Oracle HDD 0x96, 0x97 are false alarm?!

[wittyet]

I got 20 Oracle sun branded HGST 4T SAS, and some of them are giving 0x96 and 0x97 errors. I was kind of paranoid, and did some search online, I did find something interesting and hopefully useful.

TLDR: It seems the codes are caused by firmware and are considered false positive. the "REAL" predicative failure code is 'asc = 0x5d/ascq = 0x90', that's when you should consider replace the HDD.

Here are the links related to the code:

Oracle ZFS Storage Appliance: Handling DISK-8000-0X FMA 'predictive-failure' Events With 'asc=0xb/ascq=0x97'

Oracle ZFS Storage ZS5-4 - Version All Versions and later: Oracle ZFS Storage Appliance: Handling DISK-8000-0X FMA 'predictive-failure' Events With 'asc=0xb/ascq=0x9

support.oracle.com

Oracle ZFS Storage Appliance: Handling DISK-8000-0X FMA events with 'asc=0x5d' or 'asc=0xb'

and one of them did mention to upgrade "appliance firmware", but is there anyway we can do this?

 

[xeromist]

We're not running Oracle appliances so the equivalent would be either some sort of patch to TrueNAS that understands this code or perhaps something that would allow us to hide/suppress certain codes. I'd peg a fix as unlikely as this is an edge case and not worth dev time, but one can hope. With the latter there would be people who would use it to ignore valid warnings and end up with data loss, so probably also a non-starter.

I too have some of these drives and I also get false positives. One option would be flashing the drives to a different firmware but in my research it seems that these drives are pretty locked down and there's only one guy who has managed to hack them and charges per drive to re-flash. So far it hasn't bothered me enough to reach out and get a quote.