- start a new cluster with 3 config servers and one shard
- create a new sharded collection
- do some finds - all are fast (less than a second)
- block the first config server with iptables --DROP
- create a new connection
- try finds on the new and the existing connections - all are fast
- try inserts on the new and the existing connections - all are 5s
Test-case in python is attached (uses my own mongo-docker framework). LogLevel 5 logs too.
- duplicates
-
SERVER-21531 Shard's connection blocks forever when attempting autosplit and first config server (SCCC) in TCP blackhole from shards
- Closed
- is related to
-
SERVER-17617 One config server being down can block read operations on config data for seconds
- Closed