Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 357
  • Last Modified:

Recovering from partial RAID failiure?

I'm building a PC to be used exclusively as a small fileserver for an office of 3 (perhaps ultimately 4) people who use standard Office 2003 apps Word, Excel and Outlook in their daily work.  The office already has a 100mbps wired network, and I'm going to run just Windows XP Pro on the new machine, sharing folders with different permissions to give public and private storage on the fileserver for documents so they can be backed up centrally.

A lot of regular motherboards now are offering SATA RAID controllers, so it is affordable to create a mirrored RAID configuration with two identical drives for reliability.  That's great ... but having already done that on my own PC, once the drives are set up using the RAID's BIOS, it all behaves like one drive.

If someone was to go wrong with one of the drives, how would I actually use the remaining drive to keep the machine and data alive? Say one hard drive starts squealling and fails, would the PC operate on the remaining disc until I get back to their office with a replacement?  Would there be an option to add a new disc and 're-sync' the data onto it so the mirroring can continue?

(Not having bought a specific product for them yet, I'm happy to be guided by you [but would be naturally inclined towards Asus motherboards otherwise], but am really looking for general principles rather than specific instruction).
2 Solutions
Basically mirroring means that the data is written simultaneously to both Hard Disks, in the event of a drive failure this allows continued operation using the other drive, until you restore the data to a new drive using the existing copy, usually while the copy is being rebuilt your I/O performance will suffer.

With the cheap, builtin raid hardware, and a drive fails, The PC will keep on running like normal. Only, during bootup you'll get a message warning you that the raid is "broken". Once inside Windows, and if you have loaded the software of the raid controller, you'll also get a similar message, and probably an icon turning red in your taskbar.

When you change such a drive, make sure you first shutdown the PC. Most of these systems don't allow for "hotswapping". Once the new drive is in your system it may or may not be automatically rebuilt. This depends on the raid controllers firmware and software. If it doesn't automatically get rebuilt, you will have to start up that utility and use it to start the rebuilding of that array.
Just make sure that the board you buy has the raid controller in the bios, not as an addon. I had a disastrous experience with an Asus board which had a Silicon image controller to run the SATA Raid. Basically it would corrupt data regularly on shutdown and would always fail if the power failed. I quickly swapped it for a Soltek board and haven't had a problem. These early issues should be resolved by now.

Chris B
Get free NFR key for Veeam Availability Suite 9.5

Veeam is happy to provide a free NFR license (1 year, 2 sockets) to all certified IT Pros. The license allows for the non-production use of Veeam Availability Suite v9.5 in your home lab, without any feature limitations. It works for both VMware and Hyper-V environments

I have had the same setup (PATA instead of SATA) for a file server for one of my clients.
Most of the major motherboard manufacturers use the PROMISE family of chips. If you have a 2 drive mirrored array, should 1 drive fail, your computer will prompt you to press F1 at the next reboot. You will then have the option to see which drive has failed, and you can replace the drive. On the next reboot, you will press F1 again, and assign the new ("free") drive to the mirrored array. After that the controller will rebuild the array by copying the data to the new drive. On a 120GB WD PATA drive this took about 45 minutes.
Some other things to keep in mind:
1) Your motherboard driver CD should have an array monitoring tool for the RAID controller. If a problem should occur, the array management software can alert you to a failed drive. Some of the retail cards come with software that can e-mail you should a drive fail. Your mileage may vary with the OEM software.
2) Enable SMART on each of the drives to detect problems that could indicate imminent drive failure.

Hope this helps!
danieloneillAuthor Commented:
Thank you to all who contributed.  I guess that I have been lucky so far and haven't had a RAID configuration fail yet - it didn't occur to me before reading your replies that I could sabotage the configuration myself by disconnecting one of the drives to simulate a 'total failiure' of the drive, and observing the machine's actions.

Hot swappable drives aren't too important, the office is open 9am - 5pm and from the responses here, the machine would happily survive until the next day that I could go with a replacement drive.  Extra protection compared with a conventional one drive machine is what I was after, 24/7/365 uptime even during failure isn't really needed.  Thanks to all once again.
thanx too.
Thanks and good luck!

Featured Post

Free Tool: Port Scanner

Check which ports are open to the outside world. Helps make sure that your firewall rules are working as intended.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now