Details
-
New Feature
-
Status: Resolved
-
Major
-
Resolution: Duplicate
-
None
-
None
Description
In a recent customer case we had the requirement to remove corrupted nodes from a repository (instead of rolling it back via oak-run check and subsequently editing the journal.log. The current ad-hoc way of doing so is via the rmNode Groovy script. Since in that case the corrupted nodes where inside a checkpoint we couldn't use this approach though.
Going forward we should implement more robust tooling around this use case:
- Remove corrupt nodes and properties and log their path
- Copy the repository (head state) to a new repository skipping all corrupt nodes and properties while logging their path.
Attachments
Issue Links
- duplicates
-
OAK-5792 TarMK: Implement tooling to repair broken nodes
- Open