When a SAN experiences loss of a storage pool, a LUN failure, or both controller failures, it can bring down the entire virtual environment or database platform simultaneously. However, in most instances, the data still exists across the individual disks, and through proper reconstruction of the pool and LUNs away from the damaged SAN, recovery is possible. We provide internal recovery services for Dell EMC, HPE, NetApp, IBM, etc., SAN-based solutions to organisations throughout Belfast and Northern Ireland.
When you have a failed SAN, doing a force rebuild, a forced expansion, or a pool reimport can over-write the exact information needed to be able to recreate the LUNs. Enterprise arrays lose data forever when they do a rebuild to a bad disk, or when a controller reinitialises a pool. Record how your disks were ordered in relation to each other (disk number & shelf), cease making any changes to the array immediately and allow us to image the disks before we begin.
Failures within a SAN occur at multiple levels; the individual pools, LUNs & controllers can all fail, and the failure cascades into the hosts above them. This is where our company recovers the most.
We are able to recover any enterprise SAN platform across both Fibre Channel and iSCSI interfaces, as well as all possible RAID and pool configurations used by each manufacturer.
Dell EMC, HPE, NetApp, IBM, Hitachi, Pure Storage, Fujitsu, Lenovo, Infortrend and Promise, with support for Fibre Channel and iSCSI interfaces; RAID 5, 6 and 10 configurations; thin-provisioned and deduplication enabled pools; dual-controller arrays; and VMFS data stores, from single-shelf systems through multi-shelf systems.
The process of recovering SAN systems is essentially enterprise RAID reconstruction on an enormous scale. We read each individual disk independently. Then, using virtual images that are read-only, we rebuild the original pool configuration and rebuild the RAID groups. Next, we provide access back to the customers to their LUNs and data stores. This way, even if there was a failure during the rebuild process and/or one or more controllers died before completion, there will be no opportunity for any data overwrites.
For each customer's system, we review each disk individually. We also verify which RAID group(s) exist within that pool; which LUN(s); what data stores may have existed; and which components failed first, and in what sequence, prior to sending them our written quotation.
Each failing disk is repaired or copied first (heads, board or firmware) before any other disks are used in the recovery process, so no disk is put under strain.
All disks are copied in a read-only state, and all work is done on the copies. Therefore, there is no danger of the original disks being modified during the process, nor will the disks making up the array be modified.
The RAID group layouts; pool layouts; as well as the thin-provisioning map are created by imaging the disks (in read-only) and then "reassembled" virtually to create the same pool and its associated LUNs as if you had not experienced a disaster.
Once your pool has been restored, each of the LUNs can have VMFS datastores, virtual machines, databases, files etc. extracted for recovery purposes. Where possible damaged files/data will also be repaired.
Prior to returning any recovered data to you we check to ensure the recovered data can open and function correctly. We will also allow you to review the results of our recovery efforts prior to transferring the data.
In addition to having access to your recovered data on storage appropriate to the size of the recovery effort, secure transfer options may also be available to suit your environment.
Whether recovering from an offline storage pool, a deleted LUN, dual controller failure or corrupted VMFS datastore, we have experience with recovering SAN arrays made by any manufacturer and rebuilding your pool using read-only images and presenting your LUNs and virtual machines.
Give us some information regarding what happened to cause this issue, and an engineer will contact you. Usually within a single business day.
We will get back to you soon. If it is urgent, call 028 9002 0144.
Free evaluation, followed by a written quotation (fixed price) prior to starting the restoration process for SAN recovery. The minimum cost for recovering from a SAN failure is £1,250 + VAT, which will be based on the amount of arrays in the SAN environment and the configuration (disks, RAID, etc.) that requires recovery. In most cases you only pay if we recover your data; however, if you require physical repair on one or more of the failed disks, a deposit of 50% of the total cost will be required for the replacement parts and/or labour costs associated with repairing the disks. Once successful completion of restoring your data has been completed, the remaining 50% will be due.
The various types of SAN arrays we have recovered from (across different platforms and fault types) are demonstrated here with their configurations and outcomes; the specific customer names are protected for confidentiality purposes.
Many disks within a single RAID group had failed. We then took images of all the disks that were part of the pool. We then restored all the RAID groups virtually and restored the LUNs and their corresponding virtual machines.
Both of the controllers had failed and therefore the LUNs were no longer available. We could still access the disks as individual disks and rebuild the entire disk pool. Then we were able to recover the previously presented volume(s).
One production LUN had been removed, but the underlying aggregate was still intact. We rebuilt the layout of the aggregate and successfully recovered nearly all of the data associated with the deleted LUN.
After a power event there was datastore damage so we restored the pool using images of the disks, repaired the VMFS & recovered the virtual machines from this pool.
