Jedis behavior during Elasticache redis failover

[Julian Medina]

Hello all -

During a planned or actual failover scenario in elasticache redis we're experiencing the following behavior.

Jedis switches to new primary node (at a very poor performance level creating thousands of new connections and spitting out large amounts of errors), e.g.

• redis.clients.jedis.exceptions.JedisClusterOperationException: Cluster retry deadline exceeded.
• redis.clients.jedis.exceptions.JedisConnectionException: java.net.SocketTimeoutException: Read timed out
• redis.clients.jedis.JedisFactory: Error while close

And after recovery of the original primary node jedis continue to create thousands of connections with the secondary failover node without ever recovering the original performance.

[Julian Medina]

Duplicate from other thread, feel free to delete.