How likely is it that large numbers of pods just randomly die as opposed to nodes dying? I also agree with the other poster, freezing/partitioning the network a

How likely is it that large numbers of pods just randomly die as opposed to nodes dying? I also agree with the other poster, freezing/partitioning the network and resuming things later on to see how things fall apart then is a good idea.