00 /What You Will Learn
By the end you will be able to
- Explain the difference between subjective down (+sdown) and objective down (+odown).
- Understand how quorum (2 of 3) prevents false alarms from a single faulty monitoring link.
- Trace the 4-step failover process: detection, leader election, replica promotion, and reconfiguration.
- Verify how a resurrected primary is demoted to replica to prevent split-brain dual-master writes.
- Configure
min-replicas-to-writeto limit the lost-write window during an asymmetric network partition.
Before you start
Finish Lab 06 (Redis Basics) and Lab 23 (PostgreSQL Replication Lag) first. You should know what primary-replica replication is and why distributed nodes need health checks. Docker Desktop is required.