Looking costs nothing · one fixed figure after
Price your RAID recovery. Tell us the set and what it says; we will tell you what it takes.
This page is for arrays: hardware RAID on PERC, Smart Array, MegaRAID, Adaptec, Intel and the rest; software RAID on mdadm, ZFS and Storage Spaces; datastores and cluster volumes on top; and the array layer inside NAS units. Two questions give you a likely cause, the price band and the one thing not worth trying tonight. The short form below opens your case; the fuller booking sheet further down goes in the box with the drives. Either way we reply by email with the posting address, the nearest drop-off point and a sheet to print. After the free look you get one figure in writing. A set of two to four members is £500 + VAT upwards; larger sets, 15K SAS sets, storage shelves and enterprise storage start at £1,250 + VAT; and on most jobs no data means no bill.
Put it to an engineer — the first look costs nothing
0800 6890668
Two questions, and a likely cause.
The order of the job →What kind of array?
Whichever comes closest will do.
Two taps and you get a plain answer, plus the one action worth taking before any other. A diagnosis it is not; for that the members have to be on an engineer's imager. Read it as the answer the telephone would give at three in the morning, when the telephone is not there to give it.
Now the detail.
What happens after you send it.
Every band, in prose →An answer inside a working day
A person reads what you write; nothing is sorted by script. Back comes a straight first view of what is likely, one instruction about what not to touch meanwhile, the posting address, the nearest drop-off point and a booking sheet to print. We do not publish the address; it comes by email in reply to the form, and the booking sheet inside the parcel is what matches it to your enquiry when it is opened. Your case number is issued that day and emailed to you.
The bench names the fault Free
A case number goes on the parcel and a number on every member, matched to the slot you wrote on it, and each member goes on the imager its interface needs, never on a controller. The metadata is read from each, the level, the order, the event counts; the fault is named and how much can honestly be read back is settled. Allow about 48 hours for a small set; shelves take longer, and you are told which at booking-in.
The figure, and your decision
The figure arrives in writing, the findings alongside it. Accept it and work begins. Decline and the drives travel home with the postage at our end. Nothing is owed for the free look in either case, and nothing further down the page quietly undoes that.
The booking sheet: on screen, printed, or both.
Getting it packed.
Run through this before taping it
- Export the log before the server is touched: OpenManage, iDRAC, iLO, SSA, storcli /c0 show all, mdadm --examine on every member, zpool status, Get-VirtualDisk. It is the best record of what failed when.
- List the drives by slot, with model string, serial and capacity. It is the single most useful thing on the sheet, and on a shelf it shortens the job by days.
- Put every key on the sheet, or send it by email once the case is open: the PERC passphrase, the HPE encryption key, the BitLocker recovery key, the KMIP details. Encryption has no side door for a laboratory either.
- Be honest about what has been tried, in order. A rebuild that ran, a cleared configuration, an F2, a --create: each changes the plan, and on images most of it is reversible.
Worth saying twice: whatever the checker said, leave the server off and the drives in their slots. A great deal of the work on this bench was made harder by one more rebuild to see whether today might be different.
The questions that come up first.
Is the checker actually a diagnosis?
It is not. What the checker does is infer from the boxes you ticked. A diagnosis means every member on an engineer's imager with its metadata read — and only that sets a price. Take the checker's word on what to leave alone tonight; take the bench's word on cost.
How much will it be?
A set of two to four members is £500 + VAT upwards. Larger sets, 15K SAS sets, storage shelves and enterprise storage start at £1,250 + VAT. The pricing page sets it all out in prose, and the free look is what produces your one written figure.
Is no data, no bill genuinely true?
On most jobs, yes: nothing recovered means nothing to pay. Two kinds of work fall outside it — a set another laboratory has already worked on, and drives that have been opened at home. If yours is one of those you are told at the outset, well before you agree to spend a penny.
Do I send the whole set or just the failed drive?
For an array job, every member, including the one that failed first and any part-written replacement. If the set is healthy and you only want one failed member imaged so your own rebuild can run, that is the drive site's job at £300 + VAT.
How soon does an answer come?
Within one working day at the outside, and usually much faster while the desk is staffed. The free look is slower: reckon on about 48 hours for a small set once it is on the imager, longer for a shelf.
A price inside two minutes.
Checker, form, parcel, bench — and the decision is yours, made with one figure in front of you and nothing to pay for the looking.