Testing drive health in a Synology NAS using S.M.A.R.T. data
Hard drives and SSDs rarely fail without warning. Long before a disk becomes unreadable, internal sensors log subtle changes in temperature, reallocated sectors, spin-up time, and vibration. S.M.A.R.T. — the Self-Monitoring, Analysis and Reporting Technology built into every modern ATA, SATA, and NVMe drive — collects those readings and turns them into numerical attributes a host system can read on demand. On a Synology NAS, this telemetry is exposed through DiskStation Manager, which makes drive health testing far more accessible than running smartctl from a telnet session.
For Australian homes and small businesses, the value of monitoring this data is amplified by local realities. Many regional properties still see brownouts during summer heatwaves, NBN connections occasionally drop mid-backup, and bushfire-prone parts of Victoria and New South Wales may leave storage gear running on UPS power for hours. Each situation stresses drives differently, so a quiet, scheduled check of S.M.A.R.T. data can catch a weakening disk before it takes a whole volume down.
What S.M.A.R.T. actually measures on a NAS drive
S.M.A.R.T. is not a single measurement but a bundle of attributes defined in the drive's firmware. Manufacturers choose which ones to expose, and for NAS-class HDDs such as Seagate IronWolf and WD Red Plus, the most informative values are usually consistent. Each attribute carries a current value, worst recorded value, threshold, and raw data count. When the current value falls below threshold, the drive is signalling imminent failure and DSM surfaces it as a critical alert.
Key HDD attributes are Reallocated Sectors Count, Current Pending Sector Count, Uncorrectable Error Count, Spin Retry Count, and Temperature Celsius. On SSDs, focus shifts to Media Wear Indicator and Available Spare. The comparison covers the most useful attributes for spinning disks and SSDs in a NAS context.
| Attribute | Tracks | HDD range | SSD range | Warning sign |
|---|---|---|---|---|
| Reallocated Sectors Count | Bad sectors swapped for spares | 0 | N/A | Any growth |
| Current Pending Sector Count | Sectors waiting to be remapped | 0 | N/A | Steady increase |
| Uncorrectable Error Count | Sectors the drive could not fix | 0 | 0 | Any non-zero value |
| Spin Retry Count | Motor start attempts | 0 | N/A | Any increase |
| Temperature Celsius | Operating heat | Under 45°C | Under 55°C | Sustained over 50°C |
| Power-On Hours | Cumulative runtime | Context only | Context only | Compare with warranty |
| Media Wear Indicator | NAND wear | N/A | 100 → 0 over life | Below 10% remaining |
| Available Spare | Spare block reserve | N/A | 100% | Below 10% |
Setting up S.M.A.R.T. monitoring in Synology DSM
DSM exposes S.M.A.R.T. inside Storage Manager, the same panel used for RAID, storage pools, and SSD caches. Open the main menu, launch Storage Manager, then click the HDD/SSD tab. Each disk listed has a small health indicator on the left; clicking it expands a window showing the S.M.A.R.T. table for that drive. The values mirror what smartctl would return, in a friendlier format.
Schedule a recurring test under Storage Manager → HDD/SSD → S.M.A.R.T. Test, choosing between a Short Self-Test and an Extended Self-Test with frequency from daily to monthly. Most Australian home users pick a weekly short test and a monthly extended test, scheduled for early morning when NBN upload contention is lower and nobody is streaming to Plex. Email notifications under Control Panel → Notification remain the most reliable channel in Australia, because SMS gateways can be throttled by Telstra and Optus during major outages.
Running short and extended self-tests
The short test typically runs for two minutes on a modern NAS HDD and sweeps the most critical mechanical and electronic subsystems. It is cheap, safe to run while the volume is in use, and works well as a weekly smoke detector. The extended test reads every sector on the disk and can take eight to twenty-four hours depending on capacity. Plan it for a window when the NAS is idle, such as overnight on a Sunday after backups to an offsite cloud such as Backblaze B2 have finished.
