Hi Himanshusince1989,
Here is the clarification reference VM auto-starts, also a couple of links to assist you in setting it up...
Nutanix AHV Virtual Machine Auto-Start (Equivalent to VMware Start-up Priority)
In AHV, virtual machine auto-start after host reboot is handled using Virtual Machine High Availability (VMHA). There’s no direct equivalent to the manual VM start-up priority configuration like in VMware ESXi, but Nutanix does offer:
1. VM High Availability (VMHA) Built-In Mechanism
Nutanix AHV supports automatic restart of VMs in case of a host or CVM failure using VM High Availability (HA).
- When an AHV host fails, the VMs on that host are automatically restarted on other healthy hosts in the cluster.
- This functionality is enabled by default in most clusters and does not require start-up priority settings like in ESXi.
- It uses the Acropolis Master Leader process to detect failure and take action.
How it works technically:
The Acropolis Leader keeps track of VM-to-host mapping. When a node goes down, the leader evaluates cluster health and reboots VMs on surviving hosts after fencing off the failed node.
Nutanix Reference: TN-2132 – Virtual Machine High Availability
2. VM Categories (for Prioritization)
In Section “Creating and Managing Recovery Plans”, the document explicitly describes how to:
- Assign VMs to categories.
- Group these into recovery stages.
- Control start-up sequence by ordering categories in recovery stages.
- Set boot delays, scripts, and dependencies per stage.
Official Statement:
“A recovery plan executes in stages, and each stage consists of one or more VM categories. VMs in a given category are started together. You can configure delays between stages or run custom scripts post-stage.”
Nutanix Source: Disaster-Recovery-DRaaS-Guide-vpc_7_3, Recovery Plan Configuration section
Can you explain why you want auto startup? Is ths for when the cluster is booting of for data protection?
- For cluster booting: This is a manual step as nutanix cluster are build to run for every and shutting them down is only for specific maintenance issues.
- Data Protection: You can create recovery plans in Prism Central where you can specify the boot order of virtual machines.