The very first thing you need to do when a SAN drops a LUN is to stop everything. That includes failing over, resetting paths, updating firmware, running vendor scripts. This "motion" will harm your data. The first real technical step for recovering a SAN is a freeze: no writes; no repairs; no maintenance, good or bad. All the rest is simply remapping (and mapping), because a SAN is nothing but a series of agreements between different levels; therefore, recovering a SAN is all about rediscovering every level of agreement.
At the back-end of a SAN management console lies layered discipline: physical disks are formed into RAID groups, those groups are split up into logical units (LUNs), those LUNs are presented as Fibre Channel, or iSCSI, to hosts that format them using their preferred file system (VMFS, NTFS, etc.) or use them raw for a database. If there is a failure anywhere along this layering (e.g. at the RAID group layer), all of the upper layers will fail. And unfortunately, the management console, which reads the corrupted metadata generated by the failed lower-layered components, often reports on the failure incorrectly. Therefore, relying upon the opinions of the failed array regarding its own functionality is how good data gets rebuilt over.
Working from the bottom-up, we create images of each component member of the RAID group(s). We recreate each group. We re-carve each LUN. Then we give each LUN to its respective file-system rebuild process. The map is reconstructed based upon evidence, not based upon the configuration that has failed.
Enterprise jobs have an enterprise weight: compliance, insurers, sometimes boardrooms. The treatment fits. There is a formal change freeze from when we start working. Every disk & shelf is logged in & imaged read-only. The working set is stored on our storage, not on the patient. Also there is a formal scope document before any paid hour. Non-disclosure agreements are signed without much drama. And progress reporting in plain language (not ticket speak).
The promise that matters most at this scale: everything stays in-house and inside the UK. No partner labs, no offshore "escalations", no third party hands to touch the data.
A controller pair quarrelled & split-brained the metadata. A RAID group lost its second disk during a firmware update. A VMFS datastore showing empty after a resignature went wrong. An iSCSI target vanished & took the accounts system's LUN with it. Same bench method underneath regardless of logo: HP MSA/EVA, Dell EMC, IBM, NetApp, Fujitsu. Disks travel fine when they leave their shelves: take them out, number them by bay, then bring them to Charing Cross or send them by courier.
As all jobs have different shapes and sizes, the quote for each job is made separately and written down as soon as the free assessment has been completed prior to any paid work commencing. Each job starts with a free diagnostic, and finishes exactly how it was initially quoted. On most jobs, we do not charge for anything unless we recover your data. We never use an hourly rate, and we do not surprise you with an evaluation fee or take a percentage of the value of your recovered files.
Not while the data remains unprotected. Vendor repairs are written to restore service. They will happily rebuild structures over recoverable data to get a green light. Freeze changes, export nothing, image first. Once safe copies of every member exist, repairs can be attempted with nothing at stake.
Usually not: an empty-appearing datastore after a resignature or partition failure means the pointers were damaged (not the VMDKs). We can build back the VMFS structure from a copy of the LUN and recover all machines intact. The issue with these jobs is when you start provisioning new VMs on what appears as empty space. Quarantine the area until completion.
Yes, and most times this approach will be the best: your internal IT staff has knowledge of the current state of your environment, vendor support will know the specifics about the firmware, and we protect the data while both talk. We will identify clearly which suggested approaches are safe to perform once images have been created and which should wait. There is one hard and fast rule here: nothing writes to the original hardware devices.
Specifically quoted to the specific job request, in writing, after a free assessment of shelf count, disk count and failure shape, because a two-shelf MSA and a forty-drive fabric are not the same week. Enterprise storage and SAN start from £1,250 + VAT; the quote sets the cap on cost prior to any work being performed.
Get a quote before posting or dropping off your storage. We can give you a quote over the phone or with the form. First remove the disks from their enclosures or shelves. Remove them and put labels on each disk showing which bay they were in. Send the disks only. Now you have two safe options to send us your disks. No vans. Drop them off personally at Tay House, 300 Bath Street (right at Charing Cross, two minutes off the M8) Monday to Friday, 9am–5:30pm. Appointments are not required. Alternatively, pack up your disks securely and use a trackable, insured postal service from Paisley, Ayrshire, the Highlands or another part of the UK. In either case, we will complete all work using in-house engineers with a documented chain of custody and your data never leaves the UK.
Every hour that you "try stuff" is an hour of taking a risk on your data being recoverable. Call first (we will tell you which pieces to lock down), then get the members to us. SAN data recovery for Glasgow and across all of Scotland, handled with bench discipline and boardroom discretion.