> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/switch-infrastructure/config-manager/llms.txt. > For full documentation content, see https://docs.nvidia.com/switch-infrastructure/config-manager/llms-full.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/switch-infrastructure/config-manager/_mcp/server. # Start Here NVIDIA Config Manager is a Helm-deployed platform for rendering, deploying, and operating NVIDIA network configuration at scale. It combines APIs, workers, Nautobot, data stores, eventing, and device-facing services behind one deployment. Use Config Manager when you need to: * Bootstrap new switch fabrics from modeled intent. * Keep rendered intended configuration tied to Nautobot data. * Review and approve Day-2 configuration changes before they reach devices. * Run operational workflows such as backup, deploy, cable validation, password rotation, OS upgrades, and reprovisioning. ## Primary Personas | Persona | Primary work | | :----------------------------------- | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Network Infrastructure Administrator | Stands up and maintains the platform. This includes the Helm release, Nautobot integration, identity provider integration, secrets, registries, observability, ZTP exposure, and workflow defaults. | | Network Infrastructure Operator | Uses a running platform to drive network outcomes. This includes launching workflows, approving changes, browsing rendered or backed-up configs, provisioning devices through ZTP/DHCP, and watching dashboards or logs. | ## Feature Areas | Feature area | Administrator focus | Operator focus | | :------------------------------------------------------------------------------------------------------------------------------- | :----------------------------------------------------------------------------------------------------------- | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | [Getting Started with Config Manager](/switch-infrastructure/config-manager/getting-started/getting-started-with-config-manager) | Deploy the Helm release on Kind or a target Kubernetes cluster, choose sizing, hostnames, services, and TLS. | Confirm the running UIs and APIs are reachable after install. | | [Airgapped Deployment](/switch-infrastructure/config-manager/deployment/airgapped) | Build, transfer, and install the offline bundle through an internal registry. | Use the installed platform the same way as a connected deployment. | | [Nautobot Integration](/switch-infrastructure/config-manager/config-manager/nautobot) | Configure bundled or external Nautobot, bootstrap jobs, tokens, and source-of-truth data loading. | Treat Nautobot as the source of truth for device, interface, cabling, IPAM, VLAN, and config context data. | | Device authentication | Seed per-site network secrets in Vault/OpenBao through ESO, or Kubernetes secrets for local development. | Run and review [device](/switch-infrastructure/config-manager/user-guides/lifecycle/device-password-rotation) or [site password rotation](/switch-infrastructure/config-manager/user-guides/lifecycle/site-password-rotation) workflows. | | [Render Service](/switch-infrastructure/config-manager/services/render/overview) | Enable template plugin paths, storage, and event plumbing. | Review intended config output and trigger safe render checks through Nautobot changes. | | [Config Store](/switch-infrastructure/config-manager/services/config-store/overview) | Provision storage, API, and UI exposure. | Browse intended and backup configs, inspect diffs, and confirm rendered output exists. | | [Temporal Workflows](/switch-infrastructure/config-manager/services/temporal/overview) | Configure Temporal, worker pools, and workflow RBAC. | Launch, approve, retry, reject, or terminate workflow runs. | | [ZTP Provisioning](/switch-infrastructure/config-manager/services/network-ztp/overview) | Configure OS image storage, boot URLs, DHCP/ZTP exposure, and Nautobot readiness. | Watch device boot and callback progress during bring-up. | | [DHCP Service](/switch-infrastructure/config-manager/services/dhcp/overview) | Enable Kea-backed DHCP and model reservations/options in Nautobot. | Inspect reservations, leases, and generated DHCP config when provisioning stalls. | | [Config Manager UI](/switch-infrastructure/config-manager/getting-started/which-interface-should-i-use) | Set `global.baseHostname`, SSO callbacks, and gateway/TLS access. | Use workflow and config views for day-0 and day-2 operations. | | Observability and troubleshooting | Enable monitoring resources and the local observability stack when appropriate. | Use [Troubleshooting](/switch-infrastructure/config-manager/deployment/troubleshooting), pod logs, and workflow status to investigate failed operations. | ## Recommended first path 1. Administrators should start with [Config Manager Concepts](/switch-infrastructure/config-manager/config-manager/overview), then follow [Getting Started with Config Manager](/switch-infrastructure/config-manager/getting-started/getting-started-with-config-manager) or [Airgapped Deployment](/switch-infrastructure/config-manager/deployment/airgapped). 2. Before pointing Config Manager at real devices, confirm network secrets, DHCP/ZTP exposure, device authentication prerequisites, and maintenance controls. 3. Operators should start with [Which Interface Should I Use?](/switch-infrastructure/config-manager/getting-started/which-interface-should-i-use), then review [Temporal Workflows](/switch-infrastructure/config-manager/services/temporal/overview), [Workflow Lifecycle](/switch-infrastructure/config-manager/user-guides/controlling-running-workflows), and [Config Store](/switch-infrastructure/config-manager/services/config-store/overview). 4. To preview Config Manager locally without a real lab, use [Local Development Quick Start](/switch-infrastructure/config-manager/getting-started/local-development-quick-start). 5. After installation, follow [First Run Tour](/switch-infrastructure/config-manager/getting-started/first-run-tour) to verify the deployment and choose the first workflows to run. The local development quick start is not a deployment path and is not a workflow validation environment. It exists only to help developers preview rendered configuration, inspect UI flows, and understand Nautobot modeling shape without touching real devices.