High Availability Home Assistant

Overview

Premium offering that leverages a Proxmox cluster (2+ nodes) to keep Home Assistant running through host failures or planned maintenance. This is a premium, installer-led option for large or mission-critical homes where short outages are unacceptable.

Customer value

• Installers: Differentiated tier with proactive resilience and service continuity, enabling higher-margin packages.
• Homeowners: Automations, scenes, and critical controls continue working even when primary node is unavailable.

Scope (first iterations)

• Failover approach centered on Proxmox capabilities (live migration and HA restarts), not active-active HA.
• Clear recovery boundaries for HA state (database, automations, integrations) to avoid split-brain scenarios.
• Health checks and alerting integrated with Remote Installation Monitoring.
• Remote access guardrails aligned with Secure Remote Access (e.g., Tailscale JIT) for maintenance operations.

Architecture (directional)

• Proxmox cluster: 2 or 3 nodes with shared or replicated storage suitable for VM mobility.
• Home Assistant VM: Pinned resources, watchdog, graceful shutdown, and boot sequencing for reliable recovery.
• Storage: Options evaluation (ZFS replication, Ceph, NFS) for simplicity vs performance vs cost tradeoffs.
• State protection: Enforce single-writer patterns; no active-active clustering of Home Assistant itself.

Open questions and risks

• Device ecosystems generally expect a single coordinator; evaluate edge cases for Zigbee/Z-Wave/Matter during failover.
• Database and add-on state consistency during abrupt host failure; document guarantees and recovery windows.
• Cost and complexity vs alternatives (fast restore, redundant hardware in a single node) for premium tier ROI.