Nutanix NCP-MCI flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Foundation

Nutanix imaging tool that discovers bare-metal nodes over IPMI or Redfish, wipes disks, installs the hypervisor and AOS, and creates the cluster from a jump host.

Prism Element

Per-cluster management UI and API running on the CVMs. It handles local storage, networking, VMs, alerts, and health for a single Nutanix cluster.

Prism Central

Separate VM or scale-out set that manages many registered Prism Element clusters, adding categories, projects, policies, and multicloud dashboards.

Cluster VIP

Virtual IP that always points to a live CVM, giving a single stable management address. It must be a free IP in the CVM subnet.

RF2 minimum nodes

Replication factor 2 requires at least three nodes so one node can fail while two copies of every extent remain available.

NCC health checks

Nutanix Cluster Check library of diagnostic tests covering disks, network, time sync, and services. Run it first when troubleshooting an unhealthy cluster.

LCM upgrades

Life Cycle Manager performs AOS, hypervisor, firmware, and NCC updates with pre-checks and staged rollout. Failed pre-checks block the upgrade.

Pulse telemetry

Formerly CallHome. Sends cluster health and configuration data to Nutanix support, enabling proactive case creation and automated health insights.

Alert severity levels

Nutanix alerts are classified as Critical, Warning, or Info. They can be routed to email, SNMP, syslog, or webhooks and acknowledged in Prism.

Prism Central registration limit

A Prism Element cluster can register with only one Prism Central at a time, and PC must have network reachability to each PE VIP.

br0 default bridge

Host bridge carrying CVM and hypervisor management traffic. Guest VM traffic should use a separate VLAN-backed network to isolate management.

balance-tcp bond mode

Bond mode that requires an LACP port-channel on the upstream switch. Misconfiguration breaks connectivity, so verify switch config before enabling.

balance-slb bond mode

Bond mode using source MAC hashing with no switch LACP configuration. It works with standard switch ports but does not aggregate a single flow.

AHV VLAN network

VLAN-backed virtual network in Acropolis. A VM NIC attaches to a network, and the VLAN ID defines its broadcast domain.

Flow microsegmentation

Prism Central security feature that applies distributed firewall policies to VMs by category rather than by IP address, across clusters.

Storage pool vs container

A storage pool groups physical disks; a container is a logical namespace presented to the hypervisor as a datastore. Multiple containers can share one pool.

Erasure coding EC-X

Container setting that reduces capacity overhead by storing parity instead of full replicas. It suits cold data and needs a minimum node count.

Distributed storage fabric

AOS log-structured, extent-based storage layer. Writes land in the local CVM staging area, replicate to peers, then are acknowledged.

Storage QoS IOPS limit

Policy that caps IOPS per VM or per container, preventing one workload from consuming cluster-wide storage performance.

Nutanix snapshots

Metadata-based point-in-time copies that are space-efficient. They are not full copies, but changed blocks consume space over time.

VM categories

Key-value pairs assigned to VMs that drive Prism Central policies for protection, recovery, security, and cost across clusters.

Nutanix Guest Tools

In-guest agent enabling file-level restore, application-consistent snapshots, and self-service restore for supported guest operating systems.

Anti-affinity rules

Scheduler rules that keep specified VMs on different hosts. They can be should or must strength and improve application resilience.

Prism Central Projects

Self-service environment with quotas, role-based access, and isolation. Users provision VMs within a project without full cluster admin rights.

Protection policy RPO

Rule defining snapshot schedule, retention, and replication target for protected VMs. RPO sets the maximum acceptable data loss window.

Metro Availability

Synchronous replication between sites with low latency, giving near-zero RPO. It requires a stretched network and compatible AOS versions.

Recovery plan

Prism Central DR object defining VM boot order and network mapping for failover. Test it with a planned failover before relying on it.

NearSync replication

Replication mode with RPO measured in seconds to a few minutes, requiring a compatible AOS version and sufficient network bandwidth and latency budget.

CVM memory pressure

High memory use on the Controller VM, a common cause of cluster instability. Monitor it and resize the CVM if consistently elevated.

Disk rebuild

Process of restoring redundancy after a disk or node failure. It consumes cluster resources, so avoid maintenance operations while it runs.

NTP drift

Time difference between cluster components. It breaks authentication, replication, and cluster services, so verify time sync on CVMs and hypervisors.

Node maintenance mode

State that evacuates VMs and data from a node before hardware service, preserving replication factor and avoiding data loss.