sector issues from the RAID-5, but I was still getting a few parity errors.

I then disabled Windows write-cache buffer flushing and set the Intel RST program to cache mode: write back.

The parity error issue is resolved. I can now reboot and run a verification and repair and get zero parity errors. I also get zero parity errors on subsequent runs of the verification and repair program.

As a general maintenance procedure for Seagate Desktop SATA drives in a Desktop RAID, I recommend periodically removing a disk from the RAID-5 and performing a full erase and long test with Seagate Seatools for DOS. This task should be completed with the BIOS set to AHCI and only the one disk attached to a SATA port. Then reset the BIOS to RAID connect the other disks to their SATA ports, boot to Windows, initialize the disk, setup a partition, full format, run Windows Check Disk with all options checked, and then allow the RST program to rebuild the RAID Volume. Then repeat this process with the next disk until all the disks have had a full erase and long test with Seatools for DOS.

The Seagate Seatools for DOS full erase and long test with the BIOS set to AHCI is the only test that can find and fix, if possible, problems with the array disks' physical sectors.

These problem do not reveal themselves any other way. I had one Seagate SATA disk that passed a Seagate Seatools for DOS short and long test, SMART reporting healthy, would format and successfully pass Windows Check Disk, but wouldn't pass a Seatools for DOS full erase. Although the SMART was reporting healthy, a utility to view the SMART data indicated the disk had SMART issues and was beginning the road to disk failure.

Regards

0 Kudos
Reply