Intel Rapid Storage Technology

[idata]

I have a home-built system running the DH77EB board, Ivy Bridge i7, 8GB RAM, WIndows 8 Release Preview, Radeon HD7870 and a Samsung 830SSD.

I saw that the most recent 11.6 version of the Intel Rapid Storage Technology driver now lists Windows 8 compatibility, the prior 11.4 version worked on Windows 8 but did not list Windows 8 compatibility. I went ahead and installed 11.6, but it does not appear to start - the icon in the system tray says Intel RST service is not running, and when I click on it to launch the app, nothing happens. I checked the error log and this is what I see:

Log Name: Application

Source: Application Error

Date: 9/23/2012 3:11:55 AM

Event ID: 1000

Task Category: (100)

Level: Error

Keywords: Classic

User: N/A

Computer: XYZ

Description:

Faulting application name: IAStorUI.exe, version: 11.6.0.1030, time stamp: 0x5042b0f9

Faulting module name: ISDI2.dll, version: 11.6.0.1030, time stamp: 0x5042b0b6

Exception code: 0xc0000417

Fault offset: 0x0004d11f

Faulting process id: 0x1ad0

Faulting application start time: 0x01cd995ab619ce9a

Faulting application path: C:\Program Files (x86)\Intel\Intel(R) Rapid Storage Technology\IAStorUI.exe

Faulting module path: C:\Program Files (x86)\Intel\Intel(R) Rapid Storage Technology\ISDI2.dll

Report Id: f412be45-054d-11e2-9b7d-000272da1ace

Faulting package full name:

Faulting package-relative application ID:

Event Xml:

Next, I try to manually start the service using the Task manager, but the service starts for only a millisecond before stopping again. Checking the Event Log, and I just see simply "The Intel(R) Rapid Storage Technology service terminated unexpectedly. It has done this 2 time(s)."

Has anyone else had this issue with 11.6?

[idata]

I just installed the final, released version of windows 8. This issue with RST still exists. It did work for a second immediately after install, but has not since.

[idata]

I'm seeing the same issue, IRST service not starting due to error. If I disable Asmedia 106x sata controller from device manager all of a sudden the service will run, then if I enable the device again it will crash iastoreui.exe.

this is on w7sp1, asus m5g (z77)

[idata]

Hello,

same problem here, but i have found a solution that worked for me.

I have an MSI P35 Neo2 with ICH9R to provide 4 SATA internal ports + a Marvell 88SE6111 Controller to provide an IDE interface and an additional SATA.

My OS is Windows 7 x64, with 3 SATA disks and a SATA DVD unit. All this stuff is connected to the ICH9R and the controller is in RAID mode. No issues before today. I have recently updated the IRST to 11.6 without problems.

This morning the system was up and working, when i decided to use the Marvell connector to add another disk. After the boot i started to notice the service crash problem.

Under the device manager, i saw that the Marvel controller did not appear in the device list. I switched the the device manager to the view for connection tree, and i noticed the new disk drive was connected to an intel IDE chip (!). So i figured that IRST attempted to take control over a non-chipset controller, causing the crash.

I had to provide the correct drivers to the marvell, so i have downloaded and installed the Marvell driver 1.2.0.68 provided for the Asus P6TH Deluxe (the only one i have found, although i have a Microstar) and... YES, it works, i see the Generic Marvell 61xx in the Controllers list and now iastorsvc starts without crashes.

Sorry for my english, but i wanted to share my experience.

Bye!

[idata]

mom4751, Thanks for your work on this. My issue was similar but the controller was a JMicron JMB362 that the Intel drivers had taken control of. Once I installed the correct JMicron driver the Intel RST drivers and services loaded correctly.

Thanks again,

Carl

[MSmit17]

We are many...

[idata]

Antonio, which motherboard do you have?

Anyway the problem seems to be clearer now. Windows 7 comes with a lot of generic drivers for many controllers and shows them as Intel peripherals on the Intel based chipset motherboards.

So if you do not add a specific driver for each 3rd party controller, they are seen as Intel devices, the IRST thinks they are Intel parts and tries to take control on these ones too. But they are not really Intel parts, so crashes.

PS: IRST 11.7.0.1013 is out

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&ProdId=3409&DwnldID=22194&keyword=IRST&DownloadType=Driver&lang=ita download

[MSmit17]

Asus P8Z68 V- Pro/Gen3

Yesterday i enabled each of these devices on my Bios (cause i got disabled things that i dont use), Marvel and jmicron, then installed the drivers for each one and exactly the same crash.

