Disk health shows an alert for failed SMART tests despite subsequent tests completing successfully

TrueNAS General
1

Hello,

I have an issue where 2 disks had failed SMART tests, which I believe occurred because I was running too many SMART at the same time.

I’ve reran these SMART tests on the two disks with failed tests and they have both passed without issue twice. I have also confirmed in smartctl that the disks are healthy as well as confirmed no ZFS errors have occurred.

This tells me that these disks are likely healthy I believe but the disk health still shows an alert. How can I clear this alert? I saw there was a bug regarding this last year for aborted SMART tests but nothing about failed with subsequent passes. I’ve updated to the latest (ElectricEel 24.10) and rebooted but the alert still is not cleared. Any ideas how can I manually clear this alert?

Thanks!

2

I’ve got a similar problem here. I had to shutdown the machine while it was running a load of long smart tests, and they all flagged as failed.

I now have 11 disks marked as having failed SMART tests, although I’ve subsequently run long tests to completion on all of them.

(I’ve also got two manually initiated smart tests “stuck” in the UI, but that’s another long-running bug, afaik)

3

I never managed to find a way to clear these false failed SMART tests. However I did learn that TrueNAS reads these from the drive itself and that drives only retain the most recent 21 tests.This means after 21 succeeded tests this warning should go away in TrueNAS. I havent run enough tests yet to confirm that, but that is what I read.

4

I can confirm the above: successful tests eventually “pushed” the failed tests off the bottom of the history, and everything is reporting as okay again in multi-report and the truenas webui