Data Recovery Case File · NAS & Network Storage · The Power Cycle Is the Event
The Planned Shutdown That Ended It
His enquiry described the most frustrating kind of failure — one that arrived through doing something correctly. An eight-bay network storage unit, bought second-hand and well out of warranty: "the other day I shut it down because my router was due to be changed." It did not come back. There is a natural assumption that a device switched off deliberately, in an orderly way, should switch back on — and for a machine that has been running continuously for years, that assumption is exactly backwards. The shutdown is not a coincidence in the story. It is the event.
| Media | Multi-bay network storage appliance, second-hand and out of warranty — shut down deliberately for network maintenance and failing to restart |
| Reported situation | Unit in continuous service for an extended period · deliberate orderly shutdown performed · unit failing to start subsequently · no fault reported before the shutdown · out of warranty |
| Fault class | Failure at power-on following prolonged continuous operation — appliance electronics or drive spin-up; member disks readable independently of the unit |
| Equipment used | Unit not repeatedly power-cycled · external supply and appliance electronics eliminated separately · members removed, labelled and imaged write-blocked (Atola TaskForce 2) · array assembled offline from images |
The decode: why switching off is the risky moment
Why continuously-running equipment fails at power-on: the counter-intuitive fact behind a great many of these. A device that has been running for years reaches a stable thermal and electrical state and simply stays there. The moments of greatest stress are starting: capacitors that have degraded quietly under years of heat are asked to handle inrush current, power supplies deliver their largest surge, and every drive in the unit spins up from stationary simultaneously — which is the most current-hungry thing any of them ever does. A unit that has been up for three years may have been failing gradually the whole time without anything showing, because nothing ever asked it to start again.
Why this catches people out: because the shutdown was correct and deliberate. He did the right thing before a network change, and the device punished him for it. That produces a reasonable but wrong conclusion — that the shutdown caused the damage — when in fact it revealed a condition that had been developing invisibly. The same failure was waiting for the next power cut.
What has most likely failed: in order of probability. The external power supply, which is the commonest fault on these units and costs nothing to establish. The appliance's own power circuitry, where degraded capacitors are a well-known age-related failure on units of this class and generation — worth having checked specifically rather than assumed. Or one or more drives failing to spin up after years of continuous rotation, which is its own version of the same phenomenon.
Why the disks are probably fine: in the first two cases entirely untouched. Even where a drive has failed to restart, a multi-bay array is usually built to survive one member — so the set may be complete without it. The contents are not lost because the box will not start.
What must not happen: repeated power-cycling in the hope it catches. If a drive is straining to spin, each attempt is that motor pulling at full torque against a mechanism that will not turn. And under no circumstances should the unit be allowed to rebuild if it does come up degraded — a rebuild reads every remaining member intensively and is precisely when a second marginal drive fails.
The lesson for anyone still running: a unit that has been up for years is not proven reliable; it is untested. Rebooting it deliberately, on a day when you have a current backup and time to deal with the consequences, is far better than discovering this during a power cut.
On the bench
The unit was not repeatedly power-cycled, since a drive straining to spin is damaged by exactly that. The external supply and the appliance's own power circuitry were eliminated separately before any drive was implicated, because a unit that fails for want of a supply should not be quoted as a recovery. The members were then removed, labelled by bay, and imaged individually write-blocked on the Atola TaskForce 2 — establishing which had failed to start and whether the surviving set was complete. The array was assembled offline from the images, where nothing could attempt a rebuild.
The outcome
The supply and appliance electronics eliminated first, the members imaged individually and the array assembled offline. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone whose always-on device died on a planned restart: the shutdown is the event rather than a coincidence, because equipment that runs continuously reaches a stable state and is stressed most at power-on — degraded capacitors meet inrush current and every drive spins up at once; so a unit up for years has been failing invisibly with nothing ever asking it to start. Check the power supply first, don't power-cycle repeatedly, and never let it rebuild.
Always-on device that won't restart
Stop power-cycling it, and don't blame yourself for shutting it down. Equipment that runs continuously reaches a stable state and stays there — the moment of greatest stress is starting, when degraded capacitors meet inrush current, the power supply delivers its biggest surge, and every drive spins up from stationary at once. A unit that's been up for years may have been failing quietly the whole time with nothing ever asking it to start again, so an orderly shutdown reveals the problem rather than causing it. Check the external power supply first, since that's the commonest fault and costs nothing. Don't let the unit rebuild if it does come up degraded. And if you have one still running, reboot it deliberately on a day when you have a current backup — far better than finding out during a power cut.
Don't keep power-cycling it — call Newcastle Data Recovery on 0191 406 1051; supply and appliance electronics eliminated first, members imaged individually, array assembled offline where nothing can rebuild.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.