Devices · Arrays and servers
Arrays rarely die because of their first fault. Arrays die because of what follows: a rebuild that forces a tired member to read cleanly until it stops working. A disk is forced back in after being rejected. Disks are swapped between bays in an attempt to help. Mirroring doesn't buy as much as people think; both halves went in on the same day and have been running the exact same hours since. Sets arrive from Hull's Old Town offices, from the Grimsby docks, from a haulier's yard at Goole. All work is done off images, so once these sets land, nothing gets worse.
An engineer will take the details
0800 6890668
| What you see | What the fault is | What to try |
|---|---|---|
| Foreign Configuration Found (Dell PERC) | Disks report themselves to be members of an array that cannot be placed by the controller. | Provide no answer. Image each member. |
| 1786 — Drive Array Recovery Needed (HP) | All remaining redundancy has been eliminated; a rebuild process has been queued or started. | Turn power off; image the entire set |
| 1784 — Drive Array Drive Failure (HP) | One of the members has died | A new disk cannot fix this. |
| 1788 — Improper Drive Replacement (HP) | The disks have been fitted into the wrong sequence. | Stop. Bay order will resolve this. |
| Virtual Drive is Degraded / Offline | Parity has been lost from the volume, or the volume has gone. | Shut the server down fully. |
| 1720 — SMART drive detects imminent failure (HP) | A remaining disk within the group claims it is going bad. | Image it today. |
Every piece of gear that shows up receives a case number on the very same day it arrives. An engineer examines it and determines what is really wrong with it and there is no charge for this. We give you straight language about which files we will be able to copy from the disk and which ones we won't be able to. Then we provide you with a single price quote, in writing, based upon these findings. We don't begin working on anything until you agree to the quoted price.
Each disk is placed upon imaging hardware made specifically for it, including those the card had previously relinquished. From here, all subsequent activity occurs against the copies. Nothing ever gets written to what was provided.
The layout itself provides details regarding the operational sequencing of the disks (which drive contains which data), the stripe size, and how parity rotates. We stand up the configuration against the images created by us on our own equipment. The controller stays with you and is not needed, and under no circumstances do we ask anything to attempt a rebuild.
We build an array, then repair the file system, followed by repairing the virtual machine containers, and finally we repair the databases. Prior to releasing the data, everything is validated via a complete inventory listing.
You will not be charged while you think about this. First, you receive a list of everything that was recovered from the drive. Once you have seen the list and given us the go-ahead, the bill comes next. Files are returned to you on media purchased specifically for this job and we cover the return postage as well. Your job is kept open by us until you can view them on your personal computer.
How did single parity get such a bad name? A consumer SATA drive will be rated by the manufacturer for being unable to read one sector in every 1014 bits, about one per 12.5TB. A rebuild of a large single parity set, however, asks for much more than that from every good disk in the set, without a single stumble, and that is the point at which rebuilds stop. This number is provided by the manufacturers, not us, and whether or not it hurts enough in real life to really matter is still an open question.
At that point another drive dropped out during the rebuild on Friday, and access to the online data became impossible. The RAID was then shut down until further notice. Each of the four drives was then imaged to our own equipment. Then using three complete reads from the clean members of the stripe plus whatever could be recovered from the fourth, the works continued uninterrupted on Monday.
Get a quote first, either via phone call or through completion of our quote request form. Then send all disks, including the one(s) previously discarded by the card: use tracked & insured post to our Leeds intake laboratory, or drop them off there (Monday to Friday, 9am to 5:30pm, no appointment needed). Often, a discarded disk will contain newer versions of some stripes. We'll retrieve each block from whichever member provided it in its most pristine state.
No. Any information needed about how disks were assigned in relation to each other, how large stripes are and which way parity rotates can be found by looking at the data contained on the disks. This is bench analysis, not an attempt at guessing what happened.
Both. First, VMDK and VHDX containers and databases that store files will be copied off the array once it is standing. Afterwards, these will be mounted locally (as they would normally) and opened for review: knowing only a filename won't tell you anything.
Typically, restoring a group of several disks will take 4-7 working days. If business has stopped, let us know upon calling. We move your project to the top of our priority list, and schedule your restoration around your desired completion date.
Leave it off. There are only a few start attempts left in your failing drive and each attempt uses up a little more of what remains. Our complimentary diagnosis will determine what files are readable.