We saw an issue in which we have bg-save disabled with save = "". This is a cluster setup 6 master 6 slaves. Now when we do a failover. The slave takes over as master but also start doing bg-save. When we check to save settings it is still empty. I need help here.

Redis version: 3.2.1

Logs: ``` 1:S 28 Apr 08:30:49.314 # Manual failover user request accepted. 1:S 28 Apr 08:30:49.405 # Received replication offset for paused master manual failover: 263533816629 1:S 28 Apr 08:30:49.435 # All master replication stream processed, manual failover can start. 1:S 28 Apr 08:30:49.435 # Start of election delayed for 0 milliseconds (rank #0, offset 263533816629). 1:S 28 Apr 08:30:49.535 # Starting a failover election for epoch 63. 1:S 28 Apr 08:30:49.538 # Currently unable to failover: Waiting for votes, but majority still not reached. 1:S 28 Apr 08:30:49.538 # Failover election won: I'm the new master. 1:S 28 Apr 08:30:49.538 # configEpoch set to 63 after successful failover 1:M 28 Apr 08:30:49.538 # Connection with master lost. 1:M 28 Apr 08:30:49.538 * Caching the disconnected master state. 1:M 28 Apr 08:30:49.538 * Discarding previously cached master state. 1:M 28 Apr 08:30:49.710 * Slave 10.101.70.247:6379 asks for synchronization 1:M 28 Apr 08:30:49.710 * Full resync requested by slave 10.101.70.247:6379 1:M 28 Apr 08:30:49.710 * Starting BGSAVE for SYNC with target: disk 1:M 28 Apr 08:30:50.486 * Background saving started by pid 17519

Comment From: antirez

This is normal, Redis requires to BGSAVE for replication. New versions of Redis are able to perform diskless replication however.