Dynamic disks import to a new server

Question

One of our servers has died recently. So I've took those disks from it and installed them all to the new server running Windows Server 2008. Along with disks I've moved two simple two-port controllers on a Silicon Image SiI 3132 chip.

So here is a problem. There was two software Windows-driven RAID5 arrays. After reinstallation those disks plugged in thought controllers are shown offline or foreign in diskpart utility. I could not import them because they are reported somehow as readonly. If, after all, I clear readonly flag from those disks and import them, diskpart reports that array is OK and resyncing. And then, soon or late, I'm getting BSOD of some random error (it was disk manager only once) and Windows won't boot.

I've returned to last known configuration several times and tried to trace an error, but I could not. Any thoughts besides hardware fault in controller, guys?

PS. Here is system log entries IMHO related:

• An online volume in an online pack has no name. This is incorrect. Error code: 1@020A000F
• VDS received bad data from a provider. Error code: 1@020A0005
• The previous system shutdown at 2:20:16 AM on ‎5/‎28/‎2009 was unexpected.
• The computer has rebooted from a bugcheck. The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff80001edc125, 0xfffff88004410000, 0x0000000000000000). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 052809-25209-01.

Answer 1

The two main candidates are hardware fault, and driver errors. A list of the STOP errors that occur over time is essential to establish a pattern.

If there are multiple different "blue screen" fault codes, chances are that it is a hardware error and you should probably skip to step 3 and start replacing gear. For a single error, there are three main methods for trying to resolve/diagnose further:

1. Upgrade Drivers: Find and install all the latest drivers for the main system devices
2. "Spray and Pray" with hotfixes: Search TechNet for all hotfixes that reference the particular STOP error (e.g. 0x3b or 0x0000003b), and install
3. General kernel debug dumps and use the kernel debug tools (Debugging Tools for Windows - Native x6) to try and locate the exact culprit
4. Change hardware: start swapping the suspect components

Hardware faults are messy to accurately diagnose, particularly on a new server - it could be just about any component. I would suggest testing with the following in mind:

• If the controller was in use in the old server it's probably OK and just a driver update might solve the problem. Unless it was a power issue that caused the old server to fail...
• Faulty RAM would easily be the most common fault I've seen that causes STOP errors. Try a complete replacement, or test one stick at a time if possible.
• Don't discount a faulty mainboard/motherboard, they do happen
• Inadequate power supplies can sometimes cause irregular faults when too many disks are active

Answer 2

If you right-click on the "Foreign" disks, do you have the option to "Reactivate" or "Import" the disks? It looks like that's the first step to get Windows to recognize them, and it should be able to see the RAID array they belonged to once they're all active.

It looks like you can do all the management you need to from the Disk Management GUI, though if you're partial to the command line tool, I'm sure the command to "Reactivate" a dynamic disk is just as easy.