ansible
Create, update, or run Ansible playbooks and inventory in this repo for multi-node changes (ansible/). Use for node bootstrap, package installs, or cluster-wide config updates.
When & Why to Use This Skill
This Claude skill empowers users to automate complex infrastructure tasks through Ansible, enabling seamless multi-node configuration management and orchestration. It focuses on repeatable, idempotent workflows for bootstrapping clusters, managing system packages, and maintaining consistent configurations across diverse environments like k3s, Rancher, and Docker hosts.
Use Cases
- Cluster Bootstrapping: Efficiently set up and configure k3s High Availability (HA) clusters and Rancher management platforms across multiple master and worker nodes.
- Automated Software Deployment: Streamline the installation and lifecycle management of essential system tools, such as Tailscale for secure networking and NFS clients for distributed storage.
- Fleet-Wide Configuration Updates: Execute synchronized configuration changes across specific host groups (e.g., proxy servers or docker hosts) while ensuring system stability and consistency.
- Infrastructure Maintenance & Validation: Perform safe, idempotent system updates and service checks using Ansible playbooks with built-in validation steps to minimize production downtime.
| name | ansible |
|---|---|
| description | Create, update, or run Ansible playbooks and inventory in this repo for multi-node changes (ansible/). Use for node bootstrap, package installs, or cluster-wide config updates. |
Ansible
Overview
Use Ansible for repeatable, idempotent changes across nodes. Keep playbooks minimal, explicit about hosts, and safe to re-run.
When to use
- You need to apply the same change on multiple hosts.
- The change touches OS packages, services, or system config.
- You are bootstrapping or maintaining k3s, Rancher, or Tailscale on nodes.
Inventory and groups
Inventory lives in ansible/inventory/hosts.ini. Common groups:
kube_masters(k3s masters)kube_workers(k3s workers)k3s_cluster(masters + workers)proxy(nuc)docker_hosts(docker-host)
Quick start
Ping all nodes in the cluster:
ansible -i ansible/inventory/hosts.ini k3s_cluster -m ping -u kalmyk
Run a playbook on all nodes in the cluster:
ansible-playbook -i ansible/inventory/hosts.ini ansible/playbooks/install_nfs_client.yml -u kalmyk -b
Limit to a single host:
ansible-playbook -i ansible/inventory/hosts.ini ansible/playbooks/install_tailscale.yml -u kalmyk -b --limit kube-worker-00
Common playbooks in this repo
install_nfs_client.yml- install NFS client tools on nodesinstall_tailscale.yml- install Tailscale packagesstart_enable_tailscale.yml- enable and start tailscaledstart_enable_tailscale_client.yml- start Tailscale client servicesk3s-ha.yml- configure k3s HA clusterk3s-oidc.yml- configure OIDC for k3srancher2.yml- install Rancherwait_for_rancher.yml- wait until Rancher is readyrancher_bootstrap_logs.yml- capture Rancher bootstrap logsstart_rancher2_container.yml- start Rancher container
Safety and idempotency
- Prefer Ansible modules over shell commands.
- Use
--checkand--diffwhen validating a risky change. - Use
--limitto scope changes during testing. - Keep playbooks idempotent so re-runs are safe.
Validation
- Service check:
systemctl status tailscaled - Logs:
journalctl -u tailscaled --no-pager -n 50 - Cluster check:
kubectl get nodes -o wide
Resources
- Reference:
references/ansible-runbook.md - Runner:
scripts/run-playbook.sh - Template:
assets/playbook-template.yml