We are using MongoDB Community Edition (6.0.1) (PSA) and planning to take a backup while the database is running on a live secondary node(full cluster level backup). Is it possible to do so without impacting production? Can someone please provide guidance and a backup plan for the secondary node? Thank you
Welcome to the MongoDB community. You might choose to schedule backups during a time when the workload is low. If you can’t find a suitable time when the workload is low, you could add the following option to make backups:
In case of disk snapshot, you need to flush cache data and lock writes so that the snapshot is consistent. Then your majority writes will always timeout.
In case of mongodump, you can track ongoing oplog entries during backup (backup may take longer in case of heavy write traffic), but data back up will cause heavy disk read. So if the secondary node is very busy (in terms of read and write), performance can be impacted.