Dell RAID Data Recovery | Multiple Drives Offline

Many of us have loved our Dell computers over the years.  As a major producer of quality laptops and desktops Dell has distinguished itself as a quality brand name that millions rely on.  Even their small business servers come with great features, hardware, and support.  However there is one important area in which dell servers have come up short of their normal quality standard and that has to do with their RAID controllers.Dell RAID controllers seem to often be sup-par when compared with the really solid controllers used in many other small business servers made by Adaptec and other brands.  Typically Dell uses their own build of LSI re-branded RAID controllers.  And while LSI isn’t necessarily a terrible brand, when it comes the reliability needed for critical storage, they often come up short.

One common failure of these controllers commonly occurs after a power failure.  The system for some unexplained reason will mark several of the drives as “failed” and take them offline.  And in many cases the drives/data is actually perfectly fine on the hard drives, but the controller will not allow them to be put back into the array.If it were only one drive doing this, as simple replacement could allow the array to rebuild.  However, often more than one will simultaneously be flagged as defective and the array can’t be rebuilt in this case (at least not by the RAID controller).  This seems to be an almost strictly Dell issue, and so it’s likely that the issue lies within their own custom firmware built into the controller.  The people at Dell probably figure it’s best to flag a failing drive earlier rather than later (which would seem to make sense) however the sensitivity is much too high and causes RAID arrays to “fail” that haven’t actually failed.

So how do you go about getting the data back on a Dell server that the RAID array has failed and multiple drives have gone offline?  Below I’ve outlined the basic steps to attempt recovery.  Please note that Data Medics takes no responsibility for the outcome of DIY data recovery attempts even if you are following our advise, so if you data is important please bring it in for professional recovery.

1. Clone each hard drive to another healthy hard drive using GNU ddrescue in Linux – You might be tempted to skip this step.  DON’T, it’s extremely important that you first clone the drives.  Very likely there is at least one drive in your RAID that has bad sectors and may fail completely if you skip this step.  While there are many options out there to clone a hard drive, ddrescue in Linux is the only one that stands a fighting chance to handle hard drives with multiple bad sectors.  If any drives can’t be cloned in ddrescue, likely you’ll need professional recovery to repair the hardware to a point where the data can be extracted.
2. Connect the cloned drives to a workstation using USB adapters or SATA
3. Build a virtual RAID array using R-Studio data recovery – This isn’t an easy process if you haven’t done it.  It requires you to find the correct drive order, array offset, stripe size, parity layout and more to build a functioning array.  With some experience this process comes down to a bit of trial and error to deduce the correct settings.  For a professional, seeing the data in hex from each drive it’s possible for many of the parameters to be determined right away greatly speeding up the process.  The software does have a feature to “auto detect” the array settings, however it rarely ever works.  But it might be helpful to get a starting point before beginning to manually adjust settings.
4. Extract the data onto another hard drive – Never attempt to save the data back to the drives you’re recovering the data from.  Any writes to the drives (especially if the settings are wrong) will destroy other data.  So it’s critical that you have another large drive available to at least temporarily store the data.  You can also create an image of the entire array if you’d like and later use it to recreate the array just as before (helpful if needed to boot the server).
5. Create New RAID Array – after you are certain you have all the necessary data (and it’s good data, not shifted or improperly striped) you can then create a new array using the old (still good) hard drives and/or new drives.  Then copy the data back over to the new array.  If you created an image of the entire array you can boot into a linux live build and again use ddrescue to burn the image back to the array and hope it’s a clean enough copy of the data to boot the system.

Data Medics in Cranston, RI offers free evaluation and can get a quote to recover the data from your failed Dell server RAID.  We’ll provide an upfront quote that will not change, and we guarantee that we’ll get your data back or else our efforts are absolutely free.  We are a full service data recovery laboratory and can even handle completely failed drives in house.Even if you’re far away from Rhode Island, we only need the hard drives to be able to get the data back.  Many customers as far away as California mail us their hard drives and even RAID arrays for recovery.  Give us a try and see how affordable professional data recovery service can be.

Check out our RAID Recovery Page to read more or visit our Prices page to see our current rates.  When you’re ready to move forward, visit the contact page.