I have tested on the server. Since a mongod node is running, a journal file needs to be maintained for database operation replay. This journal file needs to occupy a certain amount of disk space, so when the disk space is insufficient, the mongod node will exit. My server originally had 4 mongod nodes and a configsvr running on it. However, after the disk became full, only 3 mongod nodes were left.
After sharding, the data is inserted into shard 1. Explain that there is a problem with sharding
sh.status() Check it
If the shard key is not established correctly, the sharded data will be uneven
But it’s not absolutely uniform
I have tested on the server. Since a mongod node is running, a journal file needs to be maintained for database operation replay. This journal file needs to occupy a certain amount of disk space, so when the disk space is insufficient, the mongod node will exit. My server originally had 4 mongod nodes and a configsvr running on it. However, after the disk became full, only 3 mongod nodes were left.