I understand that perhaps this would work, but is that really the point of the recovery key? If we're using the recovery key post-install on the same installed system, manually unmounting /target etc, is that reflective of the situations the recovery key is supposed to be used in?
I don't have a problem with the process you suggest but in terms of tests for the isotracker, they should be reflective of user experience. It's my assumption, and I may be wrong, but would a user not use the recovery key to recover a drive they no longer have access too by usual methods (on the originally installed system)? Is there a use case in which a user would have access to the original system, but not to the encrypted zfs data?
We can also include both steps in the isotracker tests - unmount and remount from the installed system, then boot into it with an iso and mount the data in a live session.
If I'm wrong in my assumptions, please do tell me :)
I understand that perhaps this would work, but is that really the point of the recovery key? If we're using the recovery key post-install on the same installed system, manually unmounting /target etc, is that reflective of the situations the recovery key is supposed to be used in?
I don't have a problem with the process you suggest but in terms of tests for the isotracker, they should be reflective of user experience. It's my assumption, and I may be wrong, but would a user not use the recovery key to recover a drive they no longer have access too by usual methods (on the originally installed system)? Is there a use case in which a user would have access to the original system, but not to the encrypted zfs data?
We can also include both steps in the isotracker tests - unmount and remount from the installed system, then boot into it with an iso and mount the data in a live session.
If I'm wrong in my assumptions, please do tell me :)