Page 1 of 1

[Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 10 Aug 2026, 04:29
by jraghon
Howdy TerraMasters.

I have a new D4-320 and six HC520 HDDs:
  • 2x HPE GWDFE (512e, HP firmware)
  • 2x HGST ALE600 (512e, WD/HGST firmware)
  • 2x HGST ALN600 (4Kn, WD/HGST firmware)
I've tested each drive in each of the four bays. [smartctl -ax] works fine on the 512e drives (GWDFE and ALE600). On the 4Kn drives (ALN600), [smartctl -x] times out, and I see this in dmesg:

Code: Select all

[Sat Aug  8 22:37:58 2026] xhci_hcd 0000:00:14.0: ERROR Unknown event condition 199 for slot 7 ep 3 , HC probably busted
[Sat Aug  8 22:37:58 2026] usb 2-2.3: reset SuperSpeed Plus Gen 2x1 USB device number 5 using xhci_hcd
[Sat Aug  8 22:39:00 2026] usb 2-2.3: reset SuperSpeed Plus Gen 2x1 USB device number 5 using xhci_hcd
The host controller isn't actually faulted -- it recovers, and the drive keeps working for normal I/O. So far, this only affects extended SMART queries.

My guess (after research) is that the ASM235CM bridge firmware doesn't correctly translate the 4Kn drive's SMART log structure, and the malformed/oversized response causes the USB transfer to stall until the host resets the port.

I couldn't find a published firmware version or update path for the D4-320 itself -- is there one?
And has anyone else hit this specific ASM235CM + 4Kn combination?

Thank you in advance!

===

System data:

Code: Select all

System: Debian, kernel 6.12.88+deb13-amd64
smartctl 7.4 2023-08-01 (r5530)

Enclosure: TerraMaster D4-320
Bridge chip: ASMedia ASM235CM (USB idVendor 0x174c, idProduct 0x235c)
Connection: USB 3.2 Gen 2 (10Gbps), SuperSpeed+

Re: [Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 10 Aug 2026, 16:36
by EriChan
Thank you for providing such detailed technical logs.

1. Lab Verification: Our engineering team is currently investigating and reproducing this issue locally (ASMedia ASM235CM + 4Kn drives + extended SMART queries via smartctl -x).

2. Workaround (Recommended): In the meantime, please avoid -x (which requests GP Logs that stall the USB endpoint) and use standard SMART commands instead:
  • Standard SMART query:

Code: Select all

smartctl -d sat -a /dev/sdX
  • Limit CDB length:

Code: Select all

smartctl -d sat,12 -a /dev/sdX
  • Read standard error & self-test logs:

Code: Select all

smartctl -d sat -l error -l selftest /dev/sdX

Re: [Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 10 Aug 2026, 21:20
by jraghon
Thank you for your reply, EriChan.

I should have provided these additional details:
  • [smartctl -a /dev/sdX] works fine on all six drives.
  • [smartctl -l <log that is part of -a> /dev/sdX] works fine on all six drives.
  • [smartctl -l <log that is part of -x> /dev/sdX] fails only on the 4Kn drives.
For example:

Code: Select all

smartctl -d sat,16 -l devstat /dev/sdX
works on all of the 512es, but triggers a single SuperSpeed reset on either of the HGST ALN600s.

Re: [Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 11 Aug 2026, 18:48
by CursaYang
Thank you for your further feedback. Additionally, please note the following:
The current version of smartctl installed on your system is 7.4 r5530. While version 7.4 natively supports extended features such as Device Statistics, subsequent upstream releases continue to roll out fixes for ATA protocol issues, USB bridge compatibility, and device recognition.
If it is convenient for you, you can also run an A/B test using a newer version of smartmontools. This test can rule out the possibility that smartctl itself incorrectly handles the ALN600 4Kn Log Directory or Log Page length.

Re: [Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 18 Aug 2026, 05:26
by jraghon
Howdy. I have re-run the same tests using

Code: Select all

smartctl 7.5 2025-04-30 r5714 [x86_64-linux-6.12.88+deb13-amd64]
with nearly identical results (the output format has changed, but the dmesg logs are identical).

Re: [Help] D4-320 -- crash on S.M.A.R.T. extended tables for 4Kn drives

Posted: 18 Aug 2026, 19:39
by CursaYang
Thank you for your cooperation and testing; we will attempt to reproduce the issue in the laboratory.