The Self-Service Portal, powered by Calm, allows developers to deploy pre-defined blueprints. These blueprints can be configured to use linked clones, which significantly reduces storage consumption and provisioning time for disposable VMs by sharing a base image. This is ideal for rapid, iterative development and testing cycles.

The VM’s share of CPU and memory resources is directly influenced by its QoS (Quality of Service) Policy priority setting. Changing it from ‘Medium’ to ‘High’ is a configuration change that takes effect immediately without requiring a reboot or migration, giving the VM a larger share of contested resources on its host.

The replication factor (RF) setting on a Protection Domain (used for Metro Availability or Async DR) is independent of the cluster’s local RF. The PD’s RF2 indicates two copies are maintained for the purpose of replication. If one copy is lost, the other copy still satisfies the RF2 requirement, so the data remains available and protected within the context of the replication stream.

Categories (tags) can be applied to VMs and hosts. You can then create VM-to-host affinity rules (either ‘should’ or ‘must’ run on hosts with specific categories) to influence or mandate placement. By categorizing the high-performance hosts with tags like ‘SSD_Tier’ and ’10Gb_NIC’, you can ensure critical VMs are placed accordingly by the scheduler.

Nutanix Flow Microsegmentation allows you to define security policies that control traffic between VMs based on categories (e.g., ‘App-Tier-1′). This inspection and enforcement happen distributively at the source vSwitch (AHV), regardless of the VMs’ IP subnet, satisfying the security requirement without network re-addressing.

A Calm Runbook is a sequence of tasks (scripts, VM operations, etc.) that execute in a defined order. It is the primary construct for orchestrating multi-step workflows, such as application lifecycle management (launch, patch, scale, delete) or the specific provisioning and configuration sequence described in the question.

The Nutanix cluster is designed to auto-heal from component failures. The first step is always to verify the alert in Prism. The Curator service will automatically trigger a rebuild of the data from the failed disk using the remaining replicas (RF2/RF3). Manual intervention is typically not required and should only be done under guidance from support if auto-heal does not proceed.

The Self-Service Portal in Prism Central is the dedicated interface for this purpose. You can publish approved Calm blueprints (which define the multi-VM application) to a catalog. Developers can then launch, manage, and delete these consistent environments from the portal. You can also assign project quotas to control their resource consumption.

In a two-node ROBO cluster, RF2 cannot be maintained if one node fails (as you’d need 3 nodes for N+1). The Witness VM, deployed at a third location (e.g., corporate HQ), acts as a tie-breaker. It contains no data but holds cluster metadata, allowing the surviving node to form a quorum and continue serving data, preventing a split-brain scenario.

The Self-Service Portal, powered by Calm, allows developers to deploy pre-defined blueprints. These blueprints can be configured to use linked clones, which significantly reduces storage consumption and provisioning time for disposable VMs by sharing a base image. This is ideal for rapid, iterative development and testing cycles.0