Platform surface
Advanced operator surfaces: security, cost, tenancy, backup and DR, HA, automation, observability.
Back to the README ยท Docs index
The quickstart above is the everyday path. Underneath it, Zorvia's CLI carries roughly 178 subcommands across operator-grade surfaces most teams grow into over time โ security, cost, multi-tenancy, backup/DR, HA, automation, observability. Per DEVELOPMENT.md's own scope note, treat these as advanced surfaces, not every-cluster guarantees: the always-supported path is create โ day-2 โ snapshots โ drift โ plan โ golden images โ Terraform. Everything below is real and shipped โ not a promise that every module fits every deployment.
Security & compliance
zorvia security security-scan prod-db --scan-type deep
zorvia security security-harden prod-db --profile cis
zorvia security compliance-check prod-db --framework soc2
zorvia security audit-list --security-only
Cost & FinOps
zorvia cost cost-analyze prod-db --period 30d
zorvia cost cost-optimize --high-priority-only
zorvia cost cost-waste --waste-type idle
zorvia cost budget-create platform --amount 5000 --period monthly --alert-threshold 80
Multi-tenancy & access
zorvia tenancy tenants-create platform-team --owner alice --email alice@example.com
zorvia tenancy users-create bob --email bob@example.com --role operator
zorvia tenancy users-assign-role bob db-admin --scope namespace:staging
zorvia tenancy quotas-create platform-quota --namespace platform --preset large
Backup & disaster recovery
zorvia backup backup-create prod-db --backup-type incremental
zorvia backup backup-schedule-create nightly --schedule daily --vm prod-db
zorvia backup backup-restore prod-db-full-20260901 --target prod-db-restore --start
zorvia backup recovery-plan primary-site-failover
High availability
zorvia ha ha-config prod-db --enable --priority critical --eviction-strategy live-migrate
zorvia ha ha-status prod-db
zorvia ha evacuate-node worker-3 --max-parallel 4 --plan
Automation & workflows
zorvia automation automation-create nightly-snapshot --trigger schedule --enable
zorvia automation workflow-create dr-failover --template disaster-recovery
zorvia automation workflow-run dr-failover --watch
zorvia automation schedule-create weekly-backup --rule nightly-snapshot --schedule weekly --enable
Observability & insights
zorvia observability metrics-query cpu_usage --aggregation p95
zorvia observability alerts-active --severity critical
zorvia observability insights-generate prod-db --insight-type performance
zorvia observability trends-analyze cpu_usage --window 24