I recently had a problem with a RAID1 element as reported by smartctl (unreadable sectors on /dev/sdb). I did a fail, remove, add on the drive and it seemed to start syncing, but then failed again a few hours later. I swapped out the drive and tried again, with the same result. The chassis has 4 hot-swap SATA bays, so on a hunch, I did a fail, remove on /dev/sdb and moved the new drive to bay 4, then added it to the array again as /dev/sdd. The resync worked correctly and everything is functioning normally again. The problem is apparently a bad bay, SATA cable, controller chip, or chip driver. (not the first time I've seen this on ASUS 1U entry-level servers) There is one little fly in the ointment though. LogWatch is still reporting SMART errors on /dev/sdb, which is now an empty bay. I can only assume that it's also NOT checking the new location /dev/sdd. My question is this: what process is starting smartctl on a daily (hourly?) schedule and where is the config file that tells smartctl to look at a particular set of drives? I guess I should know this, but somehow I'm drawing a blank.