We use Redisson (3.40.0) RReadWriteLock on a Redis Cluster. As traffic grows, we sometimes see:

None of slaves were synced. Try to increase slavesSyncTimeout setting or set checkLockSyncedSlaves = false.

My understanding is that Redisson waits for a certain number of replicas to acknowledge the write (similar to Redis WAIT) before it considers the lock acquired.
Questions:

1. If I disable checkLockSyncedSlaves, will this reduce load on Redis and improve lock acquisition latency under high traffic?
2. Is the main downside that, during a failover, the newly promoted master may not have the most recent lock state (since replication is asynchronous), so we could “lose” the lock?
3. We use Redisson in cluster mode with default reads set to slaves. If we disable the checkLockSyncedSlaves flag, will it affect our reads? From what I understand, locks always run on the master. This flag mainly protects us during failover when a replica becomes master. Since we already read from slaves and accept that risk, it probably doesn’t matter much for us.
4. Would moving to Redisson PRO materially improve lock throughput/latency in this scenario, or is this mostly a replication/consistency trade‑off independent of edition?

Any best practices for tuning slavesSyncTimeout / required replica count vs. latency would be appreciated.