Hi, On Tue, Jan 02, 2024 at 08:17:55PM -0700, Charles Curley wrote: > On Wed, 3 Jan 2024 00:29:42 +0000 > Andy Smith <a...@strugglers.net> wrote: > > If a SMART long self-test came back clean then it already has been > > re-mapped as a long self-test reads every user-accessible sector. > > I'm not so sure about that. See the journalctl output at the bottom of > this email.
I don't see anything but smartd repeatedly warning you about the 1 pending sector. As I said, it's annoying when a drive doesn't decremement its pending sector count after remapping. If you can read the whole drive then it certainly has been remapped (or was a transient error that isn't "pending" any more). None of the logs you presented show any error coming from the drive, but then they won't as you've only selected logs from smartd. smartd will complain about that 1 pending sector count until the end of time unless: - Drive just decides to clear it, or; - You reconfigure smartd All smartd is doing here is reading the attributes of the drive and reporting them to you. It will never show you the actual error that caused those attributes to change. > > Like to live dangerously, huh… > > No. That's what fast networks, good and multiple backup programs, a > good RAID array on another computer, and multiple off-site backups are > for. It's not my view as in my experience storage is one of the most failure-prone parts of a computer, and an outage from non-redundant storage typically annoys me more than making it redundant does. That is in most cases really easy and cheap these days, so I do regard going without it as living dangerously. Not always a good cost-benefit trade off though, I grant you. Thanks, Andy -- https://bitfolk.com/ -- No-nonsense VPS hosting