To run an ad hoc test, highlight a disk in Storage Manager, choose Action → S.M.A.R.T. Test → Short or Extended, then confirm. DSM logs the outcome under Log Center. If a test reports an error, back up the volume immediately and consider replacing the drive. RAID 5 and RAID 6 tolerate a single disk loss, but a second failure during rebuild can wipe the array, and rebuilds are the most stressful event a drive ever faces. SSD caches behave differently: a read-write cache must be disabled before running an extended test on the underlying SSDs, otherwise DSM will refuse to start.
Reading the S.M.A.R.T. values that matter most
After a test finishes, DSM shows the post-test attribute snapshot. Focus first on Raw Read Error Rate and Reallocated Sector Count: even a single non-zero raw value on a desktop-class WD Blue is a sign to retire the disk from a NAS role. WD Red, Seagate IronWolf, and Toshiba N300 are rated for 24/7 vibration environments typical of multi-bay enclosures and tolerate more variation, but the principle holds — any upward trend is meaningful.
Key attributes worth highlighting in DSM:
- Reallocated Sector Count — anything above zero means sectors have already been replaced
- Current Pending Sector Count — sustained growth indicates read instability
- Uncorrectable Error Count — any non-zero value is grounds for immediate replacement
- Temperature Celsius — sustained readings over 50°C accelerate mechanical and NAND wear
- Power-On Hours — context value used to spot unusually stressed drives
Temperature deserves close attention in Australia. Garage installations in Brisbane or Perth can see ambient temperatures above 38°C in summer, and a poorly ventilated Melbourne closet can climb past 45°C. DSM's dashboard shows live temperature; correlate it with the S.M.A.R.T. record after a hot day. The Health Status field at the top summarises everything DSM sees: Normal means all attributes are above threshold; Warning means at least one is approaching failure; Critical means a value has crossed threshold or a self-test failed. Treat Warning as the cue to order a replacement drive, Critical as the cue to migrate data tonight.
Interpreting results and warning thresholds
S.M.A.R.T. thresholds are conservative by design, and a single reading does not give the full picture. Firmware smooths noise, and identical models can report slightly different raw values without being defective. Look at the trajectory: re-ran the same extended test in a week and saw the same raw number? The drive is stable. Watched the number climb? Replace it before the volume is stressed. Australian technicians commonly report slow growth in Current Pending Sector Count on drives older than four years, sudden Reallocated Sector jumps after a long power outage in regional South Australia, and inflated Power-On Hours on used ex-lease disks imported from overseas.
When an alert appears, the following sequence keeps recovery predictable and data safe.
Sequence to follow when a SMART alert appears:
- Log in to DSM and confirm the offending drive in Storage Manager
- Trigger an immediate extended self-test rather than relying on passive counters
- Verify the latest offsite or external backup is intact and recent
- Order a compatible replacement drive through an Australian reseller such as Scorptec, Mwave, or Centre Com
- Schedule the swap during a quiet window and hot-spare if the chassis supports it
Automating tests and remote notifications for Australian setups
Synology supports scheduled S.M.A.R.T. tests natively, but owners who want richer reporting often add Active Insight, a cloud service hosted in Singapore and Frankfurt that aggregates telemetry across multiple units. For households preferring local-only tools, the open-source package "S.M.A.R.T. Monitoring" in SynoCommunity pushes results to Grafana on a Raspberry Pi, a popular setup in tech-savvy suburbs around Adelaide and inner Brisbane.
Power management matters as much as test scheduling. A CyberPower or Eaton UPS connected via USB lets DSM shut the NAS down gracefully during a blackout, preventing the abrupt power loss that causes the worst S.M.A.R.T. incidents. Configure the UPS under Control Panel → Hardware & Power, set the early shutdown timer to 5 minutes, and test the routine every few months by pulling the plug at the wall. A scheduled self-test should pause automatically when the UPS signals low battery, otherwise DSM may run a long test while running on reserve power and brown out mid-scan.
Build a habit of glancing at the Storage Manager dashboard once a week. S.M.A.R.T. is most powerful when treated as a trend, not a single snapshot, and a thirty-second review of the attribute table is often enough to spot trouble early. Pair it with a quarterly verification that backups are restorable, and the NAS will remain a quiet, dependable corner of the household or office for many years.
Browse compatible replacement drives and UPS units suited to Australian conditions through WhichNAS, and lock in a routine that keeps your Synology healthy through every heatwave, brownout, and summer storm.