How do I inspect a degraded Linux RAID array without changing it?
A degraded RAID array has lost part of its intended redundancy or membership. It may remain online while being more vulnerable to another failure. Inspect the array and member identities before planning replacement; an absent member can reflect cabling, controller, power or metadata problems rather than only a failed drive.
Before you start
- Authorised access to the dedicated server and the named inspection tools.
Instructions
1. Read the kernel’s array summary
Start with the current array state and any recovery progress. Identify the exact md device before using it in a detailed query.
cat /proc/mdstatRecord missing members, active recovery/resync and the named array. Do not interpret a progress bar as permission to reboot or remove another drive.
2. Inspect the selected array
Replace /dev/md0 with the verified Linux md array. This reports its metadata and members without assembling, recreating or modifying it.
sudo mdadm --detail /dev/md0Compare member device identifiers, state and redundancy with the expected layout. Hardware RAID needs its own supported read-only controller report instead.
Check the result
- The array, affected members and current redundancy are accurately identified, and the provider’s replacement or recovery plan preserves the remaining valid data.
Troubleshooting
- The array is inactive rather than merely degraded
- Use specialist recovery guidance for that exact layout. Force assembly or create can destroy useful metadata.
- Several drives disappeared together
- Investigate a shared controller, enclosure, cable or power path before treating them as independent drive failures.
Official references
Use the documentation for your installed version when a flag, service name or platform feature differs.