Drive-health interpretation
S.M.A.R.T. is useful evidence, not a guarantee that a drive is safe. Synology’s current FAQ distinguishes test results and troubleshooting, and it also notes that NVMe SSDs use a different built-in monitoring mechanism rather than the same SMART test or bad-sector warning path. A good diagnostic workflow combines drive health, logs, I/O behavior, bad-sector trends and whether errors follow the drive.
Quick answer
Use SMART to support a replacement decision, not postpone backups
An Attention Required result deserves investigation and a current backup. A Normal result does not erase repeated I/O timeouts, reconnections or filesystem symptoms.
Current Amazon listings
Replacement and migration DiskStation options
Product cards are supporting purchase references only; they do not determine whether your failed drives are recoverable or whether a specific migration pairing is supported.
Technical decision
Trend and context matter more than one green indicator
A drive that repeatedly disconnects or accumulates media errors can be unsafe even if a quick test passes. Conversely, one historical attribute may not justify panic without understanding the value and trend.
Interactive recovery tool
Drive Health Risk Checker
This tool routes the symptoms you select to a safer next step. It does not issue destructive commands. Before initializing a disk, deleting a pool, reinstalling DSM or moving the only copy of important data, verify the exact live Synology procedure for your model and DSM build.
Recovery safety checklist
Four rules before you change a failing Synology system
Preserve readable data first
If unique files can still be opened, copy or back them up before changing storage metadata, reinstalling DSM or running stressful diagnostics.
Record drives and bays
Capture drive serial numbers, bay positions, health states and DSM messages before swapping disks or moving them to another chassis.
Match the action to the failure layer
A degraded pool, crashed volume, failed system partition, bad drive and failed bay have different supported recovery paths.
Keep a rollback path
During migration or recovery, do not erase the source system until the destination data, permissions, applications and backups have been verified.
SMART is a monitoring system, not a repair tool
The drive reports health attributes and self-test results; DSM presents those signals so you can judge risk. Running a test does not fix sectors, restore redundancy or repair filesystem metadata. Treat it as part of diagnosis.
Within a SMART-health review, smart is a monitoring system, not a repair tool matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Quick and extended tests answer different questions
Short tests are useful for routine checks and obvious failures, while deeper tests take longer and exercise more of the device. On a stressed array, test timing should consider backup status and the additional load placed on aging disks.
Within a SMART-health review, quick and extended tests answer different questions matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Attention Required deserves a backup-first response
When DSM reports a concerning health state, confirm that important data exists elsewhere before spending hours on further diagnostics. If the drive deteriorates during testing, the backup is more valuable than a perfectly explained failure mode.
Within a SMART-health review, attention required deserves a backup-first response matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Normal does not overrule repeated I/O errors
A disk can pass a self-test while the system logs command timeouts or reconnections. Those errors may come from the drive, slot or cable path. Combine SMART with Log Center and controlled physical diagnostics.
Within a SMART-health review, normal does not overrule repeated i/o errors matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Bad sectors should be interpreted as a trend
A stable historical count and a rapidly increasing count present different risk profiles. Any growth on a drive holding important data should increase urgency for backup and replacement planning rather than waiting for the next threshold.
Within a SMART-health review, bad sectors should be interpreted as a trend matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
NVMe SSD monitoring is different
Synology notes that NVMe SSDs use built-in health monitoring and do not follow the same SMART test or bad-sector-warning workflow. Do not conclude that an NVMe drive is unmonitored simply because the familiar HDD test option is absent.
Within a SMART-health review, nvme ssd monitoring is different matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Do not run every test during an active emergency
If the pool is degraded, the volume is read-only or drives are disappearing, the priority may be evacuation and redundancy restoration rather than a long health test. Tests create workload and time exposure.
Within a SMART-health review, do not run every test during an active emergency matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Record serial numbers and attribute history
A screenshot tied to a specific serial number is more useful than a generic note that Drive 2 was unhealthy. Bays can change during troubleshooting; the physical device identity should remain traceable.
