## Purpose
After many years of using Material for MKDocs and it being updated with new features and security updates, it finally reached EOL around the end of 2025.  The project maintainers started pivoting to a new successor called [Zensical](https://zensical.org/docs/get-started/).  This document outlines my particular process for setting up a standalone documentation server within a virtual machine.

!!! info "Assumptions"
    It is assumed that you are deploying this server into `Ubuntu Server 24.04.2 LTS (Minimal)`.  It is also assumed that you are running every command as a user with superuser privileges (e.g. `root`).

    You are generally safe to have a GuestVM with 16GB for the virtual disk, and expand it over-time based on your needs.  CPU count and RAM allocation can also be extremely low based on your preferences, since this is simply a static page website at the end of the day.

## Architectural Overview
It is useful to understand the flow of data and how everything inter-connects, so I have provided a sequence diagram that you can follow below:

``` mermaid
sequenceDiagram
    autonumber
    actor Author as Doc Author
    participant Gitea as Gitea (Repo + Actions)
    participant Runner as Act Runner
    participant Zensical as Zensical Server (watch + build)
    participant NGINX as NGINX (serves static site)

    Author->>Gitea: Push to main
    Gitea-->>Runner: Trigger workflow job
    Runner->>Zensical: rsync docs → /srv/zensical/docs
    Zensical-->>Zensical: Watch detects change
    Zensical->>Zensical: Rebuild site → /srv/zensical/site
    NGINX-->>NGINX: Serve files from /srv/zensical/site
```

## Setup Python Environment
The first thing we need to do is install the necessary python packages and install the zensical software stack inside of it.

```sh
sudo apt update && sudo apt upgrade -y
sudo apt install -y nano python3 python3.12-venv
mkdir -p /srv/zensical
cd /srv/zensical
python3 -m venv .venv
source .venv/bin/activate
pip install zensical
zensical new .
deactivate

# Remove Placeholder Example Docs
rm -rf /srv/zensical/docs/{*,.*}
```

## Zensical
### Configure Settings
Now we want to set some sensible defaults for Zensical to style it to look as close to Material for MKDocs as possible.

```sh
sudo tee /srv/zensical/zensical.toml > /dev/null <<'EOF'
[project]
site_name = "Bunny Lab"
site_description = "Server, Script, Workflow, and Networking Documentation"
site_author = "Nicole Rappe"
site_url = "https://kb.bunny-lab.io/"
repo_url = "https://git.bunny-lab.io/bunny-lab/docs"
repo_name = "bunny-lab/docs"
edit_uri = "_edit/main/"

[project.theme]
variant = "classic"
language = "en"
features = [
    "announce.dismiss",
    "content.action.edit",
    "content.code.annotate",
    "content.code.copy",
    "content.code.select",
    "content.footnote.tooltips",
    "content.tabs.link",
    "content.tooltips",
    "navigation.indexes",
    "navigation.instant",
    "navigation.instant.prefetch",
    "navigation.instant.progress",
    "navigation.path",
    "navigation.tabs",
    "navigation.tabs.sticky",
    "navigation.top",
    "navigation.tracking",
    "search.highlight",
]

[[project.theme.palette]]
scheme = "default"
toggle.icon = "lucide/sun"
toggle.name = "Switch to dark mode"

[[project.theme.palette]]
scheme = "slate"
toggle.icon = "lucide/moon"
toggle.name = "Switch to light mode"

EOF
```

