It survives a disk dying. It does not survive a deletion, a corruption, a ransomware run or a rebuild started on a member that was also failing — because whatever happens above the array happens identically to both copies.
Where one member cannot read an address, the other frequently can. The work is merging both images into one complete picture — not rebuilding onto a replacement while the survivor is under load.
RAID 1 mirrors. Every write goes to both members, so each disk holds a full, independent, readable copy of the entire volume. That makes it the most recoverable configuration in common use and the one most likely to resolve without any specialist work at all.
In most RAID 1 failures, one disk is fine. It can frequently be connected to an ordinary machine and read directly, because a mirror member is not written in some proprietary arrangement — it is the file system, complete, with a little controller metadata at one end.
That is worth trying before paying anybody, and we will tell you so on the phone.
Both disks failed, at different times. The usual pattern. One failed months ago unnoticed, the array ran unprotected, then the second went. Both are imaged and the better copy used, filling gaps from the other.
The mirror was silently broken. One disk dropped out of sync without alerting anyone, so it holds a copy from months ago. Recoverable, but you may get an old version rather than the current one.
Corruption was mirrored faithfully. RAID 1 copies whatever it is given. Ransomware, deletion and file system damage all replicate perfectly to both members. Redundancy is not backup and this is why.
A rebuild was started onto a good disk. Rebuilding in the wrong direction overwrites the surviving copy with the failed one. This turns a straightforward recovery into a difficult one and it happens more than it should.
Take one disk out of the array and connect it to a computer directly, using a SATA port or a USB adapter. On a hardware RAID 1 the member frequently mounts as an ordinary volume, because the mirror is just the file system with metadata appended.
If it mounts, copy your data off immediately — do not put it back in the array and do not attempt a rebuild until you have a copy somewhere else entirely.
If it does not mount, stop there. Do not initialise it, do not let Windows format it, and do not try the same thing on the second disk before the first has been imaged.
You have no protection until it is replaced, and rebuilding is the moment of highest risk. Copy the data off the surviving disk before rebuilding, not after.
Frequently yes, on hardware RAID 1. Each member holds a complete file system. If it mounts, copy everything off before doing anything else.
Usually recoverable. The two rarely failed at the same moment, so one is normally in considerably better condition. Both are imaged and the better used as the base.
More recoverable than it sounds, depending on how far it got. Stop the rebuild immediately and send both disks. Do not restart it.
No. It faithfully mirrors whatever happens, including damage. RAID protects against a disk failing and nothing else. It is not a backup.
From £500 +VAT for array work. Where one member reads and simply needs extracting, it may be quoted as single-drive work from £300 +VAT instead.