Air is compressable while water isn't. That's why it makes water a good testing medium. Water also contains oxygen. The oxygen in water eats metal just as the oxygen in free air.

I have never done a hydro test or fired a boiler that didn't have some weeps here and there, much of which took up when heated. Hell, I had a water fountain that leaked when chilled but was watertight when at room temperature. There are a lot of variables in action here, some of which we don't know about until we try something and learn.

Great discussion, folks.
There may be dozens or more who read and follow without posting.

To apply, establish and maintain a nitrogen blanket is fairly simple and easy.
I know of a few locomotive operators who hand truck a 200 Cubic Foot cylinder
up to a parked and cooling locomotive when it is needed. They connect the hoses
or tubing and open the cylinder valve. They check or set the regulators.
When the steam dome valve is opened, the check valve prevents backflow.
It acts as a vacuum breaker. The boiler draws in nitrogen instead of air.

Folks are using this for a couple weeks to a couple months.

The nitrogen pressure does not need to be maintained below 100 F.
Simply open the valves and give it a 15 minute purge weekly.

Alternate and new technology, better that sulfites and resultant foaming.
A liquid oxygen scavenger that chemically strips oxygen from boiler water.
As stated before, it has been demonstrated, monitored and verified
to protect a large locomotive boiler, larger than 2,500 gallons,
in cold wet layup, for three months or more against oxygen corrosion.

My favorite line from Star Trek;
Scotty: I cann'a change the laws of Physics ( or Chemistry ) .

... But we can learn to work with it and manage it ...