[MadMax]

I was thinking something along those lines, but if you really look at how a spring guide sits in a mechbox it's difficult to load the centre of the cross bar as it bears directly on the inside of a mechbox. The hole thru the bar reduces the cross section that so much that I doubt there's enough material to make much a difference in terms of strength in spreading load to the mechbox. The bar looks like it'd yield until the centre of the spring guide bears directly on the box, a kind of load the box is supposed to easily tolerate.

I suspect that there may have been a casting flaw which caused the rear to fail first. This would remove any support for the crossbar so it'd pop in short order. It's possible to tell where the crack ended. Look for two small sharp bent lips on the mechbox. Two small bits of yielded material can indicate a latest region of failure. Fatigue cracks start typically at a flaw like a void or microcrack and propagate until there isn't enough material to support the cyclic load. This region usually has a profusion of parallel ridges caused by incremental crack growth with each fatigue cycle.

At some point the crack grows far enough that the remaining material cannot support another fatigue cycle so you get a fast failure. This typically looks like a glassy surface which ends with two small sharp yielded points where the last material bent out and tore when the chunk falls off. This bent sharp bit is particularly useful for showing where a crack ended as it's often pretty obvious. It may be present on the larger fragment (mechbox) or smaller (bit that fell off).

A failure analysis (with a good magnifying glass or other optics) is necessary to really get to the cause of this failure. Without it it's difficult to discern cause from effect.