It depends on how pervasive the issue is. Is it just two slightly nonconforming holes, or is it random sampling that suggests many nonconformances? It is not r

It depends on how pervasive the issue is. Is it just two slightly nonconforming holes, or is it random sampling that suggests many nonconformances? It is not really feasible to test an installed rivet without destructive testing, because usually we are talking about blind rivets, which are remarkably complex fasteners. Think of them more like "small, ultra-precise one-time use machines", not "nails". This is part of the reason aircraft have so many rivets: each rivet is often capable of supporting a substantial fraction of the load where it is installed (sometimes even the entire load), but we install a huge number of them. The fact that this defect was caught during what seems like normal inspection processes also moves it closed to being a "non-issue". Manufacturing defects happen all the time and liability is massive, so there are extensive inspection and quality regimes in place.