What is the difference between a degraded and a crashed volume?

These two states sound similar but mean very different things for your data. Knowing which you are in decides your next move.

Degraded

A degraded volume has lost its redundancy but not its data. In a RAID 5 / SHR array, one drive has failed; the array keeps running from parity and your files are still fully accessible. It is fixable: insert a replacement drive and run Repair, and the array rebuilds back to full protection. The catch is that while degraded, a single-parity array has no further fault tolerance: a second failure before you repair pushes it to crashed.

Crashed

A crashed volume has lost more drives than its redundancy could cover (e.g. two drives in a RAID 5). The array can no longer reconstruct the missing data, so it cannot be self-repaired. The data is not necessarily gone, but recovering it means restoring from backup or, for critical data with no backup, professional recovery.

What to do in each case

• Degraded: act promptly; replace the failed drive and repair before a second failure.
• Crashed: stop writing, do not reinitialize, and move to backup restore or recovery.

The practical lesson: repair a degraded array quickly, and use RAID 6 / SHR-2 on large arrays so a single extra failure does not mean a crash. Replace failed drives with CMR NAS drives; the table lists options.