I migrated a replica set from mLab to MongoDB Atlas.
mLab offered an automated process to reclaim disk space after dropping large amounts of documents from a collection. Compacting (or resyncing) basically happened with the click of a button with a rolling node replacement process that does not require downtime.
I didn't find such a process in the MongoDB Atlas UI.
In the mongoDB docs the steps to "automatically sync a member" require deleting data manually from the shut-down secondary, and possibly other steps that seem quite delicate.
What is the easiest way to reclaim disk space on MongoDB Atlas for a replica set with 1 primary and 2 secondaries?