In VMware vSphere High Availability (HA), the Master-Slave architecture plays a crucial role in ensuring the availability of virtual machines (VMs) in the event of a host failure. Let’s explore how the Master-Slave mechanism works in VMware HA:

1. Cluster Formation:

• When you enable HA on a cluster, one of the ESXi hosts is elected as the Master host, and the remaining hosts become Slave hosts.
• The Master host is responsible for managing the cluster’s state, monitoring the health of all hosts, and coordinating VM failover events.

2. Heartbeat Mechanism:

• To maintain communication and monitor the health of the hosts in the cluster, a heartbeat mechanism is established.
• Each host, including the Master, sends heartbeat signals to the other hosts in the cluster at regular intervals (default is every 1 second).

3. Master Host Responsibilities:

• The Master host is responsible for managing the election process, monitoring the heartbeat responses from all Slave hosts, and determining the health of each host in the cluster.
• The Master host maintains a list of available Slave hosts and their VM workloads.

4. Slave Host Responsibilities:

• Slave hosts receive heartbeat signals from the Master and respond back to confirm their availability.
• If a Slave host fails to receive the heartbeat from the Master within a specified time (default is 15 seconds), it considers the Master as failed, and the election process for a new Master begins.

5. Election of New Master:

• If the Master host becomes unresponsive or fails, the Slave hosts detect the absence of heartbeat signals from the Master.
• The Slave hosts initiate an election process to select a new Master from among themselves.
• The election is based on a priority system, where the host with the highest priority becomes the new Master.
• Host priority can be configured based on factors like resource utilization, host hardware, or administrative preference.

6. Master Duties Transition:

• Once a new Master is elected, it assumes the responsibilities of the former Master, including managing the cluster’s state and VM failover events.
• The new Master takes over the heartbeat monitoring and keeps track of the available hosts in the cluster.

7. VM Failover:

• In case a host fails, the Master host is responsible for coordinating the failover process to restart the VMs on other available hosts within the cluster.
• The Master selects the best-suited host (based on resource availability) to restart each VM, ensuring optimal resource utilization.

8. Admission Control:

• Admission control is a mechanism used by HA to ensure that sufficient resources are available to accommodate VM failover during a host failure.
• Admission control policies prevent VMs from being powered on if there are insufficient resources to guarantee VM failover in case of a host failure.

The Master-Slave architecture in VMware HA ensures that a single point of control is maintained in the cluster, preventing issues like split-brain scenarios and ensuring orderly VM restarts during host failures. The Master host actively manages the cluster and VM failover, while the Slave hosts are ready to assume the Master role if the current Master becomes unavailable. This robust architecture enhances the overall reliability and availability of virtualized environments in VMware vSphere.