mirror of
https://github.com/Memo-2023/mana-monorepo.git
synced 2026-05-16 15:59:40 +02:00
447dfe276e
- PRODUCTION_LAUNCH.md: Step-by-step guide for deploying to mana.how - URL_SCHEMA.md: Define naming conventions for subdomains - Landing pages: plural (calendars.mana.how) - Web apps: singular (calendar.mana.how) - APIs: singular + -api suffix Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
296 lines
6.2 KiB
Markdown
296 lines
6.2 KiB
Markdown
# Production Launch Guide - mana.how
|
|
|
|
Diese Anleitung beschreibt alle Schritte um die Staging-Umgebung zur Production zu machen.
|
|
|
|
**Server:** 46.224.108.214 (Hetzner)
|
|
**Domain:** mana.how
|
|
|
|
---
|
|
|
|
## Schritt 1: DNS-Einträge anlegen
|
|
|
|
Bei eurem DNS-Provider (wo `mana.how` registriert ist) folgende A-Records anlegen:
|
|
|
|
### Erforderliche DNS-Einträge
|
|
|
|
| Subdomain | Typ | Ziel | TTL |
|
|
|-----------|-----|------|-----|
|
|
| `@` (root) | A | 46.224.108.214 | 300 |
|
|
| `www` | A | 46.224.108.214 | 300 |
|
|
| `auth` | A | 46.224.108.214 | 300 |
|
|
| `chat` | A | 46.224.108.214 | 300 |
|
|
| `chat-api` | A | 46.224.108.214 | 300 |
|
|
| `todo` | A | 46.224.108.214 | 300 |
|
|
| `todo-api` | A | 46.224.108.214 | 300 |
|
|
| `calendar` | A | 46.224.108.214 | 300 |
|
|
| `calendar-api` | A | 46.224.108.214 | 300 |
|
|
| `clock` | A | 46.224.108.214 | 300 |
|
|
| `clock-api` | A | 46.224.108.214 | 300 |
|
|
|
|
**Alternative mit Wildcard:**
|
|
| Subdomain | Typ | Ziel | TTL |
|
|
|-----------|-----|------|-----|
|
|
| `@` (root) | A | 46.224.108.214 | 300 |
|
|
| `*` | A | 46.224.108.214 | 300 |
|
|
|
|
> **Hinweis:** Nach dem Anlegen kann es bis zu 24h dauern bis die DNS-Einträge weltweit propagiert sind. In der Praxis meist schneller.
|
|
|
|
### DNS prüfen
|
|
|
|
```bash
|
|
# Prüfen ob DNS korrekt ist
|
|
dig mana.how +short
|
|
dig auth.mana.how +short
|
|
dig chat.mana.how +short
|
|
# Sollte jeweils 46.224.108.214 zurückgeben
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 2: Server vorbereiten
|
|
|
|
SSH auf den Server:
|
|
|
|
```bash
|
|
ssh -i ~/.ssh/hetzner_deploy_key deploy@46.224.108.214
|
|
```
|
|
|
|
### 2.1 Backup der aktuellen Staging-Daten (optional aber empfohlen)
|
|
|
|
```bash
|
|
cd ~/manacore-staging
|
|
|
|
# Datenbank-Backup erstellen
|
|
docker compose exec -T postgres pg_dumpall -U postgres > ~/backup_$(date +%Y%m%d_%H%M%S).sql
|
|
|
|
echo "Backup erstellt: ~/backup_*.sql"
|
|
```
|
|
|
|
### 2.2 Staging Container stoppen
|
|
|
|
```bash
|
|
cd ~/manacore-staging
|
|
docker compose down
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 3: Production Konfiguration deployen
|
|
|
|
### 3.1 Verzeichnis umbenennen (optional)
|
|
|
|
```bash
|
|
# Von staging zu production umbenennen
|
|
mv ~/manacore-staging ~/manacore-production
|
|
cd ~/manacore-production
|
|
```
|
|
|
|
### 3.2 Production docker-compose kopieren
|
|
|
|
Vom lokalen Rechner:
|
|
|
|
```bash
|
|
# Aus dem Repo-Root
|
|
scp -i ~/.ssh/hetzner_deploy_key \
|
|
docker-compose.production.yml \
|
|
deploy@46.224.108.214:~/manacore-production/docker-compose.yml
|
|
```
|
|
|
|
### 3.3 Production Caddyfile kopieren
|
|
|
|
```bash
|
|
scp -i ~/.ssh/hetzner_deploy_key \
|
|
docker/caddy/Caddyfile.production \
|
|
deploy@46.224.108.214:~/Caddyfile
|
|
```
|
|
|
|
### 3.4 Caddy neu laden
|
|
|
|
Auf dem Server:
|
|
|
|
```bash
|
|
# Caddy Config neu laden
|
|
docker exec caddy caddy reload --config /etc/caddy/Caddyfile
|
|
|
|
# Prüfen ob Caddy läuft
|
|
docker logs caddy --tail 20
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 4: Environment Variables anpassen
|
|
|
|
Auf dem Server die `.env` Datei anpassen:
|
|
|
|
```bash
|
|
cd ~/manacore-production
|
|
nano .env
|
|
```
|
|
|
|
Die bestehenden Staging-Werte können bleiben. Nur sicherstellen dass:
|
|
|
|
```env
|
|
NODE_ENV=production
|
|
|
|
# Diese Werte bleiben gleich (Staging Secrets weiterverwenden):
|
|
POSTGRES_PASSWORD=<behalten>
|
|
