high-availability.storageDir clean up?

[Elias Levy]

I noticed in one of our cluster that they are relatively old
submittedJobGraph* and completedCheckpoint* files.  I was wondering at what
point it is save to clean some of these up.