A follow-up repair did not run on a test server. The cause was a mismatch with the convention used elsewhere in the same script: other repair steps tolerate individual failures, and the new ones did not, so a failure would have stopped the rest of the action.
We aligned the new steps with the existing convention so a failure in one repair can never prevent the remaining steps from running. The lesson: new code in an existing script should follow that script’s error-handling convention.
Tested (2506/2506).