On 2020-09-05 13:41, Hendrik Boom wrote:

> This seems to say we now have two independent levels of bad block
> checking.
>
> When the low-level one fails, the other takes over. I't expect the
> file system to be unlikely to see bad blocks until the drive is hosed.

That is true if the drive is SMART capable _and_ you use SMART tools to
check it. In this thread we saw that SMART over USB is not that
reliable, and in fact I had very mixed results with it even over SATA. I
admit that was in the early years of SATA. So now I rely completely on a
regular 'e2fsck -c -c' to check my drives. Each has a handful of bad
blocks but the count remains stable and the drives usable.

Back when I was trying SMART, it would uselessly spam me in this
situation, while it completely missed an approaching catastrophic
failure.

--
Ian