A Buffalo is a small Linux computer running a software array behind a web page, so when one stops the trouble is nearly always in the firmware, the volume or a single drive rather than in your files. Here are the blinking codes and what they separate, the reinitialise offer that should always be refused, why the box travels complete, and what puts a network unit and a USB DriveStation in different bands.
A LinkStation or TeraStation is the array band: from £500 + VAT, climbing with the drive count. A single-disk USB DriveStation is £300 + VAT. The unit travels complete with its disks still in their bays, and the examination that produces your figure is free.
A Buffalo box is a small Linux computer wearing a web page, and it fails in Linux ways. The light on the front confirms that something is wrong and says almost nothing about what, so these are the states people describe when they ring.
Lights look ordinary, the network cable is fine, and the discovery utility comes back with nothing whatever. Generally the firmware has failed to bring the volume up and the unit has halted before starting any network service. Disks inside are usually in good order, and that is precisely why nothing needs taking apart at this stage.
Faults get signalled by blinking, and the pattern separates a failed drive from a volume that will not mount from firmware that will not load. Note the pattern down first. What counts far more is the next decision, and there are two right answers: do not reformat, and do not let the box rebuild onto a disk it has only just thrown out.
Buffalo tools raise that option readily whenever they cannot mount a volume, and somebody tired at the end of a long day will take it. Reinitialising writes a new array and a new file system across your data. When a unit starts asking that question and the contents matter, the correct response is the plug rather than the button.
A mirrored pair loses one and keeps working. A striped pair loses everything the instant either stops. Four bays with parity survive one absence and not two. Whichever it is, the rebuild is the risky part, and running one across drives bought as a matched set on a single afternoon is how one fault becomes two.
Nothing sits between a Windows machine and a network share unless a bin was configured on the box itself, so the delete happens immediately and completely. It is also very often recoverable, and what decides it is whether the unit was switched off promptly or left serving the rest of the office while freed blocks were handed out again.
The single-disk USB units are a different animal. Inside is a plain SATA drive behind a bridge board, and the fault is nearly always one or the other of those two. That puts it in the single-drive band at £300 + VAT rather than the array band, and a free examination establishes which of the pair gave up.
Behind the web interface, a LinkStation or TeraStation is a small Linux computer running a software array on a Linux file system — the older boxes on ext3, later ones on XFS, with the array handled by the same md layer that runs on Linux servers. That is good news twice over. The layout is described by superblocks written on the drives rather than by anything peculiar to Buffalo hardware, and the file system is one this bench opens every week. A unit whose electronics have died has not taken your files with them, and the drives can be read here without a working chassis existing anywhere on earth.
It follows that the failures are Linux failures in fancy dress: an XFS log left dirty by a power cut, an md set that will not assemble because two members disagree about which is current, a volume that mounts and shows a directory tree with pieces missing. Familiar, every one, and handled the same way — image each drive, rebuild the set on the copies, then repair the file system on a copy rather than on your disks. The Linux recovery page goes further into the file system half of it.
The first is saying yes to the reinitialise offer. It does exactly what the wording says and it is offered far more readily than it should be. If your box is asking, switch it off and ring somebody instead.
The second is cycling the power on a unit with a dying disk in it, again and again, in hope. Every restart makes the box try to assemble the volume, which spins the failing member up and reads it hard. On a drive shedding surface that is measurable damage each time, and it accumulates. A unit that has not returned after one restart will not return on the fifth.
There is a third and it is quieter. Removing every drive to read the labels, then putting them back in a different order, confuses the box and complicates the recovery that follows. If they are already out and the order has gone, say so honestly in the note — it can be worked out from the data, but knowing it saves time you would otherwise be paying for in days.
Nearly everything else stays at home while the drive alone makes the journey, but a network box goes into the parcel complete: chassis, disks still in the bays they were living in, and the power supply. An external drive sealed in its own case is the other exception. It saves you dismantling a unit whose bay order matters and it means nothing lands here as a bag of anonymous disks with a shrug attached.
