Guns submitted for proof have to lock up MUCH tighter than for normal service.
They are generally submitted with no bolting spring and the bolt is pushed home with a brass hammer! If they then come off the face or show any looseness after proof firing, they can undoubtable be said to have failed.
The submitting 'smith then eases off the bite when the gun has passed proof.
My 'smith who does all my reproofing work returns them to me for easing which is no big deal as long as you proceed slowly and carefully and also appreciate that the gun will undoubtable settle in further with the first 100 or so shots.
In recent years the proof houses have both insisted on barrels being struck off on the external surface if there are any irregularities in the surface so you can wave goodbye to original browning or blacking.
Pitting is not the automatic fail that some would have you believe. I have recently had guns pass proof with substantial pitting, albeit not deep.
The kiss of death is pitting 'rib-side', ie in the barrel area under the ribs. This obviously can not be checked for depth nor the wall thickness checked so the proof houses usually fail such guns on 'view'.
Also bulges and dents are a no-no as are cracks in the action and very heavy pitting around the striker hole.
There is no measuring of MWT at proof and you can submit a gun down to any MWT and if it doesn't rivel or bulge, it will pass.
The MWT figures often quoted as a minimum for proof are just guidelines and don't let anybody tell you otherwise.
If your action has a right angle at the joint between flats and breech face (i.e. not radiused), it is more likely to fail at this point: known in engineering terms as a 'stress razor'.
Early actions (pin fires etc) were often thin in this area plus having the right angle and simply will not take the nitro proof pressures. I have had a couple fail repeatedly in the past and had to finally give up on them.