[idata]

You have a very complex mainboard. 9 SATA, Lucid, Marvell, Jmicron, Asmedia... Surely one of these is the origin of your problem.

You should turn off everything you can disable through BIOS, connect the disks to the Z68 ports only and verify if IRST works with a minimal configuration.

If not, i think there is no hope: only a fix from Intel can help you.

Otherwise, try to re-enable them one at a time, until you find what is causing the problem and then look for a different driver or keep this device disabled.

I'm sorry i can't help you more than this, but my situation was much simpler.

[MSmit17]

Thanks for trying to help.

No i did that las night, enable and disable all the SATA related (marvell and jmicron) that didn't work. Lucid is software which im not using it. Asmedia is usb3 related which i wont disable. The other 6 SATA are Intel.

[idata]

Hello,

i have the same Problme under Windows Server 2012.

Sys Details:

Asus P8B-X with 6x C202 SataII RAID with 6 2,5" SATA Disks as Raid 5

IBM M1015 with 2x RAID 1 Volumes

Marvel 91xx PCIe Sata Controller for SSD System Disk

Any tips?

[idata]

Getting the same on my system. Latest drivers/software as of Dec 27 '12 on Windows 8; get the same on Windows 7 on same system.

P8Z68 Deluxe - latest BIOS (3603)

2 Samsung 830 Series 256Gb ssds, 1tb WD Blue on Intel controller

3Tb Hitachi and 1TB Constellation on Marvel controller

bunch of Barracuda XTs on a LSI MegaRAID 9260-8i controller card.

IRST worked fine with this setup up until updating to 11.6.

[idata]

This topic of the Intel RST service not running has been discussed here for a while, and I would like to contribute my experience. Actually, thanks to this forum, I was able to solve my problem.

I recently made a fresh installation of Windows due to an upgrade of my machine to SSD for the system drive.

However, after this new installation, the RST did not work anymore. The service tried to start, but was stopped and could not run.

In the forum I learned that this might be caused by RST trying to take over control of non-Intel storage controllers. So, I installed the specific drivers for the additional storage controllers on my mobo, and - RST works!

Hence, according to my experience it appears to be a problem of the additional hardware on the mobo and the (incomplete) installation of special drivers, but not a problem of Intel RST.

Here a few more details:

Mobo Gigabyte GA-X58A-UD7 V2.0

This mobo has the following storage controllers: - Intel ICH10R - Marvell 9128 - Gigabyte SATA2 controller GBB36X (JMicron chip) - JMicron JMB362 (for two eSATA2 channels) - Intel Rapid Storage (RST) version 11.7.0.1013

I had to install the special drivers for the Marvell 9128 and for the Gigabyte controller. There was no need for a special driver for the JMicron JMB362 controller.

Since I did not use the "additional driver" feature during Windows installation, I had to update the driver from the Device Manager. In this case it is important to know that as long as the controllers use the standard driver (which conflicts then with RST), they show up under "IDE ATA/ATAPI controllers" in the Device Manager. It might not be always obvious, which of the controllers is which device in the Device Manager, and I needed some trial and error, since there was a multiple choice. Fortunately, the special drivers only install on the correct device.

After installation of the special drivers, the controllers are not listed under "IDE ATA/ATAPI controllers" anymore, but under "Storage controllers" (as also the Intel SATA RAID controller). There, they show up with the correct name and the manufacturer.

Hope this may help some others who still struggle with getting the RST service to start.

UOB

Details of my system:

Gigabyte GA-X58A-UD7 V2.0Asus graphic card AMD Radeon EAH6950/2DI2S/2GD5Windows 7 Ultimate 64-bit, English1 Kingston SSD HyperX SH100S3/240 GB, connected to Intel ICH10R (system and programs)1 Kingston SSD HyperX SH100S3/240 GB, connected to Intel ICH10R (fast video editing etc.)2 x 1TB Seagate HD, connected to Intel ICH10R in RAID 02 x 1TB Seagate HD, connected to Intel ICH10R in RAID 01 x 1TB Seagate HD, connected to Gigabyte controllerAsus BluRay burner, connnected to Gigabyte controller (SATA)Samsung DVD burner, connected to Gigabyte controller (IDE)

2 eSATA channels connected to Marvell 9128

[MSmit17]

No some of us already tried that and it didnt work either. In my personal case i have a SSD as cache so i think it is somekind related with that.

