SERVER-101261 is using the config fuzzer to stress timeseries collection configurations to induce more reopening. However, a small number of tests (< 10) inserted enough measurements to meaningfully exercise the reopening path.

There were limitations with changing some parameters in SERVER-101261 that could enable us to account for these (specifically, decreasing timeseriesBucketMaxSize and timeseriesBucketMinSize while seeming to be helpful for inducing rollovers but led to much smaller numbers or reopenings).

We should evaluate if some of our concurrency timeseries tests can insert more measurements so we can see test more rollover behavior.