This ticket changes RemoveShardProgress document field returned as part of transitionToDedicatedConfigServer command. Earlier when we waited for pending range deletions, we returned their count. Now we return the single range deletion document as BSON we wait for orphanCleanupDelaySecs to complete on.

I have selected teams impacted based on my assessment - I could be wrong here so I am erring on the side of overcommunication.

Description of Linked Ticket

Today, when a cluster transitions from embedded to dedicated config servers we use transitionToDedicatedConfigServer() which uses removeShard(). 

We drain sharded collections and then use range deletions to remove any data from the embedded config server safely in case it gets added back to the cluster as a config shard. For any other removeShard operation, we don't complete range deletions because the shard will be removed. 

Waiting on range deletions to transition to dedicated config servers is unnecessary. Atlas cannot complete the transition and reduce the tier to an M30 for a config server until range deletions are complete which can be very slow and resource intensive.