Runbooks and Operational Automation
Runbooks and Operational Automation focuses on repeatable operational procedures in Prism Central. A runbook packages the steps an operator would otherwise perform manually, with scope, approvals, and an audit trail.
Core rules
- Runbooks are for operational procedures such as patching checks, snapshot cleanup, or VM lifecycle tasks.
- Scope by category or project so a runbook does not touch unintended VMs.
- Steps can include checks, approvals, scripts, API calls, and notifications.
- Execution can be manual, scheduled, or triggered; scheduled runbooks need overlap and failure handling.
- Every run should leave an audit record of inputs, steps, and outcomes.
Worked example
A monthly snapshot runbook selects VMs with Backup:Daily. A pre-check confirms the VM is powered on and has no existing snapshot from the same day. An approval step asks the storage owner. A snapshot action runs, and a notification reports success or failure. If the pre-check fails, the runbook stops before creating a snapshot. This prevents duplicate snapshots and makes the run auditable.
Traps
- Running a broad scope without a category filter.
- Skipping pre-checks and discovering a conflict mid-run.
- Scheduling overlapping runs that contend for the same objects.
- Treating a failed step as success because later steps continued.
- Omitting approvals for destructive or high-impact steps.