Servers do not often fail as servers; they fail as an array (usually having lost more than one member of the redundant array). Retrieving the files is only half the job; returning a SQL/Exchange store in a state the application will accept is the other half of the job, where file level recovery falls short.
A failed restore that writes back onto the source can destroy exactly what a recovery would have retrieved. Image the array first before restoring if the backup is uncertain; you always can restore afterwards but never un-write a failed attempt.
The problem rarely lies with the server; storage below it fails (and usually more than one member).
Getting your data recovered is usually much more than just getting the files back; it’s about recovering those files so that they can be accepted by your application.
The store that contains a SQL Server, MySQL or an Exchange database cannot always be immediately used after being pulled off a failing server. Databases have many types of integrity structures within them and all transaction log information. Therefore, even though the contents may be perfectly intact byte-by-byte, they can still fail to “mount” when you try to attach them if there were transactions going on at the time of failure; or if any page(s) are missing in the middle of the file. Therefore, instead of just extracting files from a restored database and trying to use them again, the database has to go through a structural integrity check before anything else can happen with it. And, as opposed to reporting to you what pages are bad and giving you back a store that won’t attach anyway, we will give you a list of what objects are affected by those damaged pages.
The same would apply to the other files like virtual machine files, email archives and everything else that has some type of internal structure or format. The main differences in this case are a count of how many files you have versus what it takes to make them all work together as part of your system.
All standard platforms and controllers.
Dell PowerEdge, HP ProLiant, IBM & Lenovo System x, Supermicro, Fujitsu and white box systems.
Dell PERC, HP Smart Array, LSI and Broadcom, Adaptec, IBM ServeRAID, even if you’ve lost the config on one of those RAID cards.
NTFS and ReFS, ext4, XFS and Btrfs. Plus LVM and Storage Spaces that layer over the array.
The VMFS datastore and Hyper-V volume that hold the VMDK/VHDX files.
Shelves attached to servers (direct attach), iSCSI targets and NAS devices all working together as one system.
They are priced based upon how much it will cost to rebuild the data versus based upon the amount of space you have used.
We begin server/RAID/NAS recoveries at £500 +VAT. After completing your free 48-hour diagnostic, we will confirm pricing. For enterprise SAN recoveries pricing begins at £1,250 +VAT. If members require physical drive level work we take a 50% deposit, with the balance payable only once the recovery has completed successfully.
Please send every disk in the same bay order, label each, include any disks previously labelled “failed” and identify the type of controllers these were using. Also please inform us of the RAID levels you know about and what else you’ve tried so far. NDAs are standard and invoicing is available. Complete write-ups include: a RAID 5 whose controller forgot the array and an 8-disk RAID 5 riddled with bad sectors.
The server manufacturer, RAID configuration and server role will be provided to our engineers as they do a complimentary diagnostic on your drives. Once the engineers have confirmed what needs to be done to complete the project, a written quote will be made available before any work starts.
Please call us or fill out the form prior to shipping or dropping off your item so that we can provide you with a price quotation. We will need the individual hard drives (labelled in bay order) to begin recovering your data; the disks need to be removed from the server. We ask that you pack your items securely, include all contact information, and upon completion of the free diagnostic assessment, we will provide you with a written price quotation before we start working on your project.
Posting it? We would suggest using a tracked and insured delivery method. Want to drop it off? That’s okay too. Monday to Friday, from 9am to 5:30pm is good; no appointment necessary: simply prepare the item as described above.
Do you need to get a price first? We ask you to complete the form providing as much detail as possible regarding the server configuration including RAID and which disk has failed so our engineers can review your request and respond back to you with a specific quote.
We’ll be in touch shortly. If it’s urgent, call 01865 593000.
What are the common questions that people have been asking us about recovering a server?
Turn off the power to the server, leave the disks in bay order, do not allow a rebuild to start and do not recreate the array or restore from a backup onto the original volume until all of the data is safely backed up somewhere else. Almost every server that could not be recovered was able to be recovered at the time it failed.
Yes. They are inspected for structure integrity (as opposed to being just pulled out). Even though you have an exact byte-by-byte copy of your database (that’s good), it could be in the middle of a transaction; there may be corrupted pages, etc. Therefore instead of giving you back a store that will fail on attachment, we report what objects were impacted by this failure.
Yes, including any that have previously been designated as failed. Often, a member of the group has retained readable data even though it was removed from service; this could be the difference between recovering a portion of your lost data versus recovering all of it. Prior to removing them, label them by their bay number.
Yes, VMFS data stores (the virtual storage), Hyper-V volume(s) & the VMDK/VHDX files contained therein. The cost for recovering a virtual machine will depend upon whether the work is needed at the VM level or at the underlying storage array level. If the work is at the VM level of your system then the starting price will be £800 +VAT.
Yes to both, as standard. For business customers we include a dedicated point of contact. Invoicing is optional instead of having to make a payment prior to receiving service.
Server/RAID/NAS work will begin at £500 + VAT and includes a fixed fee in writing once we complete a free 48-hour diagnostic. Enterprise SAN environments will be priced independently of this pricing structure beginning at £1250+VAT.
Free to diagnose and priced on a tier beginning at £500; in most cases you only pay if we recover your data. We also recover servers and associated systems regardless of manufacturer: Dell, HPE, Windows Server, VMware etc. We encourage you to move forward with getting your recovery started today.