Your shares have disappeared. Your “volume” has crashed, according to the volume screen. Or your “NAS” is responding as if it is alive on the network but is not providing you with any data. While your NAS is usually still holding onto every single file you are concerned about, the menu for your NAS includes at least one button that can change that forever.
Before anything else, separate two issues from each other that appear to be similar from a laptop perspective. When the problem is a network one, the data is good and the networking/communication is bad. The issue could include a Windows update disabling an older SMB version, an expired password, a changed IP address, or a router that has renumbered the network. The unit’s own admin page should still load, the volume will say everything is okay, and there is no risk involved. Just reconnect and continue.
When the problem is a storage one, the NAS is online but the data is not. Synology says Volume Crashed or Degraded; QNAP will say "pool unmounted" or "volume inactive"; a WD My Cloud will be flashing lights and won't show any available shares. Usually the admin pages still work, which is why this type of issue is so hazardous: it allows you to press a nice clean line of buttons right before losing your entire array.
By far the most frequent reason is the failure of a single disk causing a redundant volume to fail as well. Synology Hybrid RAID, RAID 1 and RAID 5 can each tolerate up to one "dead" member; however, these will be used in a degraded state. A NAS which has run for eight months in a degraded mode will have lived off the last good thing it had. Once the second disk fails, or if a rebuild attempt hits a bad sector which cannot be read (the rebuild issue with RAID 5), the entire volume goes down. This is the RAID 5 rebuild problem showing itself in a home office cabinet.
Secondly, the most frequently occurring case after this would be file system corruption (as opposed to disk failure): damaged metadata from btrfs or ext4 caused by either loss of power during writing to disk, or a firmware update rebooting the unit while still trying to write to disk. All disks remain healthy, the array remains intact and the file system above it no longer knows what to do. Lastly, you have the delete case (a deleted shared directory, a snapshot policy that never ran, a sync that copied the mistake), none of which involved hardware failure. All appear in the same manner.
All NAS interfaces offer to assist; their assistance will be based on a fully operational system. Repair Volume, Rebuild and Resync all start writing data across the remainder of the disks (those holding the sole copy of your data), and a rebuild that fails halfway creates an even greater mess to untangle than the original failure. Create Volume and Initialise are yet worse. They create a new empty structure on top of what exists now; there is no forum support magic that can bring that back.
Two additional traps, both with good intentions. Removing each disk and testing it individually in a USB docking station while attempting to recover the problem destroys the one item recovery relies upon (the order which each member was originally in) if you did not label them by bay number beforehand. Factory resetting the device to "start clean" removes the configuration information describing how the array was built. If the volume is already down, do not press any buttons. Press the power switch. A Hamilton Synology arrived here just as described, and it left completely whole because repair was never pressed.
We recover data from a NAS without "touching" the original disks. Each drive is "imaged" by itself with hardware that will only read (no writes). The array is then created virtually from these individual images. Everything including stripe order, parity rotations, SHR's multiple layer structure as well as btrfs metadata etc... is recreated in the new virtual array. Your shares are also retrieved from this newly built volume. Thus nothing is written back to the drives you sent, as they have not been modified for the recovery. Therefore if an upstream rebuild fails you would experience inconvenience but not failure, and therefore a My Cloud whose own rebuild had previously failed could still be recovered in full.
Once you have a quote to send your NAS, it's easy. Take the drives out of the NAS if you can and label them by bay, or, if you can't, simply post the NAS as is. The one version of the NAS that takes time will be an unlabelled heap. All NAS recovery is carried out in-house and nothing leaves the UK. NAS recovery starts from £500 + VAT, with a free diagnostic and a fixed price quotation in writing before any work, and no fee on most jobs unless the data comes back.
Data is generally intact, until a rebuild writes over it. Shut the device down. Label your drives (if they are removed) and allow the complimentary assessment to inform you of what can be recovered. NAS & RAID cost from £500 + VAT. All quotes will be fixed in writing prior to commencement.