HGST SAS SSD gives Sense Key : Recovered Error [current] and Add. Sense: Grown defect list not found

[Frank Tarczynski]

I have 3 HGST SAS 1.6TB SSDs. They each have different versions of firmware.

All of them seem to work fine and smartctl reports that everyone is fine too.

I see this from just one of the SSDs:
[36687.895038] sd 0:0:1:0: [sdb] tag#116 Add. Sense: Grown defect list not found
[36987.893382] sd 0:0:1:0: [sdb] tag#97 Sense Key : Recovered Error [current]
[36987.901089] sd 0:0:1:0: [sdb] tag#97 Add. Sense: Grown defect list not found
[37287.902935] sd 0:0:1:0: [sdb] tag#100 Sense Key : Recovered Error [current]
[37287.910706] sd 0:0:1:0: [sdb] tag#100 Add. Sense: Grown defect list not found
[37587.907537] sd 0:0:1:0: [sdb] tag#89 Sense Key : Recovered Error [current]
[37587.915244] sd 0:0:1:0: [sdb] tag#89 Add. Sense: Grown defect list not found

This is on TrueNAS Scale (5.10.93+truenas kernel) with an LSI 9300-8i SAS card.

Does anyone have some pointers to what this means and how to fix it?

Frank

 

[redeamon]

Since you have 3, swap this drive and another one. See if the errors follow the drive or the port.

 

[Frank Tarczynski]

Already done and It follows the drive. Booted the TrueNAS box from Ubuntu 20.04 on a USB stick and I don't see any errors from smartmontools/smartctl from any of the SSDs.

I came across some posts suggesting that this could be a linux kernel issue in older kernels.

Does anyone have any knowledge about it?

 

[redeamon]

Could be something with smart tools then see- #1448 ("Defect list not found" SAS SSD) – smartmontools

 

[John T Davis]

tl;dr: With possibly the same disks, I'm seeing the exact same errors, but only on the drives with the HP OEM firmware as compared to the HGST firmware.

I hope this isn't considering necro'ing, but this was the first result on Google for this error, and I wanted to contribute what I've seen.

1. OS: Proxmox 8.1.4 (uname : 6.5.13-1-pve #1 SMP PREEMPT_DYNAMIC PMX 6.5.13-1 (2024-02-05T13:50Z) x86_64 GNU/Linux ).
2. HBA: LSI 9500-16i (Latest Firmware, BIOS, UEFI, PSOC)
3. Impacted disks:
   1. HGST HUSMM1616ASS200 (1.6 TB)
   2. The catch: They're HP OEM drives, and report a model number to the system of HP MO1600JEFPC. Which I would not have bought but the eBay seller didn't specify that, and I was dumb enough not to realize this would be a problem later.
   3. HP Firmware is a mix of HPD1 and HPD2.

I still haven't set up a pool yet with these, so I don't know if this is going to be a problem or not. I'm actually still working on trying to figure out how to update to the firmware on all of these drives. I think I almost have it nailed down for HGST, but trying to find firmware downloads (or information on whether i can cross-flash them to the HGST firmware) is proving a bit of a nightmare.

My current theory is that on startup (as these messages are not sent to the console when I manually trigger `smartctl`), these errors/warnings are generated and sent to the console. They're visible in the `dmesg` output.

Sorry I don't have more useful information. But, I am motivated to cross-flash the firmware over this.
Disks shouldn't throw strange warnings and errors if it can be avoided, and I have a sneaking suspicion that these errors come down to some variant of "This HP re-branded SSD is not in an HP server/attached to an HP re-branded HBA.")