Skip to main content

Overview

Xloud Instance HA monitors compute hosts and automatically recovers protected instances when a host becomes unreachable. As a project member, you configure which instances are protected and track recovery events through the Dashboard or CLI. This guide covers protection concepts, assigning policies to instances, and monitoring ongoing recovery workflows.
Instance HA protection requires your administrator to have configured at least one failover segment and host monitor. Contact your administrator if protection policies are unavailable in your project. Your administrator can configure this through XDeploy.

In This Guide

How It Works

Understand the failover detection and recovery sequence from host failure to instance restoration.

Protection Segments

Learn what failover segments are and how to view the segments your instances belong to.

Recovery Workflows

Understand the recovery process and the steps involved when an instance is recovered.

Monitoring Status

Track notification events, view recovery history, and verify instance protection status.

Troubleshooting

Diagnose protection assignment issues, failed recoveries, and notification errors.

Key Concepts

Failover Segment

A group of compute hosts monitored together as a unit. When a host in a segment fails, all protected instances on that host are recovered.

Protection Policy

Attached to an instance to define its recovery behavior — whether to recover automatically and how quickly to begin.

Host Monitor

An agent that continuously checks host availability and triggers the recovery engine when a host becomes unreachable.

Recovery Notification

An event emitted by the host monitor that initiates the recovery workflow for all protected instances on the failed host.

Next Steps

Instance HA Admin Guide

Configure failover segments, host monitors, and recovery policies platform-wide.

Instance HA Overview

Service overview, feature summary, and getting started with Instance HA.