Purpose: What is VMHA and it’s Master/Slave, requirements of VMHA, Best practices for VMHA
vSphere HA detects failures and provides rapid recovery for the vm’s running within the cluster. Core functionality of VMHA includes hosts and vm’s monitoring to minimize the downtime. When heartbeats cannot be detected HA must be turn on to use FT.
- Quickly bring back up the critical applications in the event of ESXi server failure.
- Decrease downtime and improve availability.
- Examples of business critical application
Exchange / Email server
Sql server /database
Corporate file server and internet/web.
- VMHA monitors not only ESXi host failure but also OS failures
- Supports upto 32 ESXi Servers in a cluster
- VMHA protects all VM guests and all applications.
- VMHA checks not only management network and default gateway during hosts failure but it also checks the datastore heartbeat which was not there in vSphere 4.x but it is there in vSphere 5.x. In vSphere 4.x VMHA only checks management network and default gateway. Because it happens sometimes that the vms are still be serviced as the hosts is running that’s why the heartbeat of datastore has been checked in vSphere 5.x HA become a lot smarter.
Requirement of VMHA:
- Shared storage for VMs running in HA cluster.
- All hosts should have access to all VM N/W.
- Can use DRS with VMHA or just VMHA only.
- All hosts must be licensed for VMHA.
- The cluster must be VMHA enabled.
- All hosts must have a static IP address.
At least one management network N/W but best practice say two with two different IP address that enhance HA.