0lab

Personal monorepo. Each subdirectory is an independent project with its own Docker setup and CI/CD pipeline. A single self-hosted GitHub Actions runner on nuc (this machine) handles all builds and deploys.

Repository structure

0lab/
├── .github/
│   └── workflows/
│       ├── deploy-ckt.yml       # WordPress staging deploy
│       └── build-sitemap.yml    # Sitemap crawler Docker build
├── ckt/                         # WordPress site (chekimthoa → ckt.d4t0.com)
└── sitemap/                     # Sitemap crawler CLI tool

Infrastructure (on this machine)

Service	How it runs	Address
Self-hosted runner	systemd actions.runner.dttad-0lab.nuc-runner	—
Zot container registry	Docker container zotregistry	localhost:5000 / cr.d4t0.com
CKT WordPress (staging)	Docker Compose in ckt/	localhost:8088 / https://ckt.d4t0.com
Cloudflare tunnel (ckt)	systemd cloudflared-ckt	routes ckt.d4t0.com → localhost:8088

Self-hosted runner

Installed at /home/dat/actions-runner, runs as user dat.

# Status
systemctl status actions.runner.dttad-0lab.nuc-runner

# Logs
journalctl -u actions.runner.dttad-0lab.nuc-runner -f

# Restart
sudo systemctl restart actions.runner.dttad-0lab.nuc-runner

Zot registry

Auth: user dat, password stored in /opt/zotregistry/config/htpasswd.

# List images
curl -s -u dat:PASSWORD http://localhost:5000/v2/_catalog | python3 -m json.tool

# List tags for an image
curl -s -u dat:PASSWORD http://localhost:5000/v2/sitemap-crawler/tags/list | python3 -m json.tool

# Pull an image
docker pull localhost:5000/sitemap-crawler:staging

GitHub Actions secrets required: ZOT_USER, ZOT_PASSWORD.

---

Projects

ckt/ — WordPress (Chè Kim Thoa staging)

Migrated from chekimthoa.com backup. Runs as a Docker Compose stack on this machine, exposed via Cloudflare tunnel.

Stack: PHP 8.3 + Apache (official wordpress image) + MySQL 8.0

Theme: wp-content/themes/chekimthoa-theme/ ← edit here
Plugins: wp-content/plugins/ (WooCommerce, Flatsome, CF7, RankMath, etc.)
Uploads: Docker volume ckt_uploads (not in git, persists across deploys)
Database: Docker volume ckt_db_data

Config file: ckt/.env (not committed — create from the table below on a new machine)

Variable	Description
DB_NAME	Database name
DB_USER	DB user
DB_PASSWORD	DB password
DB_ROOT_PASSWORD	MySQL root password
WP_ENV	staging or production
WP_HOME	Full URL e.g. https://ckt.d4t0.com
AUTH_KEY … NONCE_SALT	WordPress salts (generate at https://roots.io/salts.html)

WordPress admin: user quanly, login at https://ckt.d4t0.com/wp-admin

Daily commands

cd /home/dat/Sources/0lab/ckt

# Start / stop
docker compose up -d
docker compose down

# View logs
docker compose logs -f wordpress

# Run WP-CLI
docker compose exec wordpress php wp-cli.phar <command> --allow-root

# Import a new DB backup
docker compose exec -T db mysql -u wordpress -pPASSWORD ckt_wordpress < backup.sql

# Search-replace domain after DB import
docker compose exec -T wordpress php wp-cli.phar \
  search-replace 'https://old-domain.com' 'https://ckt.d4t0.com' --allow-root

Deploy flow

Edit theme in VS Code
  → git add / commit
  → git push origin staging
  → GitHub Actions (deploy-ckt.yml) triggers
  → Runner: git pull + docker compose up -d --build
  → Live at https://ckt.d4t0.com

Theme and plugin changes are bind-mounted — they reflect immediately after git pull without a Docker rebuild. Only changes to Dockerfile or wp-config.php require a rebuild.

Restore on a new machine

git clone git@github.com:dttad/0lab.git
cd 0lab/ckt

# Create .env (fill in values)
cp /path/to/backup/.env .env

# Start stack
docker compose up -d --build

# Import database
docker compose exec -T db mysql -u wordpress -pPASSWORD ckt_wordpress < backup.sql

# Copy uploads into volume
docker run --rm \
  -v ckt_uploads:/uploads \
  -v /path/to/backup/uploads:/src:ro \
  alpine sh -c "cp -r /src/. /uploads/"

# Search-replace domain if needed
docker compose exec -T wordpress php wp-cli.phar \
  search-replace 'https://old-domain.com' 'https://ckt.d4t0.com' --allow-root

---

sitemap/ — Sitemap crawler

CLI tool that fetches and parses XML sitemaps from websites, saving results as gzip-compressed JSON files. Useful for bulk URL extraction, SEO audits, and web indexing.

Language: Python 3.12
Base image: python:3.12-slim
Registry image: localhost:5000/sitemap-crawler

Environment variables

Variable	Default	Description
URLS	required	Comma-separated list of site URLs to crawl
THRESHOLD	50000	URLs per output file before splitting
MOUNTED_OUTPUT_PATH	/opt/output	Directory to write output files

Output format

Files are named {domain}_{batch}_{count}.json.gz and contain:

[
  {"url": "https://example.com/page", "last_modified": "2024-01-01T00:00:00"},
  ...
]

Run from registry

docker run --rm \
  -e URLS="https://example.com/" \
  -e THRESHOLD=50000 \
  -v $PWD/output:/opt/output \
  localhost:5000/sitemap-crawler:staging

Run multiple URLs in parallel (prorun.sh)

cd sitemap
bash prorun.sh -u "https://site1.com,https://site2.com" -o "/opt/output" -t 200000

Each URL gets its own container, all run in parallel.

Build flow

Edit sitemap/
  → git push origin staging (or main)
  → GitHub Actions (build-sitemap.yml) triggers
  → Runner: docker build → push to localhost:5000
  → Tags: :{git-sha} + :staging (or :latest on main)

GitHub Actions secrets required

Secret	Value
ZOT_USER	dat
ZOT_PASSWORD	registry password

---

Adding a new project

1. Create a subdirectory: mkdir 0lab/myproject
2. Add a Dockerfile and whatever the project needs
3. Add a workflow at .github/workflows/build-myproject.yml with paths: ['myproject/**']
4. Push to staging or main — the self-hosted runner picks it up automatically

Branch strategy

Branch	Behaviour
staging	Deploys CKT to https://ckt.d4t0.com; builds sitemap image tagged :staging
main	(future) Deploy to production; builds sitemap image tagged :latest