### Create Watchdog Service
Since NGINX has taken over hosting the webpages, this does not need to be accessible from other servers, only NGINX itself which runs on the same host as Zensical.  We only want to use the `zensical serve` command to keep a watchdog on the documentation folder and automatically rebuild the static site content when changes are detected.  These changes are then served by NGINX's webserver.

```sh
# Create Service User, Assign Access, and Lockdown Zensical Data
sudo useradd --system --home /srv/zensical --shell /usr/sbin/nologin zensical || true
sudo chown -R zensical:zensical /srv/zensical
sudo find /srv/zensical -type d -exec chmod 2775 {} \;
sudo find /srv/zensical -type f -exec chmod 664 {} \; # This step likes to take a while, sometimes up to a minute.
```

```sh
# Make Zensical Binary Executable for Service
sudo chmod +x /srv/zensical/.venv/bin/zensical

# Add Additional User(s) to Folder for Extra Access (Such as Doc Runners)
sudo usermod -aG zensical nicole

# Create Service
sudo tee /etc/systemd/system/zensical-watchdog.service > /dev/null <<'EOF'
[Unit]
Description=Zensical Document Changes Watchdog (zensical serve)
After=network-online.target
Wants=network-online.target

[Service]
Type=simple
User=zensical
Group=zensical
WorkingDirectory=/srv/zensical

# Run the venv binary directly; no activation needed
ExecStart=/srv/zensical/.venv/bin/zensical serve

Restart=always
RestartSec=2

[Install]
WantedBy=multi-user.target
EOF

# Start & Enable Automatic Startup of Service
sudo systemctl daemon-reload
sudo systemctl enable --now zensical-watchdog.service
```

## NGINX Webserver
We need to deploy NGINX as a webserver, because when using reverse proxies like Traefik, it seems to not get along with Zensical at all.  Attempts to resolve this all failed, so putting the statically-built copies of site data that Zensical generates into NGINX's root directory is the second-best solution I came up with.  Traefik can be reasonably expected to behave when interacting with NGINX versus Zensical's built-in webserver.

```sh
sudo apt install -y nginx
sudo rm -f /etc/nginx/sites-enabled/default
sudo tee /etc/nginx/sites-available/zensical.conf > /dev/null <<'EOF'
server {
    listen 80;
    listen [::]:80;
    server_name _;

    root /srv/zensical/site;
    index index.html;

    # Primary document handling
    location / {
        try_files $uri $uri/ /index.html;
    }

    # Static asset caching (safe for docs)
    location ~* \.(css|js|png|jpg|jpeg|gif|svg|ico|woff2?)$ {
        expires 7d;
        add_header Cache-Control "public, max-age=604800, immutable";
        try_files $uri =404;
    }

    # Prevent access to source or metadata
    location ~* \.(toml|md)$ {
        deny all;
    }
}
EOF

sudo ln -s /etc/nginx/sites-available/zensical.conf /etc/nginx/sites-enabled/zensical.conf
sudo nginx -t
sudo systemctl reload nginx
sudo systemctl enable nginx
```

## Gitea ACT Runner
Now is time for the arguably most-important stage of deployment, which is setting up a [Gitea Act Runner](https://docs.gitea.com/usage/actions/act-runner).  This is how document changes in a Gitea repository will propagate automatically into Zensical's `/srv/zensical/docs` folder.

```sh
# Install Dependencies
sudo apt install -y nodejs npm git rsync curl

# Create dedicated Gitea runner service account
sudo useradd --system --create-home --home /var/lib/gitea_runner --shell /usr/sbin/nologin gitearunner || true

# Allow the runner to write documentation changes
sudo usermod -aG zensical gitearunner

# Download Newest Gitea Runner Binary (https://gitea.com/gitea/act_runner/releases)
cd /tmp
wget https://gitea.com/gitea/act_runner/releases/download/v0.2.13/act_runner-0.2.13-linux-amd64
sudo install -m 0755 act_runner-0.2.13-linux-amd64 /usr/local/bin/gitea_runner
gitea_runner --version

# Generate Gitea Runner Configuration
sudo mkdir -p /etc/gitea_runner
sudo chown gitearunner:gitearunner /etc/gitea_runner
sudo -u gitearunner gitea_runner generate-config > /etc/gitea_runner/config.yaml
```

