我们正在使用mongos 3.2.6并且最近发现当与数据库节点的网络连接失败并在几分钟后恢复时，我们的mongos进程将不会重新建立与我们的数据库节点的连接。
我们的日志如下：
- 网络问题已经开始 -
2017-06-15T08:43:10.723Z I NETWORK [replSetDistLockPinger] All nodes for set config_server are down. This has happened for 2 checks in a row. Polling will stop after 28 more failed checks
2017-06-15T08:43:10.723Z W SHARDING [replSetDistLockPinger] pinging failed for distributed lock pinger :: caused by :: FailedToSatisfyReadPreference: could not find host matching read preference { mode: "primary" } for set config_server
2017-06-15T08:43:10.723Z I SHARDING [Balancer] couldn't create config.actionlog collection: :: caused by :: FailedToSatisfyReadPreference: could not find host matching read preference { mode: "primary" } for set config_server
- 时间过去了类似的套接字和连接异常 -
2017-06-12T08:57:45.806Z I NETWORK [ReplicaSetMonitorWatcher] SocketException: remote: (NONE):0 error: 9001 socket exception [RECV_TIMEOUT] server [<IP>:<PORT>]
- 出现类似的消息 -
- 解决了网络问题并继续显示这些日志条目，但mongos仍无法运行 -
2017-06-15T10:13:37.024Z I ACCESS [conn2704] SCRAM-SHA-1 authentication failed for admin on admin from client 127.0.0.1 ; ReplicaSetNotFound: None of the hosts for replica set config_server could be contacted.
Mongos服务重新启动，它正常工作。
为什么mungos在池停止后没有检查连接到DB？ 网络恢复后如何连接？