I’ve seen a lot of discussion about the internal technical mechanisms of why an SSD might fail, which is fascinating by itself. But I haven’t seen a whole lot about how to predict that failure.
-
Does a failed SSD really just “disappear”? Either not found at all, or found but entirely useless with everything gone, when it was working perfectly yesterday?
-
Does it slowly decrease its capacity until there’s no more free space and random data starts “falling off”, but the rest is okay? (wear leveler doing all it can, with a decreasing number of good cells)
-
Does new data start to read back incorrectly while old data is still good? (might lose a day or two of work, and maybe some recent settings changes, but nothing else)
-
Does random data start to read back incorrectly regardless of age, while other data is still correct? (not quite “disappeared”, but not really usable either)
-
Does the entire drive suddenly switch to read-only? (bad for ongoing work, but at least you don’t lose anything…unless the OS must write as part of the boot process, in which case that will fail and appear as a completely dead system instead of simply read-only)
-
Something else?
I know there are explicit tools to check the detailed health of a drive, but a casual user is probably not going to use any of them. And they’re probably not going to back up their system either, until they see a definite failure coming. And even then, they’ll have to wait for shipping (or a trip to the store) because they don’t have one on hand.
And yes, I know the standard answer is to call them idiots and their data unimportant, because they didn’t already have an entire second computer just lying around that they spent as much effort to keep in sync as they did to use their primary. (I know that’s not the intent of that answer, but it is how they interpret it: a computer to them is not a collection of parts that can have one or more of them swapped out, but a monolithic whole device…)
But anyway, is there actually something that they can use, that gives them enough time to choose the cheap slow shipping option on a new drive (or whole computer) while continuing to use the failing one, and still be able to copy everything over with zero loss once it arrives? Maybe something like a steadily decreasing overall size while continuing to work normally otherwise, or one of the other less-catastrophic options above…
Do modern systems monitor the details automatically and give that warning? If one OS does and another doesn’t, that would be nice to know too.
