Seagate 14TB intermittently disconnects in TerraMaster D4-320 — fault follows drive between bays, SMART clean
Posted: 12 Oct 2026, 18:17
I’m trying to diagnose an intermittent disconnect/reconnect issue with a 14TB Seagate HDD in a TerraMaster D4-320 USB DAS.
The drive is a Seagate ST14000NM0121, originally shucked from a Seagate Expansion Desktop 14TB STKP14000400. It’s being used in a TerraMaster D4-320 connected directly to a Mac.
The problem is that the 14TB drive will occasionally disappear from macOS and then remount automatically a short time later.
SMART/DriveDx looks very clean:
- SMART health 100%
- 0 reallocated sectors
- 0 pending sectors
- 0 offline uncorrectable sectors
- 0 reported uncorrectable errors
- 0 UDMA CRC errors
- no SMART self-test failures
- no obvious media errors
At first I thought this might be a bad bay or bridge inside the D4-320, so I started logging macOS kernel and Disk Arbitration events.
On one failure, the 14TB was attached through a TerraMaster USB/UAS endpoint identified as:
TDAS@01220000
The logs showed a USB request timing out after 5 seconds with 0 bytes transferred, followed by the TDAS device re-enumerating and the disk appearing again. In other words, this did not look like macOS voluntarily unmounting the disk — the USB mass-storage endpoint itself stopped responding and disappeared temporarily.
To test whether the problem was the bay or the drive, I physically swapped the 14TB and 18TB drives between two bays.
I then copied roughly 307GB from an external SSD to both the 14TB and 18TB simultaneously. Both transfers completed successfully.
Shortly afterwards, the 14TB disconnected again.
This time the failure was on a different TerraMaster endpoint:
TDAS@01240000
The logs showed the TDAS device terminating, /dev/disk5 disappearing, and macOS explicitly identifying the lost volume as my 14TB volume. The drive later re-enumerated, macOS recreated the disk device, replayed the HFS journal, and remounted the volume.
So the important part is:
Before swap:
14TB → TDAS@01220000 → disconnect
After swap:
14TB → TDAS@01240000 → disconnect
The 18TB drive moved into the 14TB’s old bay and has not shown the same behaviour so far.
That makes a single bad D4-320 bay much less likely, because the failure appears to follow the physical 14TB drive.
What I’m unsure about is whether this points to:
- an intermittent fault in the 14TB drive’s SATA/interface electronics despite clean SMART,
- a compatibility issue between this specific Seagate model and the D4-320’s USB/UAS bridge,
- or some firmware/power-management quirk with this drive model.
The drive is a shucked ST14000NM0121, which I believe is one of the enterprise/Exos/Mach.2-style models people commonly find inside some 14TB Seagate Expansion externals.
Has anyone seen similar behaviour with this model in a USB DAS, NAS, Unraid, TrueNAS, or server setup?
In particular, has anyone seen a drive remain completely SMART-clean while periodically causing the USB/UAS endpoint to disappear, time out, and re-enumerate?
For now I’m backing everything off the 14TB onto another drive before doing more testing.
The drive is a Seagate ST14000NM0121, originally shucked from a Seagate Expansion Desktop 14TB STKP14000400. It’s being used in a TerraMaster D4-320 connected directly to a Mac.
The problem is that the 14TB drive will occasionally disappear from macOS and then remount automatically a short time later.
SMART/DriveDx looks very clean:
- SMART health 100%
- 0 reallocated sectors
- 0 pending sectors
- 0 offline uncorrectable sectors
- 0 reported uncorrectable errors
- 0 UDMA CRC errors
- no SMART self-test failures
- no obvious media errors
At first I thought this might be a bad bay or bridge inside the D4-320, so I started logging macOS kernel and Disk Arbitration events.
On one failure, the 14TB was attached through a TerraMaster USB/UAS endpoint identified as:
TDAS@01220000
The logs showed a USB request timing out after 5 seconds with 0 bytes transferred, followed by the TDAS device re-enumerating and the disk appearing again. In other words, this did not look like macOS voluntarily unmounting the disk — the USB mass-storage endpoint itself stopped responding and disappeared temporarily.
To test whether the problem was the bay or the drive, I physically swapped the 14TB and 18TB drives between two bays.
I then copied roughly 307GB from an external SSD to both the 14TB and 18TB simultaneously. Both transfers completed successfully.
Shortly afterwards, the 14TB disconnected again.
This time the failure was on a different TerraMaster endpoint:
TDAS@01240000
The logs showed the TDAS device terminating, /dev/disk5 disappearing, and macOS explicitly identifying the lost volume as my 14TB volume. The drive later re-enumerated, macOS recreated the disk device, replayed the HFS journal, and remounted the volume.
So the important part is:
Before swap:
14TB → TDAS@01220000 → disconnect
After swap:
14TB → TDAS@01240000 → disconnect
The 18TB drive moved into the 14TB’s old bay and has not shown the same behaviour so far.
That makes a single bad D4-320 bay much less likely, because the failure appears to follow the physical 14TB drive.
What I’m unsure about is whether this points to:
- an intermittent fault in the 14TB drive’s SATA/interface electronics despite clean SMART,
- a compatibility issue between this specific Seagate model and the D4-320’s USB/UAS bridge,
- or some firmware/power-management quirk with this drive model.
The drive is a shucked ST14000NM0121, which I believe is one of the enterprise/Exos/Mach.2-style models people commonly find inside some 14TB Seagate Expansion externals.
Has anyone seen similar behaviour with this model in a USB DAS, NAS, Unraid, TrueNAS, or server setup?
In particular, has anyone seen a drive remain completely SMART-clean while periodically causing the USB/UAS endpoint to disappear, time out, and re-enumerate?
For now I’m backing everything off the 14TB onto another drive before doing more testing.