[Help] Repeated ATA bus errors
Posted: 16 Aug 2026, 06:12
Hi everyone,
I have an issue with my F4-424 Pro running TOS 7 and would like to know if anyone has experienced something similar.
TOS started reporting an I/O Access Status error for one of my HDDs, while the S.M.A.R.T. status itself remains normal.
All four drives are WD Red Pro WD40EFPX 4TB drives in RAID5.
I checked dmesg and found repeated errors on ata4:
Sometimes the errors also result in actual I/O errors, for example:
Initially, I suspected the HDD itself. However, its important S.M.A.R.T. attributes look fine:
I then performed several tests.
First, I completely powered down the NAS, removed all four drives, inspected the drive bays and backplane area, cleaned everything with compressed air, and carefully reseated all drives.
There was no visible dust or contamination, but I cleaned everything anyway.
The problem returned.
Next, while the NAS was completely powered off, I swapped the suspected HDD with another known-good WD40EFPX.
This was the important test: the problem did NOT follow the HDD.
After swapping the physical drive positions, a different HDD was connected through ata4, but exactly the same errors remained on ata4:
This made me suspect the SATA path rather than the HDD itself.
I then completely disassembled the NAS, disconnected and reseated the backplane connection to the motherboard, cleaned the connection and surrounding area again, reassembled the NAS, and performed another test.
After reassembly, ata4 initially negotiated normally at 6.0 Gbps:
However, after approximately 40 minutes under disk load, the same errors returned:
The errors also resulted in actual I/O failures:
During an earlier test, after several ATA bus errors and link resets, ata4 even renegotiated from 6.0 Gbps down to 3.0 Gbps:
This was immediately followed by an actual I/O error.
After rebooting the NAS, the SATA link negotiated at 6.0 Gbps again, but the ATA bus / Handshk errors eventually returned.
The RAID5 array itself is currently healthy and all four members are active:
At this point I have:
Based on these tests, it looks like the issue is somewhere in the SATA path rather than with a specific HDD.
Has anyone experienced something similar on an F4-424 Pro?
Does anyone know whether the SATA/backplane board in the F4-424 Pro can be purchased separately as a replacement part?
I have also opened a ticket with TerraMaster Support and generated a full TOS system report for them.
I will update this thread when I receive their response.
Thanks!
I have an issue with my F4-424 Pro running TOS 7 and would like to know if anyone has experienced something similar.
TOS started reporting an I/O Access Status error for one of my HDDs, while the S.M.A.R.T. status itself remains normal.
All four drives are WD Red Pro WD40EFPX 4TB drives in RAID5.
I checked dmesg and found repeated errors on ata4:
Code: Select all
ata4.00: irq_stat 0x08000000, interface fatal error
ata4: SError: { Handshk }
ata4.00: failed command: READ/WRITE FPDMA QUEUED
Emask 0x10 (ATA bus error)
ata4: hard resetting link
Code: Select all
I/O error, dev sdd, sector 1631577088
I/O error, dev sdd, sector 1631577816
Code: Select all
Reallocated_Sector_Ct 0
Reallocated_Event_Count 0
Current_Pending_Sector 0
Offline_Uncorrectable 0
First, I completely powered down the NAS, removed all four drives, inspected the drive bays and backplane area, cleaned everything with compressed air, and carefully reseated all drives.
There was no visible dust or contamination, but I cleaned everything anyway.
The problem returned.
Next, while the NAS was completely powered off, I swapped the suspected HDD with another known-good WD40EFPX.
This was the important test: the problem did NOT follow the HDD.
After swapping the physical drive positions, a different HDD was connected through ata4, but exactly the same errors remained on ata4:
Code: Select all
interface fatal error
SError: { Handshk }
ATA bus error
hard resetting link
I then completely disassembled the NAS, disconnected and reseated the backplane connection to the motherboard, cleaned the connection and surrounding area again, reassembled the NAS, and performed another test.
After reassembly, ata4 initially negotiated normally at 6.0 Gbps:
Code: Select all
ata4: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Code: Select all
ata4.00: exception Emask 0x10 SErr 0x400000 action 0x6 frozen
ata4.00: irq_stat 0x08000000, interface fatal error
ata4: SError: { Handshk }
ata4.00: failed command: READ FPDMA QUEUED
ata4.00: failed command: WRITE FPDMA QUEUED
Emask 0x10 (ATA bus error)
ata4: hard resetting link
Code: Select all
I/O error, dev sdd, sector 1631577088
I/O error, dev sdd, sector 1631577816
Code: Select all
ata4: hard resetting link
ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
After rebooting the NAS, the SATA link negotiated at 6.0 Gbps again, but the ATA bus / Handshk errors eventually returned.
The RAID5 array itself is currently healthy and all four members are active:
Code: Select all
[4/4] [UUUU]
- Checked the S.M.A.R.T. data of the suspected HDD
- Completely powered down the NAS
- Removed and reseated all four HDDs
- Cleaned the drive bays and backplane area with compressed air
- Swapped the suspected HDD with another known-good WD40EFPX
- Confirmed that the problem does not follow the HDD
- Confirmed that the same errors remain on ata4 with a different HDD
- Completely disassembled the NAS
- Disconnected, cleaned and reseated the backplane-to-motherboard connection
- Reassembled and tested the NAS again under disk load
Based on these tests, it looks like the issue is somewhere in the SATA path rather than with a specific HDD.
Has anyone experienced something similar on an F4-424 Pro?
Does anyone know whether the SATA/backplane board in the F4-424 Pro can be purchased separately as a replacement part?
I have also opened a ticket with TerraMaster Support and generated a full TOS system report for them.
I will update this thread when I receive their response.
Thanks!