I just finished watching the discussion acknowledging the controversy surrounding removing the SMART test function being removed from the GUI in 25.10: https://www.youtube.com/watch?v=IuKwwPp8K-U&t=4s
I came in frustrated but leave attempting to keep an open mind (Apple comes to mind, pulling controversial features in the past that were often [not always] in hindsight, the correct decision), but with some questions:
- I understand ZFS and TrueNAS monitor more health components behind the scenes, including some SMART functionality, instead of just pass/fail, which is the reason for GUI removal. But if I purchase a used/refurbished disk, in the past I have always, before anything, run a long SMART test against it. What is the best practice with this now, just insert it normally and wait, or do it through the CLI? Is
badblocksuseful here? Or do I just runmidclt call disk.smart_test LONG ...? That command returnsnullbut then my disks go crazy so I guess something is happening. Which leads me to my next question: - How can I see the status and/or results of past tests?
- How do I get notified of integrity failure? Under System → Alert settings I’ve seen a few that suggest:
- Failed Selftest (Hardware)
- Uncorrected Errors Detected (Hardware)
- Pool Status Is Not Healthy (Storage)
- Are these “behind the scenes” tests are thorough as a LONG SMART test?
Thanks!