I’m to write some operating manual for an application relying on MongoDB for its database. As far as my simple tests are concerned (on a single node), stopping and starting MongoDB on Linux is done with the “systemctl” command:
# systemctl stop mongod # systemctl start mongod
But the customer I’m writing this manual for is using a 3 nodes Replica Set cluster (one primary and one secondary node on Linux and one arbiter on Windows). Is the procedure still the same in that case ? I.e stopping and starting MongoDB on each server ? And is the order important ?