Some more:
Kutter, I think you already have answered your question: it is because Krags operate at much lower pressures the the .30-06. The danger is caused by the intersection of 19th century metallurgy with 20th century cartridge pressures.
It is really hard to discuss things like this (rifle safety) rationally and scientifically because the data is just not sufficient. Once we get away from the military use of the rifles, accidents are only occasionally reported, so who really knows what the failure rates are? What we are left with are stories, web postings or magazine articles and our own personal experiences. All that adds up to not much more than mythology.
With the that disclaimer, it is still apparent that the conventional wisdom about these actions changed dramatically at some point. Mr. Petrov's magazine quotes are undoubtedly accurate but they pre-date the official board which examined the matter in 1927. It is correct that the arsenals experimented with reworking these actions in the early 20's, but by early 1926 they had both decided that that approach would not work (see Hatcher's book published in 1947). I have no comment on the commercial reworks by Sedgley, I know nothing about them beyond what was advertised. Crossman writing in 1931 (The Book of the Springfield) says (in some pretty politically incorrect language) that only a fool could break a low number rifle. Twenty years later Roy Dunlap in a supplement added to that same book says: "The low-number (below 800,000) Springfields are close to oblivion, I hope. They just aren't safe with any cartridge in the 50,000 lb. pressure class, which takes in practically all modern bolt-action cartridges above the .22 Hornet class." Dunlap's over the top statement is on par with the internet quote Mr. Petrov has already given us. What could have caused this change? Could it have been Hatcher's book published in 1947? Were the 1927 board findings well known before that point? I don't know.
I have personally have never experienced a case failure or pressure event with a Springfield. I have only experienced two case body splits in my lifetime: one in a pre-64 model 70, and one in a High Wall with a Mann-Niedner firing pin. In both cases the split was lengthwise on the body (but not to the head) and the result was the same: minor gas leakage, a wisp of smoke, and no damage to me or the rifle.
I do agree with Mr. Petrov that the listed action is almost certainly a double heat treat one, even if it is quite close in number to the change point.
