As described in SERVER-51397, mongos can take up to 30s to discover that the communication to configsvr replica set node have been lost and to ends all the ongoing request to that specific node. Thus primary_config_server_blackholed_from_mongos.js needs to waits for the mongos to be aware of the failed node before to route commands through it.
- is related to
-
SERVER-51397 Mongos fails to serve requests for 30 secs when losing comm with one config replica set node
-
- Closed
-