After you recover the backup images that were taken from the hidden MongoDB nodes, the hidden nodes become primary nodes. Update all such primary nodes in the shard list and restart the mongos process using the following command:

db.getSiblingDB('config').shards.updateOne({ "_id" : "shard1" },{ $set : { "host" : "ShardName/repl1.example.net:27018,repl2.example.net:27018,repl3.example.net:27018" } })

After the recovery process is complete, manually add the secondary nodes to the cluster. Before adding the nodes, ensure that ownership and permissions on the MongoDB data path on secondary nodes are set properly.

For more information, refer to the following article: add-members-to-the-replica-set

After the recovery operation, the mongod or mongos process is started using the configuration files from the /tmp location. Ensure that you move the configuration files to a selected location and restart the services from that location.

Remove the configuration files from the /tmp location so that the restore or recovery operations can restore files at the /tmp location with the same name for different users. If you do not remove the files, the subsequent recovery operation using a different user fails with error 2850 because the configuration files cannot be restored at the /tmp location.

You can add more MongoDB configuration parameters if there are any changes from the backup data that is restored.

Before recovery, if you start the MongoDB services with systemctl command, then systemctl status mongod command may show the mongod status as dead after recovery. This happens because after recovery, the mongod services are brought up with the config file under /tmp location.

In such scenarios, bring up the service again with a use of systemctl start mongod commands.