How Wipora Erase Handles Drives Coming and Going

An erasure bench is not one drive at a time. It is a rack of docks with drives cycling through all day: one finishes, the operator pulls it out and puts the next one in, while the others carry on.

Older erasure software treats a run as one indivisible job. You load the tray, press Start, and the machine is tied up until the slowest drive in that tray finishes. A drive that takes eight hours holds up nineteen docks that finished in twenty minutes, and a drive plugged in halfway through cannot be started until the whole job ends.

Wipora Erase removes that limit. Drives can be added and removed at any time, including while erasures are running.

A batch is a tray of drives that went in together and will come out together. Each batch has its own erasure method, its own list of drives, its own completion and its own certificate. Several batches can run at the same time, independently of each other.

How a batch is created

Drives that are connected but not yet started sit in the Not started area at the top of the screen. Anything you plug in appears there.

When the operator presses Start, everything in that area is locked into a numbered batch. The button tells you what it is about to create, for example "Start Batch 3 (12)": batch number three, with twelve drives. The moment Start is pressed, a fresh, empty waiting area appears above it, ready for the next drives.

A drive belongs to one batch, for life

Once a drive is in a batch, nothing can take it out or start it again. Pressing Start a second time cannot pick it up, and the operator cannot untick it. A drive that is being written to must never be handed to a second erasure.

Different batches can use different methods

Batch 1 can be running a fast Clear on one customer's hard disks while Batch 2 runs a Purge on another customer's SSDs. Different jobs, different requirements, the same bench, at the same time, and each batch produces its own certificate for its own drives.

The screen is a stack of panels, one for each batch, with the Not started area at the top. Each panel shows its batch number, its method and a live count of how it is doing, for example "Batch 1, NIST Clear, 6 running, 4 done". Its Finish button sits in its own heading, so nothing is hidden behind a right-click.

With one batch open, the panel fills the screen. With two or more, each takes half, and the stack scrolls for the third and fourth. A batch with more drives than fit in its panel scrolls on its own.

Drive numbers belong to positions, not drives

Drives are labelled Drive 1, Drive 2, Drive 3, matching the order they sit on the bench. When a drive is removed, its number goes back into the pool, so if you pull out Drive 2 and plug in a replacement, the replacement becomes Drive 2, because that is where it physically sits.

This sounds small, but on a bench where forty drives pass through in a shift, numbers that only ever count upwards soon stop matching anything the operator can see. A number still in use is never given to another drive, including one that has been unplugged while its batch is still open.

Drives start as soon as you press Start, as many at once as the machine and its connections can usefully run. If a batch is larger than that, the rest wait in a queue and each waiting row says why. The instant one drive finishes, the next one takes its place, so drives come off the bench steadily instead of all at the end.

Why the connection matters

Every drive connects through a controller, and each controller can move only so much data per second. Drives that share a controller share that capacity.

• On separate ports of a proper disk controller, each drive gets its own full-speed connection, so eight drives really do run eight times faster than one.
• On a single USB hub, all the drives share one connection. Eight drives on one USB 3.0 hub each run at roughly an eighth of their normal speed.

That does not make the job longer. The total work is the same either way, because the connection was the limit and not the number of drives. It means the drives finish together instead of one after another. The software records how each drive is connected in its log, so a tray that took longer than expected has a documented reason.

Use this as guidance when planning a bench. The software adapts to whatever hardware it finds, and the same installation behaves correctly on a laptop with one USB hub and on a rack with a SAS backplane.

For hot-swapping to work at all, the software has to keep looking for new drives while erasures are in progress. Scanning a drive that is being written to could interfere with it, so the scan works around the busy drives:

• A drive that is being erased is left completely alone. The scan does not touch it.
• Its row is never removed, even if a scan briefly fails to see it because of a loose cable or a drive that has gone quiet mid-command. That row is where its progress and final verdict are shown.
• Rows are updated in place rather than rebuilt, so a running erasure never loses track of where it is reporting.

New drives show up within seconds of being plugged in.

Every drive must have an answer

The Finish this batch button only becomes available when every drive in the batch has reached a final state: either the software completed it, or the operator made a decision about it. There is deliberately no button that closes everything at once. A tray cannot be signed off with one click, and every drive that did not finish by itself needs its own deliberate action and its own warning.

When a drive will not finish

Sometimes a drive hangs, loses power or is pulled out mid-job. The operator marks that drive on its own, and a warning appears first, explaining three things:

1. The drive will be recorded as NOT COMPLETED.
2. Its certificate will say the erasure was not confirmed, and the drive must be treated as still holding data. Do not resell or release it on this certificate.
3. Marking it does not stop the erasure, because an instruction already handed to a drive cannot be called back. What changes is that the batch stops waiting for it.

The rest of the batch carries on untouched.

The certificate is issued when the operator says so

Pressing Finish is what produces the certificate, not the last drive quietly ending. The confirmation shows the batch summary, says plainly how many drives were not erased, and warns that a closed batch cannot be reopened. Any drive still connected afterwards goes into a new batch.

The order is deliberate. A certificate is a legal record, and it should be issued by a person who has looked at the result.

• You cannot walk away from a running job by accident. The Back button and the Reports tab stay locked while any batch is running or any certificate is being produced, and unlock together only when everything has genuinely finished.
• Every message names its batch. With two batches open, a plain "Upload complete" or "Erasure failed" could belong to either. Every popup says which batch, or which drive, it is about.
• Batches never affect each other. Each batch keeps its own results, progress, audit records and certificate. One batch finishing, failing or uploading cannot disturb another that is still running.

On a busy bench the machine stops being the bottleneck. The operator's hands are, which is the right place for it to be.