[idata]

With this thread i managed to pinpoint my "problem", like uob's was the extra Via VT6330 controller of my m/b (1394 & 2port IDE), the problem is that i cant find any driver for that controller for windows 7, and i confirmed that by disabling it through bios, RST gui works ok...

Is there any way to "un-link" ("somehow") the RST from checking this controller... as to make it work ok with default MS driver?

[idata]

i am glad that it is working for you.

i have done that solution from you but it doesnt work for me =(

on both of my systems there are no "Standard Controllers" left and the problem

is still there ;(

[JPerk5]

The Rapid Storage Technology service (RST) seems to want to enumerate all SCSI controllers\disks and some cause the service to crash. In my case it was the Microsoft iSCIS driver. Disabling the driver fixed "Faulting application name: IAStorUI.exe" and "The Intel(R) Rapid Storage Technology service terminated unexpectedly".

[idata]

Similar issues which could not be solved by any one of the current releases of Intel RST.

Constant crashes, the RST console not working ,the Service starting and then failing with BEX errors (buffer oferflows) and corrupted RAID disks which occur randomly, cans cause RAID volumes to be dismounted abruptly while Windows is running (the RAID controler device is still working though, but volumes are in error and on next Boot waindows wants to reformat them as if they were empty).

Each time I need to press CTRL+I at boot to recover the volume, and it does not work reliably (even if the GUI console is not working). Recover sometimes works, but most of the time it cannot complete its work before a new driver crash and corruption occurs on one disk. These drivers are extremely unstable !

My installation:

- Intel ICH10 SATA RAID controler, connected to:

* 5 SATA II hard disks of 1TB each (used to create two RAID5 volumes of 1.8 TB, for data volumes D: and E: formated in NTFS), on ports 0-4 of the controler

* a SATA II BlueRay disk on port 5 of the controler (working apparently in AHCI mode through the Intel RAID controler)

- Gigabyte SATA III controler (in AHCI mode)

* a 512GB SATA III SSD (no RAID, want to use backups on the 2nd RAID volume)

- a JMicron SATA III / USB3.0 controler (in AHCI mode)

* connected

- processor: Intel Core i7 (4 cores x 2 threads, VTD enabled) at 3.20GHz (QPI link at 4.8GHz), no overclocking.

- RAM 24 GB (1.6 GHz), no overclocking

- Windows 7 SP1

The only solution is to not use ANY one of the IRST versions after 1.2 (all versions 11.4, 11.5, 1.6, and 11.7 have severe unrecoverable BEX errors, either in the GUI controle, or worse in the controler driver).

BEX errors (BUFFER OVERFLOW) are simply unacceptable at the driver level (which is supposed to run in fail safe mode, and that CANNOT be uninstalled when running in RAID mode). Unfortunately, the Microsoft driver does not work, and the old Intel driver 8.6.9 is dramatically slow.

Intel suffers now from low quality code with insufficient testing in its IRST drivers (supposed to be used on mission-critical servers) : their programmers should be fired, or Intel should ask to Microsoft to write and test a decent version of IRST.

I just solved MOST of my problems by finding a copy of version 11.2 (which works with the Intel GUI Console).

DO NOT USE ANY VERSION of IRST after 11.2, your data is at sever risk, and BSOD will occur randomly.

[idata]

Also IRST will crash with any other SCSI or AHCI controler installed. IRST should NOT attach to disks not connected directly to an Intel controler, or not working with the default generic Microsoft drivers for IDE/SATA/RAID : this includes iSCSI (even fom Microsoft), some BlueRay disks, some CD/DVD emulators (mounting ISO images, including the tools from Microsoft for Windows, or virtual image mounters for accelerated hypervisors).

Visibly the IRST driver attempts to enumerate all SCSI devices using a very small buffer (of 144 bytes) when the data blocks returned by drivers will frequently be more than 1500 bytes). However it seems that some SCSI drivers are correctly returning the "buffer size too small" error when calling the Win32 API, but they then return sometimes this error by indicating a buffer size which is still too small (by some bytes) to safely perform the query. When the IRST driver perform such calls, it should allocate a bit more than what is indicated by the error value returned (notably because there are also alignment contraints): ideally the driver should round up the sizes to the next memory page size, i.e. if the driver returns that the 144-byte buffer is too small and the buffer should be AT LEAST 1434 bytes, and the system memory page size is 4KB, then the driver should allocate a full 4KB page as the query buffer, not trying to allocate ONLY the suggested minimum !