Where the unit is a heavy rack-mounted TeraStation, or simply too awkward to post comfortably, ring the freephone first and the practical options can be talked through. What has never existed anywhere in this network is a collection round.
Anything with more than one drive in it, and anything that plugs into the network, is the array band: from £500 + VAT, climbing with the member count. LinkStation and TeraStation units of every size sit there, and they sit there because each drive is imaged separately and the volume has to be reassembled before a single filename becomes visible to anybody.
A single-disk USB DriveStation is a different proposition, being one hard drive behind a bridge board, so it belongs in the single-drive band at £300 + VAT. If you are unsure which of the two is in front of you, the test takes a second: network cable means the array band, USB cable almost certainly means the drive band. Either way the free examination settles it and nothing is charged for asking.
For completeness, the other rungs are £250 + VAT for cards and sticks, £400 + VAT for recorder disks and for encrypted volumes where the key can be produced, £800 + VAT for forensic casework written up, and £400 + VAT for the verified image with deletions extracted and no report. A Buffalo hit by ransomware is charged as ordinary array work from £500 + VAT.
Looking is free and closes on the second working day after the unit is logged in, producing one fixed written figure beside a realistic completion date. Nothing begins until you have accepted both of them. Single drives usually finish within two to four working days of authorisation; multi-disk boxes take longer, for the obvious arithmetic reason.
No fix, no fee applies to logical work — a volume that will not assemble although the drives are sound, a corrupt XFS file system, deleted shares, ransomware. Chip-level work, DVR jobs, forensic jobs and mechanical or electronic failure sit outside it, and a drive that has failed physically takes half the quoted figure upfront.
The Buffalo units that reach this bench from Coventry and Warwickshire are mostly the ones bought to solve a problem and then successfully forgotten. The shared drive in a two-room design practice. A camera archive behind the bar of a pub. Parish and club records living on one box in one cupboard in Kenilworth or Southam. Twenty years of survey photographs in a small building firm outside Atherstone. Four bays under a desk in a Bedworth workshop holding every job sheet since the business opened. They run for six or seven years without anybody laying a finger on them, which is a compliment to the hardware and precisely the reason no one has a current backup on the day one finally stops.
Post it covered, or drive it down — Oxford takes drop-offs on weekdays between 9:00am and 5:30pm, about fifty-five miles of M40, roughly an hour. Coventry has no counter and nobody collects. Synology, QNAP and the rest of the cupboard live on the general NAS recovery page.
Which is good news: the layout is recorded on the drives, so a Buffalo whose electronics have died has not taken anything with them.
Multi-disk jobs travel as a set of bare drives sharing one padded box, and that parcel wants tracking and covering for what is written on it rather than what it weighs. Booked over a Coventry post office counter in the afternoon, it usually reaches the Oxford bench the next working morning — faster than most firms manage to spare somebody for the run down the M40.
As a rule the storage comes out and the machine stays where it is. That applies to a laptop, a tower, an iMac and to the recorder sitting under a counter. Taking equipment apart is not something this bench does, and a repair shop will free a drive in a few minutes. Two things go the other way: an external drive stays sealed inside its own case, and a NAS travels as a complete unit with its disks still in their bays. A Fusion Mac is a third case — both of its drives come out and travel together, each one labelled. The single situation nobody can work around is memory soldered flat onto a mainboard, which is how Apple Silicon Macs and a good many slim laptops are built: if the storage will not unbolt, there is no parcel to send.
↓ Print the shipping & booking-in form (PDF)
Put Oxford Data Recovery on the label. From Coventry it is roughly fifty-five miles straight down the M40, about an hour if you would rather drive it in than post it. Either way you are told the moment it is logged, and the free diagnostic finishes two working days later.
Unsure what ought to go in the box? Ring 0800 689 0668 before you seal it, or work through the free online diagnostic and let it do the asking.