HA restarts VM’s after failure of an ESXI host. But the time taken by HA to restart VM’s is different in case of a slave ESXi failure and master ESXi failure. We will discuss here the case when master ESXi has failed.

In case of a failure of a master ESXi, restart of VM’s are delayed TILL the time a new master is elected because only a master can PERFORM VM restart.

The timeline is explained as follows:

• T0 – Master failure.
• T10s – Master election process initiated.
• T25s – New master elected and reads the protected list.
• T35s – New master initiates restarts for all virtual MACHINES on the protected list which are not RUNNING.

At T0 seconds master ESXi has failed, the election process is initiated by slave ESXi hosts after 10 seconds at T10. At T25 the newly elected master first reads the protected list file to find out which VM were protected by HA and are currently not running. At T35 seconds the master ESXi initiates the VM restart.

HA restarts VM’s after failure of an ESXi host. But the time taken by HA to restart VM’s is different in case of a slave ESXi failure and master ESXi failure. We will discuss here the case when master ESXi has failed.

In case of a failure of a master ESXi, restart of VM’s are delayed till the time a new master is elected because only a master can perform VM restart.

The timeline is explained as follows:

At T0 seconds master ESXi has failed, the election process is initiated by slave ESXi hosts after 10 seconds at T10. At T25 the newly elected master first reads the protected list file to find out which VM were protected by HA and are currently not running. At T35 seconds the master ESXi initiates the VM restart.