Drive dissapears in Windows, Error 9 in System Log

Posted on


At what seems to be at random intervals my secondary drive in my workstation just disappears. The OS is not installed on this drive (it’s my “E:” drive). I’m not using RAID.

  • The PC is a Dell Precision T3500
    running Windows 2008 R2.
  • It contains a Intel Matrix Storage
    on the motherboard
    (v8.5.2.1002). I’m running the
    windows software for that controller
    v8.9 (which I guess contains windows
    drivers – which is why I mention it).
  • The failing drive is a Western Digital WD4000AAJS-0 which is a 400 GB 7200 RPM drive SATA (firmware 12.01C01)

I notice system event log entries that seem to correlate to the times the drive goes away:
Source isStor, Event ID 9, Description:
The device, DeviceIdeiaStor0, did not respond within the timeout period.

This drive was working fine for more than a year in another PC and was moved to this machine for about 3 months. However, this problem started only a week ago.

Googling for this problem seems to point at the intel matrix controller. Yet none of the workarounds I’ve found help. (renaming a registry key, updating the matrix software, switching sata ports)

Any ideas?

The drive started to not just disappear but stay visible and deny access to some files on the drive. I ultimately copied what I could to the working main drive and disconnected it. This workstation’s primary drive is too small for the work I do so I’ll hook up other drive at some point and post the results here. (Why add 12GB of ram for hosting virtual machines but choose 250GB drives so i have no room for VHDs??)

Also I tracked down to the event log that the problem started the same day I installed Remote Desktop Services role (aka terminal services) in attempt to get Aero glass working over remote desktops to this PC (don’t ask). I ended up removing that role since it didn’t help. I wonder if some type of power management setting was flipped on/off during the add or remove of that server role?

Update 2
After a few years of running this machine with only one drive (the one dell ships) after never really resolving this issue I ended up installing a second drive again. This drive, a 2012 WD Caviar “black” (WD1002FAEX), seems to be running fine with none of these issues. I assume this issues was caused by bugs in the SATA power management of older drives with this controller/windows 2008.


The following Intel article seems to address your problem:
Intel Rapid Storage Technology : Possible issues with Windows Vista* and Intel® RAID

Some excerpts from the text:


The following issues may occur on
Windows Vista* with a supported Intel®
RAID controller hub:

  • Timeout errors reported by iastor or iastor(v) in Microsoft Event Viewer


Various desktop Serial ATA devices
such as hard drives and CD/DVD drives
have been reported that do not comply
with the Serial ATA Link Power
Management (LPM) device specification
and display erratic behavior when LPM
is enabled.


Install Intel Matrix Storage Manager
version 7.5 or later. These versions
workaround this issue by disabling
Link Power Management (LPM).

If you are experiencing installation
issues, you can also use one of the
following additional workarounds:

  • If available, install updated firmware for the device.
  • Attach the device(s) to ports 3-5 on the Intel® 82801HR/HH/HO I/O
    controller hub (ICH8R) or port 3 on
    the Intel® 82801GR/GH I/O controller
    hub (ICH7R)
  • Switch to IDE mode in the system BIOS.

    Note : This change will disable RAID
    functionality. Contact your BIOS
    manufacturer for instructions on how
    to switch to IDE mode.

I’ve had a similar problem and think I’ve found a solution.


  • Dell M6500 laptop
  • Vista 64-bit Business SP1
  • Intel ICH8M-E/ICH9M-H/PCHM SATA RAID controller
  • Intel Matrix Storage Manager
  • 2 Seagate ST9250410ASG disks
  • Raid 1 configuration

After running without problems for a week after the purchase, one day the drives started failing more or less arbitrarily. About once a day, the Intel Matrix Storage Console would report one of the drives as failed. After a reboot, the drive would reappear and the controller would successfully rebuild to the failed drive, and return to normal state.

I noticed these events in the Windows event log, correlated to the times a drive disappeared: Source isStor, Event ID 9, Description: “The device, DeviceIdeiaStor0, did not respond within the timeout period.”

I read this article:

but since I was already running a much newer version of the Intel Matrix Storage Manager, this does not solve the problem. However, after reading about the relation to Link Power Management in the article, it struck me that on the exact day that the problem started, I had switched the Windows power plan from the default “Dell” plan to the “High performance” plan.

I compared the advanced settings of the power plans, and noticed that they differ in the option PCI Express | Link State Power Management. The high performance plan has Link State Power Management turned off, while the Dell plan has it to “Moderate power savings” while plugged in, and “Maximum power savings” while on battery.

So, I switched the power plan back to “Dell”, and since then (for about a month), I’ve had no problems and no entries in the event log. I’ve been running plugged in all the time, so it’s the “Moderate power savings” setting that I’ve been using.

So, the solution seems to be to use a Windows power plan that has PCI Express | Link State Power Management set to “Moderate power savings” (or perhaps “Maximum power savings”).

put a small page file on the drive that is getting disconnected–it could be an orphan power management issue and a page file will keep the drive spun up

actually, re-reading your OP, if the drive is showing up in explorer but you can’t access some of the data, it sounds like the drive itself is failing. can you reproduce the issue if you attach a different drive to the same port?

also try replacing the SATA cable first since it’s cheaper than a new drive

Leave a Reply

Your email address will not be published. Required fields are marked *