Hi, running raidz2 x 7 x 20TB, 1 cache SSD, 1 log SSD. I migrated to 24.10.0.2, and after that was complete, on the dashboard I was seeing that I had 1 ZFS error, on the SSD cache disk. ‘sudo zfs status’ shows one read error on that disk. ‘zfs clear’ does not remove the error. SMB shares and apps seem to be working, but I’m not clear if there’s really a problem with the cache or if it’s just cosmetic.
What should I try to verify that disk, and clear that ZFS error?
A reboot cleared the ZFS error.
A read error on a cache disk would result in a cache-miss, and the read being serviced from the pool vdevs.
Check the SMART data of the disk perhaps, see if it has incremented an error counter for remapped sectors.
I ran a short test, no error.
Short tests are almost useless. And remapped sectors don’t necessarily show up as test failures.
You’ll want to look at the smartctl -a /dev/sdX output at the least, potentially smartctl -x /dev/sdX for the full output