### Configure Registration Token
- Navigate to: "**<Gitea Repo> > Settings > Actions > Runners**" 
    - If you don't see this, it needs to be enabled. Navigate to: "**<Gitea Repo> > Settings > "Enable Repository Actions: Enabled" > Update Settings**"
- Click the "**Create New Runner**" button on the top-right of the page and copy the registration token somewhere temporarily.
- Navigate back to the GuestVM running Zensical and run the following commands.

```sh
# Start Token Registration Process
sudo -u gitearunner env HOME=/var/lib/gitea_runner /usr/local/bin/gitea_runner register --config /etc/gitea_runner/config.yaml

# Gitea Instance URL: https://git.bunny-lab.io
# Gitea Runner Token: <Gitea-Runner-Token>
# Runner Name: zensical-docs-runner

# Move Runner Config to Correct Location & Configure Permissions
sudo mv /tmp/.runner /var/lib/gitea_runner/.runner
sudo chown gitearunner:gitearunner /var/lib/gitea_runner/.runner
sudo chmod 600 /var/lib/gitea_runner/.runner
```

### Create Service
Now we need to configure the Gitea runner to start automatically via a service just like the Zensical Watchdog service.

```sh
# Create Gitea Runner Service
sudo tee /etc/systemd/system/gitea-runner.service > /dev/null <<'EOF'
[Unit]
Description=Gitea Actions Runner (gitea_runner)
After=network-online.target
Wants=network-online.target

[Service]
Environment=HOME=/var/lib/gitea_runner
User=gitearunner
Group=gitearunner
WorkingDirectory=/var/lib/gitea_runner
ExecStart=/usr/local/bin/gitea_runner daemon --config /etc/gitea_runner/config.yaml
Restart=always
RestartSec=2

[Install]
WantedBy=multi-user.target
EOF

# Remove Container-Based Configurations to Force Runner to Run in Host Mode
sudo sed -i \
  '/^[[:space:]]*labels:/,/^[[:space:]]*cache:/{
    /^[[:space:]]*labels:/c\  labels:\n    - "zensical-host:host"
    /^[[:space:]]*cache:/!d
  }' \
  /etc/gitea_runner/config.yaml

# Enable and Start the Service
sudo systemctl daemon-reload
sudo systemctl enable --now gitea-runner.service
```

### Repository Workflow
Place the following file into your documentation repository at the given location and this will enable the runner to execute when changes happen to the repository data.

```yaml title="gitea/workflows/gitops-automatic-deployment.yml"
name: GitOps Automatic Documentation Deployment

on:
  push:
    branches: [ main ]

jobs:
  zensical_deploy:
    name: Sync Docs to https://kb.bunny-lab.io
    runs-on: zensical-host

    steps:
      - name: Checkout Repository
        uses: actions/checkout@v3

      - name: Sync repository into /srv/zensical/docs
        run: |
          rsync -rlD --delete \
            --exclude='.git/' \
            --exclude='.gitea/' \
            --exclude='assets/' \
            --exclude='schema/' \
            --exclude='stylesheets/' \
            --exclude='schema.json' \
            --chmod=D2775,F664 \
            . /srv/zensical/docs/

      - name: Notify via NTFY
        if: always()
        run: |
          curl -d "https://kb.bunny-lab.io - Zensical job status: ${{ job.status }}" https://ntfy.bunny-lab.io/gitea-runners
```

## Traefik Reverse Proxy
It is assumed that you use a [Traefik](../edge/traefik.md) reverse proxy and are configured to use [dynamic configuration files](../edge/traefik.md#dynamic-configuration-files).  Add the file below to expose the Zensical service to the rest of the world.

```yaml title="kb.bunny-lab.io.yml"
http:
  routers:
    kb:
      entryPoints:
        - websecure
      tls:
        certResolver: letsencrypt
      service: kb
      rule: Host(`kb.bunny-lab.io`)

  services:
    kb:
      loadBalancer:
        servers:
          - url: http://192.168.3.8:80
        passHostHeader: true
```