We currently have a bug in our republishing event process post migration test. It looks like it could be a race condition based on fetching the old publish state rather than the new, thus passing the wait and asserting on data that hasn't yet been refreshed. It normally passes, but this edge case could be a cause for the sporadic failures.

We should fix the republishAllEventBasedProcessesAndEnsureTheyArePublishedAndHaveInstanceData test by fixing the test waiting condition and making sure that the old publish state and its event index is properly deleted before making assertions