blumeops/ansible/roles/minikube/tasks/main.yml
Erich Blume d76d675b29 Fix minikube role skipping start when kubelet/apiserver are stopped (#137)
## Summary
- After a power loss, minikube's Docker container (host) restarts but kubelet/apiserver remain stopped
- The ansible role's status check used `--format='{{.Host}}'` which only examined the host VM state
- When host=Running but kubelet/apiserver=Stopped, the role skipped `minikube start`
- Fixed to use full `minikube status` exit code (returns non-zero when any component is unhealthy)
- Simplified all downstream conditions to use exit code instead of string matching

## Test plan
- [x] Verified the fix correctly skips `minikube start` when cluster is already fully running
- [x] Pre-commit hooks pass (ansible-lint, yamllint, etc.)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Reviewed-on: https://forge.ops.eblu.me/eblume/blumeops/pulls/137
2026-02-09 23:03:01 -08:00

222 lines
8.1 KiB
YAML

---
# Minikube installation and cluster setup for indri
# Uses docker driver - requires Docker Desktop to be installed manually first
# (Docker Desktop requires GUI setup, so it's not automated in this role)
#
# Prerequisites:
# 1. Install Docker Desktop: brew install --cask docker
# 2. Launch Docker Desktop and complete setup wizard
# 3. Configure Docker Desktop with at least 12GB memory
#
# NOTE: minikube start may have issues when run via SSH.
# If cluster fails to start, manually run on indri:
# minikube start --driver=docker --container-runtime=docker \
# --cpus=6 --memory=11264 --disk-size=200g \
# --apiserver-names=k8s.tail8d86e.ts.net --apiserver-names=indri \
# --apiserver-port=6443 --listen-address=0.0.0.0
- name: Install minikube via homebrew
community.general.homebrew:
name: minikube
state: present
- name: Install kubectl via homebrew
community.general.homebrew:
name: kubectl
state: present
- name: Check if Docker is running
ansible.builtin.command:
cmd: docker info
register: minikube_docker_status
changed_when: false
failed_when: false
- name: Warn if Docker is not running
ansible.builtin.debug:
msg: "WARNING: Docker does not appear to be running. Please start Docker Desktop manually."
when: minikube_docker_status.rc != 0
- name: Check minikube cluster status
ansible.builtin.command:
cmd: minikube status
register: minikube_status
changed_when: false
failed_when: false
- name: Start minikube cluster
ansible.builtin.command:
cmd: >
minikube start
--driver={{ minikube_driver }}
--container-runtime={{ minikube_container_runtime }}
--cpus={{ minikube_cpus }}
--memory={{ minikube_memory }}
--disk-size={{ minikube_disk_size }}
{% for name in minikube_apiserver_names %}
--apiserver-names={{ name }}
{% endfor %}
--apiserver-port={{ minikube_apiserver_port }}
--listen-address={{ minikube_listen_address }}
register: minikube_start
changed_when: minikube_start.rc == 0
failed_when: false # Don't fail - may need manual intervention
when:
- minikube_docker_status.rc == 0
- minikube_status.rc != 0
- name: Check minikube status after start attempt
ansible.builtin.command:
cmd: minikube status
register: minikube_final_status
changed_when: false
failed_when: false
- name: Warn if minikube failed to start
ansible.builtin.debug:
msg: "WARNING: minikube may not have started properly. Run 'minikube start' manually on indri if needed. Status: {{ minikube_final_status.stdout | default('unknown') }}"
when: minikube_final_status.rc != 0
# Configure containerd to use zot registry as pull-through cache
# With docker driver, use host.minikube.internal to reach the host
# Zot runs on indri:5050 and caches images from docker.io, ghcr.io, quay.io
- name: Create containerd registry mirror directories
ansible.builtin.command:
cmd: minikube ssh --native-ssh=false "sudo mkdir -p /etc/containerd/certs.d/{{ item }}"
loop:
- registry.ops.eblu.me
- docker.io
- ghcr.io
- quay.io
changed_when: false
when: minikube_final_status.rc == 0
# Private registry (registry.ops.eblu.me) - direct to zot
- name: Check registry.ops.eblu.me config
ansible.builtin.command:
cmd: minikube ssh --native-ssh=false "cat /etc/containerd/certs.d/registry.ops.eblu.me/hosts.toml 2>/dev/null || echo ''"
register: minikube_registry_config
changed_when: false
when: minikube_final_status.rc == 0
- name: Configure registry.ops.eblu.me mirror
ansible.builtin.command:
cmd: |
