I focus on all of the above — prevention is key.
My weekly VM health check routine usually includes:
Performance metrics: CPU, RAM, and disk I/O trends (vCenter / Grafana dashboards).
Backups: Verify last job status, run a quick test restore with Vinchin Backup & Recovery.
Security patches: Apply hypervisor and guest OS updates, check for outdated packages.
Storage health: Monitor datastore usage, check for latency or early warning signs.
Tip: Automating reports with scripts saves a ton of time — I run a PowerShell job every Monday that emails me a summary of all key metrics and backup statuses.