Various calls to the Win32 API are not checked. When this affects the IRST management services (just used to control the recovery process in a RAID configuration) or the IRST GUI icon launcher, this is not so critical, but in the kernel-mode driver code (iaStor*.sys), this bug is extremely critical !

[idata · Accepted solution]

Intel Rapid Storage Technology 12.0.0.1083 at station drivers does seem to fix the problem, at least for me it seems to be working...

[idata]

I notice that there are much more competent contributors to this forum. Nevertheless, I would like to report a recent finding, which perhaps might help in one or the other case. For details, please see also my original report of January 14, 2013.

Today, I wanted to use for the first time also the eSATA controller on my mobo (Gigabyte GA-X58A UD7 Rev. 2.0). This is a JCMicron controller JMB362 with two eSATA connections.

However, when I attached two external harddisks (in one casing and over one eSATA cable), only one of the disks was recognized, and the Intel RST crashed again (it was running fine, before I used the JCMicron controller for the first time).

I noticed that the JCMicron controller did not have its specific driver installed, but the generic driver from Microsoft. Hence, I checked on the Gigabyte website for the proper driver. However, Gigabyte does not provide the driver for the JCMicron controller, only for the Marvell controller and the Gigabyte controller on the same mobo. This is obviously bad and not correct.

So, I had to check on the website of JCMicron directly and found a suitable driver. JCMicron mentions that the license for these drivers is "only in order to test the controllers". However, I downloaded and installed the driver for both channels of the JCMicron controller. And - first, Intel RST is running properly again. Also, both harddisks in the casing are correctly recognized. Hence, this driver is necessary, in order to properly operate the controller. I am disappointed that Gigabyte does not provide the drivers on their website.

Perhaps, it may be necessary to check for all non-Intel controllers on the mobo, even if the mobo manufacturer does not provide specific drivers, you might want to check on the website of the controller manufacturer directly. Perhaps there is a specific driver that could help you to solve the problem(s).

UOB

[BK7]

(I could not have tracked down my RST issues without the guidance I found on this forum. Thank you all).

I just updated from an older RST (version 8) to RST version 12.8.0.1016.

I immediately ran into the RST service won't start issue. (but ***SEE UPDATEs BELOW, in RED***)RST 12.8.0.1016 is apparently not fixed in regards to previously mentioned Marvell conflicts. My experience is that 12.8.0.1016 RTS will not stay started if the Marvell sata/esata is enabled in either AHCI or RAID mode, RTS will stay started if Marvell is in the IDE mode. (On my MoBo, the Marvell SATA is called "secondary SATA" in the bios). ***SEE UPDATEs BELOW, in RED***I have included a screenshot that describes my situation and my workaround.

SUMMARY for my situation: the Marvell is composed of a SATA bus and eSATA bus. When the Marvell is set to AHCI or RAID in the bios, RST will not start. My testing shows (that for me) it is the Marvell eSata bus (with or without an actual eSata HDU attached) that prevents RST from starting, while the Marvell sata bus seems to co-exist OK with RST (however, this may not be true if HDUs instead of optical devices were attached on the sata bus: untested by me). Again, see screen shot for the full picture and for my workaround.

Hope this helps someone else to save time in tracking down this issue...byron k.(image below can be clicked on to see a sharper version)

mobo: intel DZ68BC

bios: BCZ6810H.86A.0027Proc: intel core i7-2600@3.4ghzram: 8gbOS: windows 7, 64bprimary storage bus: intel desktop/workstation/server Express Chipset SATA RAID onboard controller (set to RAID in BIOS) with 1x100gb SSD system/boot drive and 3x1TB in RAID 5 for storage.

secondary storage bus: Marvel 91xx sata/esata onboard controller (set to AHCI in bios) with 2 DVD burners on the SATA bus and occasionally an eSata drive on the eSata bus.

RST v: 12.8.0.1016Marvell Driver: v 1.2.0.1012 when I wrote the original version of this post. Then I updated it to 1.2.0.1019 and that fixed it.

**UPDATE: (12-15-2013) I finally tracked down a newer version of the Marvell driver (1.2.0.1019, from the Intel website), which moved the Marvell SATA bus device from under the "IDE ATA/ATAPI" tree in device manager TO the "Storage Device" tree. Like previous poster UOB mentions, that seems to be the ticket. Once the Marvell SATA was positioned there, the RST startup conflicts went away.

Message was edited by: Byron K (on 12-15-2013) to indicate how I finally "fixed" this problem. Fixes are discussed in the RED text.