minikube ssh --native-ssh=false 'echo "server = \"http://host.minikube.internal:5050\"
[host.\"http://host.minikube.internal:5050\"]
capabilities = [\"pull\", \"resolve\", \"push\"]
skip_verify = true" | sudo tee /etc/containerd/certs.d/registry.ops.eblu.me/hosts.toml'
changed_when: true
when:
- minikube_final_status.rc == 0
- "'host.minikube.internal:5050' not in minikube_registry_config.stdout"
notify: Restart containerd in minikube
# Docker Hub (docker.io) - zot pull-through cache
- name: Check docker.io config
ansible.builtin.command:
cmd: minikube ssh --native-ssh=false "cat /etc/containerd/certs.d/docker.io/hosts.toml 2>/dev/null || echo ''"
register: minikube_dockerio_config
changed_when: false
when: minikube_final_status.rc == 0
- name: Configure docker.io mirror through zot
ansible.builtin.command:
cmd: |
minikube ssh --native-ssh=false 'echo "server = \"https://registry-1.docker.io\"
[host.\"http://host.minikube.internal:5050\"]
capabilities = [\"pull\", \"resolve\"]
skip_verify = true" | sudo tee /etc/containerd/certs.d/docker.io/hosts.toml'
changed_when: true
when:
- minikube_final_status.rc == 0
- "'host.minikube.internal:5050' not in minikube_dockerio_config.stdout"
notify: Restart containerd in minikube
# GitHub Container Registry (ghcr.io) - zot pull-through cache
- name: Check ghcr.io config
ansible.builtin.command:
cmd: minikube ssh --native-ssh=false "cat /etc/containerd/certs.d/ghcr.io/hosts.toml 2>/dev/null || echo ''"
register: minikube_ghcr_config
changed_when: false
when: minikube_final_status.rc == 0
- name: Configure ghcr.io mirror through zot
ansible.builtin.command:
cmd: |
minikube ssh --native-ssh=false 'echo "server = \"https://ghcr.io\"
[host.\"http://host.minikube.internal:5050\"]
capabilities = [\"pull\", \"resolve\"]
skip_verify = true" | sudo tee /etc/containerd/certs.d/ghcr.io/hosts.toml'
changed_when: true
when:
- minikube_final_status.rc == 0
- "'host.minikube.internal:5050' not in minikube_ghcr_config.stdout"
notify: Restart containerd in minikube
# Quay.io - zot pull-through cache
- name: Check quay.io config
ansible.builtin.command:
cmd: minikube ssh --native-ssh=false "cat /etc/containerd/certs.d/quay.io/hosts.toml 2>/dev/null || echo ''"
register: minikube_quay_config
changed_when: false
when: minikube_final_status.rc == 0
- name: Configure quay.io mirror through zot
ansible.builtin.command:
cmd: |
minikube ssh --native-ssh=false 'echo "server = \"https://quay.io\"
[host.\"http://host.minikube.internal:5050\"]
capabilities = [\"pull\", \"resolve\"]
skip_verify = true" | sudo tee /etc/containerd/certs.d/quay.io/hosts.toml'
changed_when: true
when:
- minikube_final_status.rc == 0
- "'host.minikube.internal:5050' not in minikube_quay_config.stdout"
notify: Restart containerd in minikube
# Configure Tailscale serve for k8s API access
# With docker driver, the API server port is dynamic (not fixed at 6443)
# We query the current port and configure tailscale serve accordingly
- name: Get minikube API server URL
ansible.builtin.command:
cmd: kubectl config view --minify -o jsonpath="{.clusters[0].cluster.server}"
register: minikube_api_url
changed_when: false
when: minikube_final_status.rc == 0
- name: Extract API server port from URL
ansible.builtin.set_fact:
minikube_api_port: "{{ minikube_api_url.stdout | regex_search(':([0-9]+)$', '\\1') | first }}"
when:
- minikube_final_status.rc == 0
- minikube_api_url.stdout is defined
- name: Check current tailscale serve config for k8s
ansible.builtin.command:
cmd: tailscale serve status --json
register: minikube_tailscale_serve_status
changed_when: false
when: minikube_api_port is defined
- name: Parse tailscale serve k8s config
ansible.builtin.set_fact:
minikube_tailscale_k8s_tcp: "{{ ((minikube_tailscale_serve_status.stdout | from_json).Services['svc:k8s'].TCP['443'].TCPForward | default('')) }}"
when:
- minikube_api_port is defined
- minikube_tailscale_serve_status.stdout is defined
- "'svc:k8s' in (minikube_tailscale_serve_status.stdout | from_json).Services | default({})"
failed_when: false
- name: Configure tailscale serve for k8s API
ansible.builtin.command:
cmd: tailscale serve --service="svc:k8s" --tcp=443 tcp://localhost:{{ minikube_api_port }}
when:
- minikube_api_port is defined
- minikube_tailscale_k8s_tcp is not defined or minikube_tailscale_k8s_tcp != 'localhost:' + minikube_api_port
changed_when: true