Hi Vilius,
I believe the commands you’re looking for are db.getReplicationInfo() and rs.printReplicationInfo(). These commands will output some statistics about your oplog utilization, from which you should be able to estimate how much data is going into the oplog per hour.
Best regards,
Kevin
Hi
Oplog divided by its window size probably gives it. I though there was a possibility to calculate it more accurately.
It can be more accurate if you know exactly how big each document inserted, and also taking into account other administrative entries in the oplog. For example, in MongoDB 3.4 and later, there is a periodic noop writes in the oplog SERVER-23892. However since the oplog is server-wide, it’s probably difficult to have a very exact number.
Best regards,
Kevin