IF the choice of shear pin/roll pin was correct AND it everything worked as designed, AND it was never overloaded, And the design took into account expected life cycle, then the roll pin should not fail within the expected life cycle.
However, deviations in any of the above could result in failure. And in many cases roll pins are choosen so they are the "weak link" and are designed to fail IF something operates far enough outside of normal design parameters. That way something more expensive or catastrophic doesn't happen. In fact sometimes/often they are called "shear pins" rather than "roll pins" for just that reason. It is the mechanical equivalent of a electrical "fuse"
And who knows what the design life cycle was for the pin in question and how close it was operating to it's design load. So, yes, just like a 15A fuse continously operating 15A occasionally will fail, I can see shear pins failing as well. If it fails again in a short period of time, then I'd starting looking for something that has changed and is causing an overload. It isn't so much an issue of "age" as "stress cycles" that will cause it it fail.
Personally I wouldn't normally replace a roll pin with a stronger (e.g. grade 8) roll pin. Now the pin may not fail, and something way more difficult to fix may end up failing.
Kinda like saying, that 15A fuse keeps blowing so I'll stuff in a 30A fuse. Now the fuse doesn't blow, but wires overheat.