Within a SMART-health review, record serial numbers and attribute history matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
SMART cannot identify a bad bay by itself
Controller and backplane issues can generate I/O symptoms even when the disk health data looks normal. If errors remain with one slot after a controlled swap, the NAS hardware path deserves attention.
Within a SMART-health review, smart cannot identify a bad bay by itself matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Replacement timing should reflect data importance
A lightly used media cache and the only copy of business records should not share the same risk tolerance. When data value is high, replace questionable drives earlier and maintain a verified restore path.
Within a SMART-health review, replacement timing should reflect data importance matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
New drives need a post-installation baseline
After replacement, record the new disk’s serial number, health state and initial test result. That baseline makes later changes easier to interpret and confirms that the new member is not immediately producing connection errors.
Within a SMART-health review, new drives need a post-installation baseline matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
SMART works best inside a broader maintenance routine
Combine periodic health review with backups, data scrubbing where appropriate, firmware awareness, temperature monitoring and spare-drive planning. No single dashboard metric can replace a resilient storage design.
Within a SMART-health review, smart works best inside a broader maintenance routine matters only when it is compared with the same drive's serial-number history, I/O logs and current pool condition. Record the result instead of treating the dashboard color as a permanent verdict. A long self-test is not automatically the next step when the array is already stressed, because diagnostic load can be less important than restoring redundancy or evacuating files. The purpose of this checkpoint is to improve a replacement decision without allowing monitoring data to become a substitute for backup.
Questions people ask
Synology SMART Test Guide questions
What does SMART mean on Synology?
It is drive self-monitoring information and self-test functionality presented through DSM. It helps assess risk but does not repair the drive.
SMART guidance should combine test results with error history and pool state. A green result never cancels repeated I/O faults or the need for backup.
Is a Normal SMART result proof the drive is healthy?
No. Repeated I/O errors, reconnections or other symptoms can still indicate a drive or slot problem even when a self-test passes.
What should I do with Attention Required?
Verify backup status, review the detailed drive health and logs, then plan replacement if the evidence indicates increasing risk.
Should I run an extended SMART test on a degraded pool?
Consider the array's risk and backup state first. A long test adds load and time exposure to already stressed disks.
Do Synology NVMe SSDs use the same SMART test?
Synology states that NVMe SSDs use built-in monitoring and do not support the same SMART test or bad-sector warning workflow.
Can SMART tell me whether the bay is bad?
Not reliably. A controlled drive-versus-slot test is more useful for separating disk faults from bay or backplane faults.
What do bad sectors mean?
They indicate media areas that cannot be used reliably. The count and trend matter, and growth should increase replacement urgency.
How often should I run tests?
Use a maintenance schedule appropriate to workload and drive age, but do not let testing substitute for backups or monitoring actual errors.
Can a new drive fail SMART immediately?
Yes. New hardware can be defective. Establish a baseline after installation and investigate any early abnormal result.
Should I replace a drive before SMART says Failed?
You may choose to replace earlier when other evidence shows risk, especially for important data. Thresholds are not the only decision input.
Official references and methodology
Verify the live DSM state before any destructive recovery step
This SMART guide focuses on evidence quality rather than a single pass-or-fail badge. It distinguishes HDD/SSD SMART testing from NVMe health monitoring, combines test history with bad-sector growth and I/O symptoms, and explains when a healthy-looking test should not override repeated disconnects. Thresholds and available tests differ by device type and DSM support, so current Synology drive-health documentation remains the final reference.
- Synology: S.M.A.R.T. test FAQ
- Synology: Determine whether errors come from the drive or slot
- Synology: Repair a storage pool
As an Amazon Associate, Cloudzat may earn from qualifying purchases. Failure states, DSM behavior, migration compatibility and recovery options can change by model and DSM build. Preserve readable data and verify Synology's current Knowledge Center instructions before initializing drives, deleting pools, reinstalling DSM or changing the only copy of important data.