clustered-fucks/docs/control-server-guide.md

339 lines
No EOL
7.9 KiB
Markdown

# Control Server Operations Guide
**Host:** control (CT 127)
**IP:** 192.168.1.127
**Location:** pve2
**User:** maddox
**Last Updated:** January 24, 2026
---
## Quick Start
```bash
# Launch interactive menu
~/scripts/control-menu.sh
# Or jump straight to multi-host SSH
~/scripts/ssh-manager.sh
# Or run Ansible directly
cd ~/clustered-fucks
ansible all -m ping
```
---
## Directory Structure
```
/home/maddox/
├── .ssh/
│ ├── config # SSH host definitions
│ ├── id_ed25519 # SSH private key
│ └── id_ed25519.pub # Public key (add to new hosts)
├── clustered-fucks/ # Git repo (synced to Forgejo)
│ ├── ansible.cfg # Ansible configuration
│ ├── inventory/
│ │ ├── hosts.yml # Host inventory
│ │ └── group_vars/
│ │ └── all.yml # Global variables
│ ├── playbooks/
│ │ ├── check-status.yml
│ │ ├── docker-prune.yml
│ │ ├── restart-utils.yml
│ │ ├── update-all.yml
│ │ ├── deploy-utils.yml
│ │ └── deploy-mealie.yml # NEW
│ └── compose-files/ # Docker compose files for services
│ └── databases/
│ └── mealie/
│ ├── docker-compose.yml
│ └── .env # NOT in git (secrets)
└── scripts/
├── ssh-manager.sh # tmux multi-host launcher
├── control-menu.sh # Interactive Ansible menu
└── add-host.sh # New host onboarding
```
---
## Managed Hosts
| Host | IP | User | Type | Ansible Group |
|------|-----|------|------|---------------|
| pve2 | .3 | root | Proxmox | proxmox_nodes |
| pve-dell | .4 | root | Proxmox | proxmox_nodes |
| replicant | .80 | maddox | VM | docker_hosts |
| databases | .81 | root | VM | docker_hosts |
| immich | .82 | root | VM | docker_hosts |
| media-transcode | .120 | root | LXC | docker_hosts |
| network-services | .121 | root | LXC | docker_hosts |
| download-stack | .122 | root | LXC | docker_hosts |
| docker666 | .123 | root | LXC | docker_hosts |
| tailscale-home | .124 | root | LXC | docker_hosts |
| dns-lxc | .125 | root | LXC | infrastructure |
| nas | .251 | maddox | NAS | legacy |
| alien | .252 | maddox | Docker | legacy |
---
## Ansible Playbooks
| Playbook | Target | Description |
|----------|--------|-------------|
| `check-status.yml` | all_managed | Disk, memory, container counts |
| `update-all.yml` | docker_hosts | apt upgrade all hosts |
| `docker-prune.yml` | docker_hosts | Clean unused Docker resources |
| `restart-utils.yml` | docker_hosts | Restart utils stack |
| `deploy-utils.yml` | docker_hosts | Deploy utils to new host |
| `deploy-mealie.yml` | databases | Deploy Mealie stack |
### Running Playbooks
```bash
cd ~/clustered-fucks
# Check all hosts
ansible-playbook playbooks/check-status.yml
# Update specific host
ansible-playbook playbooks/update-all.yml --limit databases
# Deploy service
ansible-playbook playbooks/deploy-mealie.yml
```
---
## Git Workflow
### Repository Info
- **Local:** `~/clustered-fucks/`
- **Remote:** `ssh://git@192.168.1.81:2222/maddox/clustered-fucks.git`
- **Web:** https://git.3ddbrewery.com/maddox/clustered-fucks
### Daily Workflow
```bash
cd ~/clustered-fucks
# 1. Before making changes, get latest
git pull
# 2. Make your changes (edit files, create playbooks, etc.)
vim playbooks/new-playbook.yml
# 3. See what changed
git status
# 4. Stage all changes
git add -A
# 5. Commit with a message
git commit -m "Add deployment playbook for service X"
