Read this before you do anything else
Do not start a rebuild. That is the action that loses arrays.
A degraded array is still holding all of your data, and the rebuild intended to restore redundancy is the single commonest way that data is lost. Rebuilding reads every sector of every surviving member under sustained load, which is exactly the condition that pushes a second marginal disk over the edge, and disks bought together tend to fail together. Worse, a rebuild writes as it goes, so a rebuild started with the members in the wrong order, or onto a disk the controller mis-identified, overwrites the array's real contents with computed parity. Record the current disk order first and change nothing.
Emergency
CCTV recorder showing a degraded array
If your CCTV recorder is showing a degraded array, the most useful thing on this page is above this paragraph, and it is free. What follows is what the fault usually turns out to be, how recoverable this class of failure is, and what it would cost, including the cases we would tell you to take elsewhere.
What this usually turns out to be
A recorder running a degraded array carries the same risk as any other degraded array, plus one the NAS does not have: it is still recording. Every hour it stays up, it writes new footage over the oldest, using the surviving members to do it, so the retention window keeps advancing while the array is at its most fragile. The two clocks run together: the one on the hardware, where a rebuild could take a second member, and the one on the footage, where the incident you need is moving steadily towards being overwritten.
Recorders write in a manufacturer's own format and loop over the oldest footage continuously. The clock on a CCTV job is retention, not hardware.
Single member disk failurebetter news
The array is doing exactly what it was designed to do. Your data is intact and readable, and the danger is entirely in what happens next.
Second disk failing during a rebuild
The classic way arrays are lost. The rebuild's sustained full-surface read is what surfaces the weakness in the next disk.
Controller or NAS unit failurebetter news
The disks are all healthy but the thing that knew how they fit together has died. Recoverable by reassembling the array from the members directly.
Members reinserted in the wrong order
The array cannot be assembled without knowing the original order and parameters. Recoverable, but only if nothing has been written since.
Accidental reinitialise or new array created
The configuration has been overwritten. Much harder, and the outcome depends on how much was written afterwards.
What to check first on a CCTV recorder
- Stop recording through the unit's menu first, which halts the overwrite without touching the array.
- Work out the retention window and the date of the incident, since that decides whether the footage is still present at all.
- Export the footage you actually need before doing anything about the failed disk, while the array is still serving it.
What not to do
Do not rebuild the array before exporting what you need, and do not replace the failed disk first. Both put maximum load on aging members while the footage you came for is still only on those members. Do not let the unit format or reinitialise its storage to clear the fault.
Try this before you pay anyone
Nothing to install, but three things to record before anyone touches it
- Power the unit down rather than leaving a degraded array running under load.
- Label the disks with their current bay positions before removing any of them, and photograph the bays. Order is one of the parameters the array cannot be rebuilt without.
- Photograph or write down the array's reported configuration: RAID level, member count, which bay is flagged failed, and any stripe size shown in the interface.
Do not accept the unit's offer to rebuild, repair, or initialise, and do not replace the failed disk yet. Each of those writes to the array. The recording above costs ten minutes and is frequently the difference between a straightforward reassembly and an expensive one, and none of it involves running any software.
How recoverable this is
Often very good, because a degraded array is by definition one that still holds your data. Arrays are reassembled from the member disks directly rather than by trusting the controller that failed, so a dead NAS unit with healthy disks is routine work. What changes the picture is what happened after the degrade: a completed rebuild in the wrong order, or a reinitialise, writes over the real contents and that is a different job entirely.
What moves it, for your device specifically:
- Whether a rebuild or reinitialise has already been attempted
- How many members have actually failed, as opposed to being flagged
- Whether the original disk order and array parameters are known or recoverable
- Whether the filesystem on top of the array is intact
- Whether the recorder has kept overwriting the surviving members while degraded, which spends your retention window on top of the fault
We will not give you a percentage for your device, and you should be wary of anyone who does. Nobody can know what is recoverable until the media has been imaged and read. What is above describes this class of failure; your drive gets its own answer after an assessment, in writing, at no cost.
What it costs
Quoted after assessment
CCTV, NVR and DVR footage
Proprietary recorder formats parsed directly from the disk. Priced on recorder type, number of disks and the time window needed.
Nothing is charged until you have seen a fixed quote for your actual device and agreed to it, and the assessment that produces it is free.
Other faults on a CCTV recorder
RAID or NAS degraded on other devices
Next step
Tell us what failed.
Free assessment. Fixed quote before recovery starts. No recovery, no fee. Sydney drop-off or Australia-wide by tracked post.