Return of used data

Bingo!
You should go there:

The solution is to optimize your disk subsystem (check & fix, defragmentation, move DB to a system drive/SSD, add an SSD as a storage tier with a PowerShell…). Or increase a timeout for that exact check (and accepting that the node may be disqualified, because of the slowness/hardware issues of this disk, especially for the readability checks…).

then you may add it, save the config and restart the node either from the Services applet or from the elevated PowerShell:

Restart-Service storagenode

P.S. The writeability check only discover that your node would lose most of uploads requests, because it’s too slow to save pieces… However, if you would be forced to increase the timeout for this check above a 5m0s just indicates that you have much more great issue with your hardware, than just crashes… You need to check this disk (S.M.A.R.T. in particular), it could be dying.