# 6. Push to Forgejo
git push
```
### After Creating/Editing Files
**Always commit your changes!** Otherwise they only exist on the control server and could be lost.
```bash
cd ~/clustered-fucks
git add -A
git commit -m "Description of what you changed"
git push
```
### Quick Reference
| Command | What it does |
|---------|--------------|
| `git status` | Show what files have changed |
| `git pull` | Get latest changes from Forgejo |
| `git add -A` | Stage all changes for commit |
| `git commit -m "msg"` | Save changes with a message |
| `git push` | Upload commits to Forgejo |
| `git log --oneline -5` | Show last 5 commits |
| `git diff` | Show what changed (before staging) |
### What Goes in Git vs What Doesn't
**✅ Commit these:**
- Ansible playbooks (`playbooks/*.yml`)
- Inventory files (`inventory/hosts.yml`)
- Docker compose files (`compose-files/**/docker-compose.yml`)
- Documentation and scripts
**❌ Never commit these:**
- `.env` files (contain passwords/secrets)
- Private keys
- Temporary files
The `.gitignore` file already excludes `.env` files.
---
## Service Migration Workflow
### Standard Process
```bash
cd ~/clustered-fucks
# 1. Create compose file directory
mkdir -p compose-files/<target_host>/<service>
# 2. Create docker-compose.yml
vim compose-files/<target_host>/<service>/docker-compose.yml
# 3. Create .env for secrets (not committed to git)
vim compose-files/<target_host>/<service>/.env
# 4. Create Ansible playbook
vim playbooks/deploy-<service>.yml
# 5. Deploy via Ansible
ansible-playbook playbooks/deploy-<service>.yml
# 6. Rsync data from alien (if needed)
ssh alien "docker stop <container>"
rsync -avP maddox@alien:/path/to/data/ root@<target>:/home/docker/appdata/<service>/
# 7. Start the service
ansible <target> -m shell -a "cd /home/docker/appdata/<service> && docker compose up -d"
# 8. Update Traefik route (on Hetzner)
# 9. Test via domain
# 10. Stop old container on alien
ssh alien "docker stop <container>"
# 11. Commit playbook to git
git add -A
git commit -m "Add <service> deployment"
git push
```
### Playbook Template
```yaml
---
- name: Deploy <Service> to <host>
hosts: <target_host>
become: yes
vars:
service_name: <service>
service_dir: /home/docker/appdata/{{ service_name }}
tasks:
- name: Create directories
file:
path: "{{ item }}"
state: directory
mode: '0755'
loop:
- "{{ service_dir }}"
- name: Ensure proxy network exists
community.docker.docker_network:
name: proxy
state: present
- name: Copy docker-compose.yml
copy:
src: ../compose-files/<host>/{{ service_name }}/docker-compose.yml
dest: "{{ service_dir }}/docker-compose.yml"
mode: '0644'
- name: Copy .env file
copy:
src: ../compose-files/<host>/{{ service_name }}/.env
dest: "{{ service_dir }}/.env"
mode: '0600'
- name: Start stack
community.docker.docker_compose_v2:
project_src: "{{ service_dir }}"
state: present
recreate: always
```
---
## Common Operations
### SSH to Specific Host
```bash
ssh databases
ssh alien
ssh nas # Uses port 44822 automatically
```
### Run Command on All Docker Hosts
```bash
ansible docker_hosts -m shell -a "docker ps -q | wc -l"
```
### Check Container Logs
```bash
ansible databases -m shell -a "docker logs mealie 2>&1 | tail -30"
```
### Copy File to Host
```bash
ansible databases -m copy -a "src=./file.txt dest=/tmp/file.txt"
```
---
## Troubleshooting
### SSH Connection Refused
```bash
# Access via Proxmox console:
# For LXC: pct enter <CT_ID>
# For VM: qm terminal <VM_ID>
# Then inside:
apt install openssh-server
systemctl enable ssh
systemctl start ssh
```
### SSH Permission Denied
```bash
# Copy your key to the host
ssh-copy-id <hostname>
# Or fix permissions on target:
chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys
```
### Git Push Fails
```bash
# Check remote is correct
git remote -v
# Should show:
# origin ssh://git@192.168.1.81:2222/maddox/clustered-fucks.git
# If wrong, fix it:
git remote set-url origin ssh://git@192.168.1.81:2222/maddox/clustered-fucks.git
```
---
## Changelog
| Date | Change |
|------|--------|
| 2026-01-23 | Initial deployment |
| 2026-01-24 | Added deploy-mealie.yml, enhanced git workflow docs |