Reattempt initially failed Sentinel connections using Master/Slave connections

[mp911de]

Master/Slave Connections can be established as long as at least one Sentinel is available (see #304). lettuce opens Pub/Sub connection upon connect to receive configuration updates. Investigate on possibilities to utilize Pub/Sub to obtain configuration updates on appearing/disappearing Redis Sentinel nodes for adaptive Sentinel monitoring.

[mp911de]

See also: https://groups.google.com/forum/#!topic/lettuce-redis-client-users/Rt1W1ciXfIU

[mp911de]

Sentinel Pub/Sub messages +sentinel and -odown/-sdown are good signals to discover changes in Redis Sentinel state changes (added Sentinel, Sentinel back up).

Consider we have two sentinels configured, and one of them is down at the time establishing the Master/Slave connection. Assuming a few things are set up the right way (firewall rules, reported IP addresses) it could be possible establishing the additional Sentinel connection from the reported events and payload reported in the event. This implies at least that the sentinel hostname resolves to the same IP address to connect the specified sentinel node. Not limiting/filtering sentinel nodes, that get available could easily lead to connections to possibly unwanted nodes.

The better alternative might be to use these signals to reattempt connections to nodes which failed to connect initially.