How dangerous are the following administrative tasks to my server and repositories?

I’m trying to make the /registry API and the web UI align and show the same things. This is an addendum to the problem I posted in this thread.

I, unfortunately, do not have a QA instance of Nexus that I test running some administrative tasks against. I working against the live, production Nexus instance here. So I’m hesitant to run potentially destructive repair commands.

How dangerous are the follow administrative tasks? Do I risk losing my database? Do I risk losing any blobs on disk? Do I risk locking my Nexus server up for an extended duration? The docs leave a lot to be desired here when it comes to the risks associated with these operations.

Repair - Reconcile component database from blob store – this appears to to update the OrionDB to align with files on disk and it seems to be one way. That is, things on disk that aren’t the DB will be added only. Is that true?

Repair - Rebuild repository browse – this appears to rebuild only the tree view that’s used in the web UI – is this correct? It wouldn’t impact /registry API calls?

Repair - Reconcile npm /-/v1/searchmetadata – this one is unclear to me. npm is calling /registry to talk with Nexus. But does /registry translate to a /v1/searchmetadata call under the hood? Is there any danger associated with this operation?