A “degraded” NAS is still working on borrowed time. Your files may open normally, but the storage pool has usually lost some or all of its protection against another drive failure. The wrong next move—pulling the wrong disk, restarting repeatedly, or launching a heavy maintenance task before confirming a backup—can turn a repairable warning into an expensive recovery job.
The problem in one sentence
A degraded RAID or SHR pool may remain online after one drive problem, but redundancy is reduced; if another required disk fails before the pool is repaired and a separate backup is confirmed, the entire volume may become unavailable.
Who is affected—and who is not?
Likely affected
- Synology, QNAP, TrueNAS, Windows Storage Spaces, server RAID, or another multi-drive storage system reporting Degraded, Critical, Failed, or a missing disk.
- A small business using a NAS for shared documents, scans, photos, line-of-business exports, or workstation backups.
- A home NAS storing the only copy of family photos or video.
- Any array with warning lights, repeated drive errors, unexpected beeping, or failed S.M.A.R.T. tests.
Not necessarily affected
- A single external USB drive with no RAID; it can fail, but “degraded array” instructions do not apply.
- A healthy NAS showing only a scheduled test or normal drive activity.
- A cloud sync folder with no local NAS. Cloud sync still needs its own recovery plan, but there is no local array to rebuild.
- A warning about storage capacity rather than drive health. Low space needs cleanup or expansion, not an automatic drive swap.
Why “RAID is not a backup” matters right now
RAID is primarily an availability tool: depending on the RAID level, it can keep a volume online after a limited number of drive failures. It does not create an independent copy. Accidental deletion, ransomware, theft, fire, controller failure, file corruption, or a bad administrator action can affect every disk in the array at once.
The U.S. Cybersecurity and Infrastructure Security Agency recommends backing up data and keeping offline or cloud-to-cloud copies as part of ransomware resilience. That separate copy is what gives you a safe exit if the rebuild fails or a second disk has unreadable sectors.
First 30 minutes: the safest order of operations
1. Stop nonessential activity—but do not yank a drive
Pause large file copies, media indexing, virtual machines, backup jobs writing to the NAS, and other heavy workloads if the business can do so safely. Do not power-cycle repeatedly. Do not remove a drive just because its front light looks different. The management console, slot number, model, and serial number should all agree before anyone touches hardware.
2. Capture the evidence
- Take a screenshot of the storage-pool status and event log.
- Record the NAS model, RAID type, pool/volume name, total drive count, failed or missing slot, drive model, capacity, and serial number.
- Write down when the warning began and whether there was a power outage, move, update, unusual noise, or performance slowdown.
- If email or phone alerts are configured, preserve the original alert.
This record prevents the most dangerous mistake: replacing the wrong disk. It also gives vendor support or an IT technician enough detail to choose the correct recovery path.
3. Confirm a separate backup before rebuilding
Check the most recent backup job, then restore a small but meaningful file to a different location and open it. Prioritize irreplaceable records: accounting exports, current customer documents, password-vault emergency data, photos, legal records, and active project files. A green “backup completed” badge is helpful; a successful restore is proof.
If the NAS contains the only copy and the data is critical, stop experimenting. Every additional read, reboot, scrub, or rebuild attempt can reduce recovery options when another disk is weak. Professional recovery may be cheaper than recreating the data—or discovering it cannot be recreated.
4. Check every remaining disk’s health
Review S.M.A.R.T. status, recent test results, bad-sector or I/O errors, temperature, and the event history for all drives—not only the one marked failed. QNAP’s documentation explains that its disk-health screen reports S.M.A.R.T. information, scan results, and self-test results; a S.M.A.R.T. attribute changes to a warning when it reaches the manufacturer or QTS threshold.
A “healthy” label is useful but not a guarantee. Watch for a second disk with rising errors, intermittent disconnects, or very long response times. Do not start several extended tests at once on a stressed array. If multiple drives look questionable, get model-specific guidance before beginning a rebuild.
5. Verify the replacement is compatible
- Capacity: the replacement generally must be at least as large as the disk it replaces; usable capacities can differ even when retail labels look the same.
- Drive type: use a NAS- or enterprise-appropriate CMR drive when required by the vendor and workload. Avoid assuming any desktop disk is a safe substitute.
- Interface and form factor: confirm SATA/SAS/NVMe, 2.5-inch/3.5-inch, tray/caddy, and supported sector format.
- Compatibility: check the NAS or controller vendor’s compatibility list and firmware notes.
- Warranty: verify by serial number, but do not delay a critical repair solely to wait for a warranty shipment if business continuity is at risk.
