EXOS 18TB drives not recognized as Seagate

[Lennert-K]

Running the following on my QNAP NAS:

docker run --rm --privileged -v /dev:/dev ghcr.io/gamestailer94/farm-check:latest ALL
=== Checking device: /dev/sda ===
FARM data not available - likely not a Seagate drive
SMART: 2394
FARM: N/A
RESULT: SKIP

=== Checking device: /dev/sdb ===
FARM data not available - likely not a Seagate drive
SMART: 2398
FARM: N/A
RESULT: SKIP

=== Checking device: /dev/sdc ===
FARM data not available - likely not a Seagate drive
SMART:
FARM: N/A
RESULT: SKIP

=== Checking device: /dev/sdd ===
FARM data not available - likely not a Seagate drive
SMART:
FARM: N/A
RESULT: SKIP

sdc and sdd disk are SSD's, so all good there, disk sda and sdb are 18TB Seagate EXOS'es

Let me know if you need additional information?

[TheRealDevKat]

Hi, I'll take a look once my PC is back together. I'm currently moving and can't do much from a phone

[Lennert-K]

No problem at all, thanks for the swift response, and good luck with your PC!

[Kovelsen]

After encountering this same message I updated smartmontools to version 7.4. This made the tool work correctly.

[goose-ws]

I just submitted a pull request that solves this issue for me by: 1. Building smartmontools from source, and 2. Changing the way the check.sh script behaves. Perhaps it can be helpful to y'all as well: #8

[Tallerik]

I had the same issue on a QNAP with QTS installed.
I edited the entrypoint.sh (using Docker) and added -d sat to every smartctl call and it worked

[TheRealDevKat]

I'm aware that -d might fix some recognition problems with HBA's and I will implement a generic -d argument that passes the value to smartctl.
But I'm currently very busy and won't have time for that for at least another week.

[graealex]

@goose-ws Tried your PR and it still failed to detect any Seagate drives, despite me having 4x Seagate Exos X18. Running the following command (since my Debian natively only offers smartmontools 7.3):

docker run --rm --privileged -v /dev:/dev -v /sys:/sys --entrypoint smartctl farm-check /dev/sdc -l farm

did work though, so it must be something wrong with the check.sh script. I could however confirm that my drives were genuine refurbished, with FARM and SMART power-on hours matching.

[goose-ws]

@graealex If you can provide me the output of your SMART and FARM data I'd be happy to update my search patterns to make sure they match your output -- However, I don't think the maintainer is planning on using my suggested changes, so I'm not sure how useful it would be to do so.

There is, unfortunately, no "standard" for SMART data reporting that I can find. So it's all about gathering as many samples as possible to account for as many patterns as possible. My drives are EXOS x14 and x16. One would think that would be similar SMART output to the x18, but maybe not?

[TheRealDevKat]

This should now be possible in the newest version if you supply -d sat to the script.