When performing validation on redirect role changes a check is made to see if the user is attempting to changing / undo an existing btrfs redirect role. This check is successful and works as intended and as documented within the code, however the message displayed is not consistent with the validation performed.

Additionally the message displayed has no direct validation counterpart. The message in question is:

"Existing btrfs partition found; if you wish to use the redirect role either select this btrfs partition or wipe it or the whole disk and re-assign. Redirecting is only supported to a non btrfs partition when no btrfs partition exists on the same device."

The missing validation check can lead to unconstrained disk role redirection choices that will in turn lead to user confusion. I.e. it is possible to configure a redirect role on a device that contains a btrfs partition that is not redirecting to that partition. This in not an intended behaviour and leads to further buggy behaviour as the user is then presented with an import icon that will fail due to this over-site / missing validation. Re-setting the redirect role to the btrfs partition will then result in the intended and freshly constrained behaviour within the remit of the existing and working validator. And imports will then work as intended.

The code in question is:
https://github.com/rockstor/rockstor-core/blob/master/src/rockstor/storageadmin/static/storageadmin/js/views/setrole_disks.js#L115-L128