A rebuild process failed and caused the pool layout to be damaged. We made an image of all the disks & restored the RAID 10 pool, then recovered the volumes.
During a write operation one or more of the disks dropped and would not allow the pool to mount. We did a rebuild of the pool from read-only images & were able to recover both the LUN and its contents.
Send or drop off your item for a free diagnostic check with a description of the issue. One of our engineers will review the item, then provide you a written quote prior to doing any repairs.
Prior to posting the disks to us or dropping them off personally, please get a quote from us by phone or via the quotation request form. The first step in this process is to remove the disks from the array’s server or enclosure, even if you are dropping them off in person. If you are posting them, make sure they are packaged securely, include your contact information, and send them to us. Our diagnostic service is complimentary, and after receiving your disks and performing the diagnostic analysis, we will provide you with a firm written cost for our services prior to accessing a single sector.
Sending it? The use of a tracked, insured delivery method would be recommended. Alternatively, dropping it off can also be done Monday to Friday, 9am–5:30pm; no prior booking required but do ensure that all items are packed accordingly.
Still waiting to send something? Please complete the form describing the failure/fault in your own words and one of our engineers will get back to you with a quotation based upon your individual circumstances.
We will get back to you soon. If it is urgent, call 028 9002 0144.
Questions we get asked most about SAN recovery.
Yes, usually. When a RAID group in a pool loses enough disks for the pool to go offline, the data is still distributed across those disks. If we make an image of each disk, reconstruct your RAID groups & pool layout, and present your LUNs again, as long as no forced rebuild or reimport overwrote your metadata prior, the data will be available.
No. The controllers provide access to the pool that contains all of your data (which resides on the disks) but we are able to read from the disks and rebuild the pool completely through software as the failed controllers do not have an impact. Therefore, there is no requirement for matching hardware to be used for replacing failed controllers in order to restore your LUNs.
Frequently. Once a LUN has been deleted or un-mapped, it will normally remain intact on the pool until such point as the space allocated to this LUN has been over-written by new data. By rebuilding the pool and locating the LUN's original data, we are generally successful in recovering the deleted LUN. If the array remains untouched immediately after deletion, we typically recover more of the lost data.
A "Yes" would be correct. We rebuild the pools & LUNs, and recover the VMFS data stores and the virtual machines (VMs) that are running on those datastores. We also can repair a bad datastore, a bad VMDK for a VM, or the file system and database of a VM.
The cost for SAN recovery depends upon the size and configuration of your array. The minimum price for SAN recovery is £1,250 + VAT. This is based on our initial free assessment of your array. Upon completion of this assessment, we quote per array depending upon both the number of disks in your array along with the RAID/pool configuration of your array and the amount of effort required to complete the recovery. Typically, you only pay for our services if we successfully recover your data. If your disks require physical repair prior to us being able to use them, a deposit may be requested to cover the costs of parts and labour, with the balance only due on success.
In most cases, yes: you only pay for our services if we successfully recover your data. When individual disks require physical repair, a deposit may be applied against parts and laboratory time; however, you still only incur the balance when we successfully complete your recovery. We also document what charges will apply to your job prior to starting any work.
All of them: Dell EMC, HPE, NetApp, IBM, Hitachi, Pure Storage, Fujitsu, Lenovo, Infortrend and Promise, using both Fibre Channel and iSCSI technologies, with each type of RAID and pool configuration and on one shelf or multiple shelves.
Yes, we cannot receive complete SAN shelves. Please send the disks as they contain the data. Please send all the disks in the pool (and from any additional expansion shelves), numbered by shelf and slot. If at some point during the recovery process we need to know which controllers were involved, we will let you know what other items may be required.
Yes, it is a very standard enterprise procedure. Usually when a rebuild or expansion goes awry it does not destroy the entire pool. We create an exact copy of the disks at time of failure and then use those images to recreate the pool while bypassing the failed/incorrect parts of the original pool created during the malfunctioned rebuild/expansion.
Generally SAN recovery takes anywhere from 8-15 working days based upon the number of disks contained in the array and its associated complexities and configurations of the pool and LUNs. Most assessments can be completed within a couple of days; we can also prioritise critical business systems. Please simply advise how much impact this has.
You can deliver disks to our Belfast facility Monday to Friday 9am-5:30pm; alternatively you can ship them using insurance; for larger systems we can work with you to find the most efficient delivery method. Number the disks by shelf and slot; package securely; include your name/contact info; and we will book them in once received and provide a quote after completing the complimentary assessment.
First, there is no cost to assess your issue; we will then provide you a fixed written price (from £1,250 + VAT) per array for the work involved; and in most cases, you only pay if we recover your data. Every enterprise SAN platform that we recover is done so in-house here in Belfast. Do not rebuild or re-import your data. Simply remove the array from service, package the hard drives and ship them to us.