This is a very grey area indeed. It depends on the local CAA how far trhey go to accept these kind of "rectification" if they are detected at all.
Also it is normal that after some corrective actions and the defect does not appear anymore on ground that the subject is closed in the TechLog, a second or even a third re-occurrence is sometimes normal.
7 times, however, is suspicious and it is more likely that "time buying" from maintenance side is on the agenda.
It may be interesting to see the limitation on APU INOP as this may be the next step.