Richard Elling

I don't understand why you have to destroy snapshots. Can you explain?

It is absurd to delete data just to make an error message go away. For an operations team, just annotate as an exception (we don't care about this particular error...

It is true that if you have the original file, then you'll know the correct data and with the block pointer we know the compression and DVAs. So it is...

bugs notwithstanding. The CLI design will be challenging, as will making it completely idiot-proof.

@aletus it is not required to change ZFS to implement this functionality. It can all be done in userland.

@aletus simply find the block in the file that is corrupted (dd can do this) and write a new block (dd can also do this). Step and repeat until dd...

this requires block pointer rewrite

see `zinject`

NB, some module parameters are done with module_param(), others with module_param_call(), and more with ZFS_MODULE_PARAM()

Perhaps we can be more clear in that if `zfs_dirty_data_max` is set, then `zfs_dirty_data_max_percent` is ignored. There are several other similar max vs percent precedents.