6. Follow the vendor’s exact repair procedure
Synology’s DSM documentation instructs administrators to replace defective drives before using the storage-pool Repair function. QNAP documents its own recovery workflow for RAID groups in a degraded state. The correct hot-swap, shutdown, replacement, and repair sequence depends on the model, RAID level, enclosure, and whether the drive is truly failed or merely disconnected.
Do not initialize, format, or create a new pool on the replacement disk unless the documented repair wizard explicitly calls for the action. “Initialize disk” and “repair array” are not interchangeable.
7. Monitor the rebuild from start to finish
A rebuild can take hours or longer and may make the NAS slower. Keep the system on stable power, preferably through a tested UPS. Do not assume “rebuilding” means “fixed.” Watch for new read errors, a second failed drive, overheating, or a stalled percentage. Avoid firmware updates and unnecessary restarts during the rebuild unless the vendor specifically requires them.
8. Verify the repaired system—and the backup
- Confirm the pool returns to Healthy or the vendor’s equivalent.
- Confirm every expected shared folder is present.
- Open representative documents, photos, databases, and archives.
- Run or verify a fresh backup to a separate destination.
- Restore another test file from that backup.
- After the pool is stable, schedule supported S.M.A.R.T. tests and data scrubbing. Synology documents data scrubbing as a maintenance function for supported pools; it is preventive maintenance, not a substitute for an independent backup.
- Replace the temporary incident notes with a permanent record of the failed drive, new serial number, rebuild time, and test results.
Good moves versus expensive moves
Phone readers: swipe the table left to see both columns.
| Good move | Bad move |
|---|---|
| Match the failed slot, model, capacity, and serial number in the management console. | Pull the drive with the amber light without confirming its identity. |
| Restore-test critical data before starting the repair. | Assume RAID itself is the backup. |
| Use the vendor’s model-specific repair workflow. | Click Initialize, Create Pool, or Format because the new disk is blank. |
| Reduce heavy workloads and keep stable power during rebuild. | Restart repeatedly, run a scrub immediately, or update firmware mid-rebuild without a documented reason. |
| Escalate when multiple disks show errors or the data has no separate copy. | Keep trying random fixes on irreplaceable data. |
Real-world examples
Example: four-bay office NAS, one failed disk
A four-bay RAID 5 or SHR-1 system reports one failed drive but shared folders remain accessible. The business pauses nonessential writes, confirms last night’s separate backup by restoring a current spreadsheet, checks the other three disks for warnings, records the failed slot and serial number, installs a compatible replacement using the vendor procedure, and monitors the full rebuild. This is the normal repairable case—but it still deserves urgency.
Example: two-bay mirror, one disk missing and the other has errors
A two-bay RAID 1 system reports one missing disk while the surviving disk logs I/O errors. The only backup is six months old. This is not the time for a blind rebuild: rebuilding reads the surviving disk heavily. The safest choice is to stop nonessential use, copy or image critical data if appropriate, and get model-specific professional guidance before stressing the remaining disk.
Frequently asked questions
Can I keep working while the NAS is degraded?
Often yes, but that does not mean you should continue normally. Minimize heavy writes, confirm a separate backup, and begin the repair plan promptly. If more than one disk is unhealthy or the NAS contains the only copy, stop and escalate.
Should I reboot first?
Not automatically. A loose connection may clear after a controlled shutdown and reseat on some systems, but an unnecessary reboot can also make an intermittently failing disk disappear permanently. Capture logs and follow the vendor’s model-specific guidance first.
Should I run data scrubbing before replacing the failed drive?
Usually not as a first reaction. Scrubbing is a valuable preventive-maintenance process on supported healthy or appropriately redundant pools, but it creates substantial disk activity. When a pool is degraded, first confirm backups, assess all disks, and follow the vendor’s repair order.
Can I replace the failed disk with a larger one?
Often yes if the vendor supports it and the replacement meets the minimum usable capacity, interface, sector-format, and compatibility requirements. The extra capacity may not become usable until additional drives are replaced or the pool is expanded.
Does a successful rebuild prove my files are good?
No. Verify shared folders, open representative files, run a fresh backup, and perform a restore test. A healthy array can faithfully store corrupted, encrypted, or accidentally deleted data.
Related The IT Guys guides
- Test one file restore before you trust your backup
- Stop opening a QuickBooks Desktop company file from OneDrive or a NAS
- Hurricane preparations for home and business computers
Your next move
If your NAS says Degraded, take screenshots, record the exact failed slot and serial number, verify a separate backup with one real restore, and stop before guessing. The IT Guys can help home and small-business customers in Port Saint Lucie, Jensen Beach, Fort Pierce, and Vero Beach assess the array, protect critical files, and plan the repair.