On Tue, Apr 20, 2021 at 03:18:05PM +0800, Haotian Li wrote:
> If some I/O error occured during e2fsck, for example the
> fibre channel connections are flasky, the e2fsck may exit.
> Try again in these I/O error cases may help e2fsck
> successfully execute and fix the disk correctly.
Why not fix this by retrying in the device driver instead? If the
Fibre Channel is that flaky, then it's going to be a problem when the
file system is mounted, so it would seem to me that fixing this in the
kernel makes a lot more sense.
- Ted
