How long does a RAID rebuild take?

It depends on the capacity of the members, the level, how busy the array is kept during the work and the controller’s own pace, so no fixed figure applies. What is predictable is that it grows with drive size and is the heaviest work the drives will do.

Nobody can give you a figure for this, and the reasons why are the useful part of the answer.

What the duration depends on

The capacity of the members. Every remaining drive is read from end to end and the replacement is written from end to end, so the work scales with size. Larger drives have grown far faster than the rates at which they can be read, which is why this window has been getting longer over the years rather than shorter.

The level. Reconstructing a mirror means copying from one surviving member. Reconstructing a parity member means reading every other member and computing. The second is more work.

How busy the array is kept. A machine still serving files is dividing its drives’ attention, and most enclosures expose a setting for how much priority the rebuild receives.

The controller. Enclosure makers do not all rebuild at the same pace, and some publish an estimate for their own models. That estimate, from the maker of the machine in front of you, is the only figure worth planning around.

Why the window matters more than its length

During a rebuild, a single-parity array has no redundancy left. A mirror has one member. Anything that goes wrong now goes wrong without a safety net.

And this is precisely when the remaining drives are being asked to do their heaviest reading, including of sectors that have not been touched since they were written. A member with a quiet, unnoticed problem is quite likely to reveal it here.

That is the whole argument for double parity on large arrays: not that failures are more likely, but that the recovery window is long and a second failure inside it would otherwise end the array.

What shortens it, and what does not

Replacing the failed member promptly shortens the exposure, because the window does not start until the replacement is in. Running degraded for a fortnight is a fortnight with no redundancy and no rebuild.

Leaving the array alone while it works shortens the rebuild itself.

A faster network link does nothing at all. The work is entirely between the drives and the controller; the network is not in the path.

What to do before it ever happens

Have the second copy already. A rebuild is the moment you find out whether the array was the only place something existed, and finding out then is the expensive way.

Keep an eye on the drives’ own reported health, and act on a warning rather than on a failure — replacing a member that is complaining is a rebuild started from a position of strength.

And know, in advance, how your enclosure signals a failure. A machine in a cupboard that has been running degraded for three months because nobody saw the light is a common and entirely avoidable story.

The manufacturer’s own documentation comes first, and this page is not a substitute for it. Every rate, capacity, class figure, endurance rating, port speed and power figure on this page is reported from the manufacturer's own documentation, with its unit written as published, and named as such — none of it is measured here. A stated capacity is reported as the maker counts it, and is never restated in another unit, because the two are not the same figure. And the second statement repeated rather than assumed: a RAID level is not a backup. It keeps an array readable through the failure of a drive, and it does nothing about a file deleted by mistake, an array encrypted by ransomware, a theft or a fire. A second copy somewhere else is a separate decision, and nothing on this site describes a machine as a place where files are safe.