REDIS_PASSWORD=<behalten>
|
|
JWT_SECRET=<behalten>
|
|
JWT_PUBLIC_KEY=<behalten>
|
|
JWT_PRIVATE_KEY=<behalten>
|
|
SUPABASE_URL=<behalten>
|
|
SUPABASE_ANON_KEY=<behalten>
|
|
SUPABASE_SERVICE_ROLE_KEY=<behalten>
|
|
AZURE_OPENAI_ENDPOINT=<behalten>
|
|
AZURE_OPENAI_API_KEY=<behalten>
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 5: Container starten
|
|
|
|
```bash
|
|
cd ~/manacore-production
|
|
|
|
# Images pullen
|
|
docker compose pull
|
|
|
|
# Container starten
|
|
docker compose up -d
|
|
|
|
# Status prüfen
|
|
docker compose ps
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 6: Health Checks
|
|
|
|
```bash
|
|
# Alle Services prüfen
|
|
curl -s http://localhost:3001/api/v1/health # Auth
|
|
curl -s http://localhost:5173/health # Dashboard
|
|
curl -s http://localhost:3000/health # Chat Web
|
|
curl -s http://localhost:3002/api/v1/health # Chat API
|
|
curl -s http://localhost:5188/health # Todo Web
|
|
curl -s http://localhost:3018/api/health # Todo API
|
|
curl -s http://localhost:5186/health # Calendar Web
|
|
curl -s http://localhost:3016/api/v1/health # Calendar API
|
|
curl -s http://localhost:5187/health # Clock Web
|
|
curl -s http://localhost:3017/api/v1/health # Clock API
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 7: SSL-Zertifikate (automatisch)
|
|
|
|
Caddy holt sich automatisch Let's Encrypt Zertifikate sobald die DNS-Einträge korrekt sind.
|
|
|
|
Prüfen:
|
|
|
|
```bash
|
|
# Logs prüfen auf Certificate-Meldungen
|
|
docker logs caddy 2>&1 | grep -i "certificate\|tls"
|
|
|
|
# Oder direkt testen
|
|
curl -I https://mana.how
|
|
```
|
|
|
|
---
|
|
|
|
## Schritt 8: Finale Tests
|
|
|
|
Im Browser testen:
|
|
|
|
| URL | Erwartet |
|
|
|-----|----------|
|
|
| https://mana.how | Dashboard Login |
|
|
| https://auth.mana.how/api/v1/health | `{"status":"ok"}` |
|
|
| https://chat.mana.how | Chat App Login |
|
|
| https://todo.mana.how | Todo App Login |
|
|
| https://calendar.mana.how | Calendar App Login |
|
|
| https://clock.mana.how | Clock App Login |
|
|
|
|
---
|
|
|
|
## Troubleshooting
|
|
|
|
### Container startet nicht
|
|
|
|
```bash
|
|
# Logs anschauen
|
|
docker compose logs <service-name>
|
|
|
|
# Beispiel
|
|
docker compose logs mana-core-auth
|
|
docker compose logs chat-backend
|
|
```
|
|
|
|
### DNS nicht propagiert
|
|
|
|
```bash
|
|
# Verschiedene DNS-Server testen
|
|
dig @8.8.8.8 mana.how +short # Google DNS
|
|
dig @1.1.1.1 mana.how +short # Cloudflare DNS
|
|
```
|
|
|
|
### SSL-Zertifikat Fehler
|
|
|
|
```bash
|
|
# Caddy Logs prüfen
|
|
docker logs caddy --tail 100
|
|
|
|
# Caddy neu starten
|
|
docker restart caddy
|
|
```
|
|
|
|
### Datenbank Verbindungsfehler
|
|
|
|
```bash
|
|
# Postgres prüfen
|
|
docker compose exec postgres psql -U postgres -l
|
|
|
|
# Datenbanken anzeigen
|
|
docker compose exec postgres psql -U postgres -c "\l"
|
|
```
|
|
|
|
---
|
|
|
|
## Rollback zu Staging
|
|
|
|
Falls etwas schief geht:
|
|
|
|
```bash
|
|
cd ~/manacore-production
|
|
docker compose down
|
|
|
|
# Alte Staging docker-compose wiederherstellen
|
|
# (müsste vorher gesichert werden)
|
|
|
|
# Caddyfile zurück auf staging
|
|
scp -i ~/.ssh/hetzner_deploy_key \
|
|
docker/caddy/Caddyfile.staging \
|
|
deploy@46.224.108.214:~/Caddyfile
|
|
|
|
docker exec caddy caddy reload --config /etc/caddy/Caddyfile
|
|
docker compose up -d
|
|
```
|
|
|
|
---
|
|
|
|
## Zusammenfassung der URLs
|
|
|
|
Nach erfolgreichem Launch:
|
|
|
|
| App | URL |
|
|
|-----|-----|
|
|
| **Dashboard** | https://mana.how |
|
|
| **Auth API** | https://auth.mana.how |
|
|
| **Chat** | https://chat.mana.how |
|
|
| **Chat API** | https://chat-api.mana.how |
|
|
| **Todo** | https://todo.mana.how |
|
|
| **Todo API** | https://todo-api.mana.how |
|
|
| **Calendar** | https://calendar.mana.how |
|
|
| **Calendar API** | https://calendar-api.mana.how |
|
|
| **Clock** | https://clock.mana.how |
|
|
| **Clock API** | https://clock-api.mana.how |
|