Windows 7 has no notification of RAID failure?
Question
All replies
-
You could add a message display for any old event that IS present, then go into the Task Scheduler and change the ID number to the one you want...
This is the trigger it added for me when I tested it using a System Uptime event...
This was the Event Action it added...
I agree with you that this is a weak workaround, but at least you're not completely out in the cold.
-Noel
- Edited by Noel Carboni Thursday, September 8, 2011 5:20 PM
- Proposed as answer by Mike DePouw Saturday, February 18, 2012 6:56 AM
-
I still don't believe that Event Viewer does not log some errors. If it's in the Disk Management, it should be in the Event Viewer. Did you check here:
I'm guessing if it's not here, it should be somewhere close. In case you find it, you can attach a task to notify the user as Noel described. If it works or doesn't please let us know. Hope it helps!
Kunal D Mehta - a Windows Server Enthusiast | My first TechNet Wiki Article -
A little update:
I get a RAID 1 set going in disk management, pull out one of the drive's power cords, and wait about 2 or 3 minutes. I get the following error messages in the system log:
I figure now all I do is r-click one at a time on the 3 distinct errors and do the following:
- Choose "Attach Task To This Event..."
- name it > Next
- Next
- Choose "Send an e-mail" > Next
- Fill out the From: email, the To: email (my alerts email address which goes to my IT dept. phones/emails), Give it a subject with the computer name and the event type, and put the ip of a mail server on my network that accepts relays from this host (a smart host). > Next
- Check the "open the Properties dialog for this task when I click Finish" box and click "Finish"
- Set the task to run even if the user isn't logged in
- Check the box for run with highest privileges
I'm rebuilding the array ("resynching" in Microsoft-anese). After that's done, I'll pull a drive and see if I get some emails.
Not ideal for the home user, but exactly what I was hoping for as an admin...
-
Understood. I originally had pulled the power on the secondary drive to get the errors. I let the array rebuild, created email tasks out of the 3 distinct errors.
Now I've pulled the power from the primary drive. It generated errors (screen clip below) and sent me the emails I told it to.
I guess the next step is let it rebuild, power down, unplug a drive and boot - In the real world, if your primary boot device is missing, you'll not be an idiot and you'll know there's a problem, so I only see merit in unplugging the secondary and booting. I'll see if that generates any knowledge of a degraded array.
I'll let you know after I try it.
-
Well - I unplugged the secondary and booted. Got 2 popups, none that were disk errors, but they're pretty descriptive. It took about 5 minutes for them to come up:
In a perfect world, this would be more descriptive and throw an entry of "error" level into the system log. It does serve the purpose though.
-
Well - I unplugged the secondary and booted. Got 2 popups, none that were disk errors, but they're pretty descriptive. It took about 5 minutes for them to come up:
In a perfect world, this would be more descriptive and throw an entry of "error" level into the system log. It does serve the purpose though.
Pop up messages? It's like you are running a completely different build of Windows. My PC was absolutely silent when a drive failed...
In the real world, if your primary boot device is missing, you'll not be an idiot and you'll know there's a problem, so I only see merit in unplugging the secondary and booting.
If the primary drive fails, Windows 7 will automatically boot from the secondary. It does advise you when booting, but then times out and continues into Windows. There was no notification in Windows. So if I boot the machine remotely and I'm not sitting at the console screen, I wouldn't see the notification.
-
Ok. So RAID failure does invoke an Event in Win7 even though an obscure one, i.e. Event 26.
It is also supposed to throw a popup, but doesnt for some reason which needs more investigation.
This feedback is already in place for Win8 and you can see the following event in Win8 dev preview. Please note that it is an unfinished product.
When the copies of a fault tolerate volume get out of sync, Windows logs event ID 142 from volmgr which looks like this:
Sumesh P - Microsoft Online Community Support- Edited by Sumesh P Friday, November 4, 2011 4:50 AM edit tech info
-
FYI, some comparitive test results with the same drives, comparing benchmark performance of varying size reads and writes between an Intel-controlled SATA II port on a Dell Precision T5400 motherboard and the same drives on two SATA III ports of a Highpoint 2720SGL controller. Note that the Intel-controlled RAID equalled or outperformed the PCIe controller solution up to 4 kbyte reads/writes. Only the larger I/Os showed the PCIe board's increased capacity with SATA III connections.
-Noel
Detailed how-to in my eBooks:
Configure The Windows 7 "To Work" Options
Configure The Windows 8 "To Work" Options- Edited by Noel Carboni Monday, February 25, 2013 4:22 PM formatting
-
A demo of the script working with BGInfo.
In theory this should work with any RAID - software, motherboard, hba - this due to the VDS subsystem support that enabled diskpart.exe and other tools.
This is the result with Windows 7/2008
This is the result with Windows XP/2003
It's just a guess, but I think 2.0 TB is the practical limit for XP/2003 - unless GPT loader from Paragon can extend that to larger volumes. But access to larger GPT spaces, 3 or 4 TB, does not guarantee access as a single volume. Disk spanning and other scenarios aside. I tend to think its best to leave well enough alone and move on towards Windows 7 for the next decade. 8/2012 concerns me at the moment, it feels like a test version that may not stick around, so I'm not too invested in looking at it for now.
To be clear, just because this used the vbscript above, nothing precludes using Powershell to access the COM interface of VDS directly. And in doing so access the finer details about status, even rebuilding status. The most advanced scenario for scripting might be to register an event that triggers an update to the BGInfo display.
-- John Willis
- Edited by jwillis84 Monday, February 25, 2013 5:50 PM
