We have an ASUS Z10PA-D8 motherboard configured as an Intel Raid. As far as I can tell we have the latest drivers from ASUS using the IRSTe. ioStorA,sys v 220.127.116.114
After replacing a failed drive in a RAID 1 array the server Blue Screens with a Bug Check 0x000000d1 during the array rebuild. The server is running Windows 2012 R2.
Does anyone have recommendations for how to rebuild the RAID1 and if there is support for newer drivers?
Thank you for contacting Intel Communities.
In this case, for C220 chipset you may use Intel® RST:
https://downloadcenter.intel.com/download/25091/Intel-Rapid-Storage-Technology-Intel-RST- Intel® Rapid Storage Technology (Intel® RST) Version: 18.104.22.1683 (Previously Released) Date: 6/4/2015
For C610 chipset, you can check the following (Intel® RSTe):
https://downloadcenter.intel.com/download/25393/Intel-RSTe-AHCI-SCU-Software-RAID-Driver-for-Windows... Intel® RSTe AHCI & SCU Software RAID Driver for Windows* Version: 22.214.171.1243 (Latest) Date: 10/2/2015
Please let me know how it goes.
Thanks for the response. I believe the motherboard supports the C610 chipset, and according to ASUS the latest drivers are version 126.96.36.1994 which are what is installed. Why would Intel's site show older drivers as the "latest"?
From ASUS's site:
Version 188.8.131.524Description-C610 INTEL RAID
-Intel Rapid Storage Technology enterprise (WHQL) SATA/sSATA AHCI Controller Miniport Driver V184.108.40.2064 for Windows 7 64 bit, Windows Server 2008 R2 64 bit.
-Intel Rapid Storage Technology enterprise (WHQL) SATA/sSATA AHCI Controller Miniport Driver V220.127.116.114 for Windows 8 64 bit, Windows Server 2012 64 bit, Windows 8.1 64 bit, Windows Server 2012 R2 64 bit, Windows 10 64 bit.
-Intel Rapid Storage Technology enterprise (WHQL) SATA/sSATA RAID Controller Miniport + RST Filter Driver V18.104.22.1684 for Windows 7 64 bit, Windows Server 2008 R2 64 bit.
-Intel Rapid Storage Technology enterprise (WHQL) SATA/sSATA RAID Controller Miniport (without RST Filter Driver) V22.214.171.1244 for Windows 8 64 bit, Windows Server 2012 64 bit, Windows 8.1 64 bit, Windows Server 2012 R2 64 bit, Windows 10 64 bit.
-Intel Rapid Storage Technology enterprise Utility V126.96.36.1992(Requires Microsoft .NET Framework 3.5 Service Pack 1) for Windows 7 64 bit, Windows Server 2008 R2 64 bit, Windows 8 64 bit, Windows Server 2012 64 bit, Windows 8.1 64 bit, Windows Server 2012 R2 64 bit, Windows 10 64 bit.
-Command Line Utility for command prompt under Windows 7 64 bit, Windows Server 2008 R2 64 bit, Windows 8 64 bit, Windows Server 2012 64 bit, Windows 8.1 64 bit, Windows Server 2012 R2 64 bit, Windows 10 64 bit.File Size337.93 MBytes 2016/05/05 update
Actually, I provided an older driver because I was afraid of any compatibility issues, but seems that the OEM made the last version compatible with your board.
The direct link to the latest driver (the one you already have) can be found https://downloadcenter.intel.com/download/26224/-Intel-Rapid-Storage-Technology-Enterprise-NVMe-Inte... here (downloadcenter.intel.com).
Please let me know if you need further assistance.
So the version on the driver link you provided shows a newer version but will not install on our system. So based on all the available info I have we have the latest drivers.
This leads me to ask if we should be using a RAID 1 for the boot drive with SSDs.
Are only Intel SSDs supported or can I use any?
Are there specific features of an SSD that would cause the crash I experienced when trying to rebuild a RAID 1 array?
Should we be investing in a separate RAID controller?
Any SSD from any SSD manufacturer should allow you to create RAID 1 volume so you may want to check with the SSD manufacturer if there is any issue when it comes to using SSDs in RAID 1 volume.
In this case, hardware RAIDs are more robust, our chipsets come with software RAID solution which is more limited.
It could happen that the OEM have modified the chipset somehow, but you have the latest version available.
I was able to recover the RAID 1 array, just not in the conventional means.
1. Perform an image backup of the working disk
2. Shut down the server and insert a replacement disk
3. Boot the server and press Cntrl-I to load the Intel Raid utility
4. Reset both disks to Non-Raid
5. Create new RAID 1 volume using the two disks (wipes both disks). Shouldn't the utility ask you which disk to use for the RAID1?
6. Reboot and restore the disk image
7. Reboot and server came up normal
Let's hope the RAID 1 will remain stable.
Thank you for the update.
If you create the RAID 1 from Ctrl + i menu, all the information in the disks will be lost but, from the same menu, you should be able to choose the disks you want to use for the RAID 1.
Please let me know if you need further assistance.
I am sorry to report that the RAID1 array has degraded again.
At 10:57 an event # 0 with source IAStorDataMgrSvc indicating that SATA SSD on Controller 0, Port 1:Failed. Just before that, event id 833 was reported indicating that SQL Server encountered an I/O request taking longer than 15 seconds to complete. A few seconds before these errors were logged Event ID 129 was recorded with a source of iaStorB, indicating a "Reset to device, \Device\RaidPort0 was issued"
It would appear that there is something wrong with SATA port 1, or these drives are somehow incompatible with the Intel RAID system.
I "think" I have finally gotten to the bottom of this, although time will tell.
To review, we have two RAID1 arrays located on this server. The first consists of two Samsung 850 EVO SSD drives (250 GB). The second being two 1 TB Seagate SATA HDD. We have never had a problem with the second array. When a drive failed on the first array, replacing and rebuilding the array caused a Stop error consistently.
I thought this might be caused by the Link Power Management, so I added registry settings DIPM, LPM, LPMDSTATE to Port0-7 under iaStorA and B. I also found settings in the BIOS where Link Power Management could be disabled. After restarting with LPM disabled, the rebuild blue-screened again.
Finally, I found an article discussing something similar. https://blogs.technet.microsoft.com/kevinholman/2013/06/21/event-id-129-storachi-reset-to-device-dev... Event ID 129 – storachi – Reset to device, DeviceRaidPort0, was issued. – Kevin Holman's System Center Blog
Looking at the device manager, I disabled the write caching on the first drive (boot/os) (see below). I successfully rebuilt the RAID1 with a blue-screen!
So the question remains, is this expected using the built in Intel RAID hardware?
I left the Write Cache enabled on the Seagate HDD Array but am wondering if it should also be disabled?
As I said in a previous post, time will tell. Unfortunately the server restarted with a Bug Check (see below), and now the RAID1 is broken. Suffice it to say, this is very frustrating and it seems that the whole purpose of mirroring drives is negated by this problem.
Log Name: System
Date: 5/4/2017 6:52:02 PM
Event ID: 1001
Task Category: None
The computer has rebooted from a bugcheck. The bugcheck was: 0x000000d1 (0x0000000000000000, 0x0000000000000002, 0x0000000000000000, 0xfffff8005aaf5b73). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 050417-13812-01.
0x000000d1 (0x0000000000000000, 0x0000000000000002, 0x0000000000000000, 0xfffff8005aaf5b73)