# Hetzner Server Inventory
**Generated:** July 9, 2026
**Purpose:** Migration planning, consolidation reference

---

## Summary: 10 Hetzner Servers

| Server | IP | Type | Uptime | Disk | RAM | Docker? | Key Service |
|--------|-----|------|--------|------|-----|---------|-------------|
| wphost02 | 5.161.62.38 | CPX21 | 7h | 67%/75G | 3.7G | No | RunCloud, Apex WP |
| unms | 5.161.225.131 | CPX21 | 4d 11h | 67%/75G | 3.7G | **Yes (9)** | UISP/UNMS, UCRM |
| unifi | 178.156.131.57 | CPX21 | 6h 42m | 14%/75G | 3.7G | No | UniFi Controller |
| hudu | 178.156.130.130 | CPX21 | 6h 42m | 23%/75G | 3.7G | **Yes (5)** | Hudu IT doc |
| fleettracker360 | 178.156.149.32 | CPX11 | 6h 46m | 25%/38G | 1.9G | No | Apache, MariaDB |
| docker | 178.156.168.35 | CPX11 | 4d 11h | **73%/38G** | 1.9G | **Yes (8)** | RunCloud, NPM, RustDesk |
| **ai** | 178.156.167.181 | **CPX41** | 4d 10h | **🔴 92%/226G** | **15G** | **Yes (6)** | **LiteLLM, Ollama, OpenWebUI** |
| app1-bu | 5.161.114.8 | CPX11 | 4d 4h | -/38G | 1.9G | No | **Warm standby** (Hermes) |
| app1 (n8n) | 87.99.144.163 | CPX11 | 6h 42m | 23%/38G | 1.9G | **Yes (2)** | n8n, n8n-postgres |
| tony-vps | 87.99.159.142 | CPX21 | 3d 6h | 11%/75G | 3.7G | No | Tony's Hermes |

---

## Server Details

### 1. wphost02 (5.161.62.38) — CPX21
- **Uptime:** 7h 35m (recently rebooted during SSH fix)
- **OS:** Debian 12/RunCloud managed
- **CPU:** 3 cores
- **RAM:** 3.7G total, 1.1G used
- **Disk:** 75G (67% used)
- **Docker:** No
- **Services:** nginx-rc (80/443), MariaDB (3306), Postfix (25), httpd (81), RunCloud agent (34210), node_exporter (9100), systemd-resolve (DNS)
- **Runs:** Apex Predators Track Experience (WordPress/WPForms), RunCloud panel
- **Notes:** Recently reconnected after SSH firewall fix. Mail relay configured.

### 2. unms.forefrontwireless.com (5.161.225.131) — CPX21
- **Uptime:** 4d 11h
- **CPU:** 3 cores
- **RAM:** 3.7G total, 1.8G used
- **Disk:** 75G (67% used)
- **Docker:** **Yes — 9 containers**
- **Containers:** unms-device-ws-1, unms-device-ws-2, unms-api, ucrm, unms-netflow, unms-siridb, unms-nginx, unms-rabbitmq, unms-postgres, unms-fluentd
- **Services:** Docker, MariaDB, containerd, fail2ban
- **Runs:** UISP/UNMS (WISP management), UCRM (customer management)

### 3. unifi (178.156.131.57) — CPX21
- **Uptime:** 6h 42m
- **CPU:** 3 cores
- **RAM:** 3.7G total, 1.2G used
- **Disk:** 75G (14% used)
- **Docker:** No
- **Services:** MariaDB, fail2ban, node_exporter
- **Runs:** UniFi Controller (Ubiquiti network management)
- **Notes:** Fresh install — only 6h uptime, 14% disk. Recently rebuilt?

### 4. hudu.itpropartner.com (178.156.130.130) — CPX21
- **Uptime:** 6h 42m
- **CPU:** 3 cores
- **RAM:** 3.7G total, 2.1G used
- **Disk:** 75G (23% used)
- **Docker:** **Yes — 5 containers**
- **Containers:** hudu2-app-1, hudu2-worker-1, hudu2-db-1, hudu2-redis-1, letsencrypt
- **Services:** Docker, MariaDB, firewalld
- **Runs:** Hudu (IT documentation platform)

### 5. fleettracker360 (178.156.149.32) — CPX11
- **Uptime:** 6h 46m
- **CPU:** 2 cores
- **RAM:** 1.9G total, 1.0G used
- **Disk:** 38G (25% used)
- **Docker:** No
- **Services:** Apache2, MariaDB, firewalld, fail2ban
- **Runs:** Fleet tracking application (BoxPilot Logistics?)

### 6. docker (178.156.168.35) — CPX11
- **Uptime:** 4d 11h
- **CPU:** 2 cores
- **RAM:** 1.9G total, 860M used
- **Disk:** 38G (**73% used** — tight)
- **Docker:** **Yes — 8 containers**
- **Containers:** RunCloud PHP (56/81), RunCloud MariaDB, nginx-proxy-manager, hbbr, hbbs (RustDesk), strongswan, uptime-kuma
- **Services:** Docker, containerd, fail2ban
- **Runs:** RustDesk self-hosted, Uptime Kuma monitoring, Nginx Proxy Manager, RunCloud staging

### 7. ai.itpropartner.com (178.156.167.181) — CPX41 ⭐
- **Uptime:** 4d 10h
- **CPU:** **8 cores**
- **RAM:** **15G** total, 3.9G used
- **Disk:** **226G (92% used — 🔴 CRITICAL)**
- **Docker:** **Yes — 6 containers**
- **Containers:** openwebui, litellm, litellm_postgres, caddy, qdrant, ollama
- **Services:** Docker, LiteLLM service, node_exporter, containerd
- **Runs:** **Key service: LiteLLM proxy for admin-ai.itpropartner.com**
- **Also runs:** Open WebUI, Ollama (local LLM), Qdrant (vector DB)
- **⚠️ 92% disk usage needs attention — 18G free on 226G drive**

### 8. app1-bu.itpropartner.com (5.161.114.8) — CPX11
- **Uptime:** 4d 4h
- **CPU:** 2 cores
- **RAM:** 1.9G total
- **Docker:** No
- **Services:** SSH, Tailscale, node_exporter, systemd-resolve
- **Ports:** 22 (SSH), 9999 (Hermes health?), 9100 (node_exporter), Tailscale
- **Runs:** Hermes **warm standby** — checks Core every 5 min, failover in 2 min. Has WG tunnel to home router.

### 9. app1.itpropartner.com / n8n (87.99.144.163) — CPX11
- **Uptime:** 6h 42m
- **CPU:** 2 cores
- **RAM:** 1.9G total, 853M used
- **Disk:** 38G (23% used)
- **Docker:** **Yes — 2 containers**
- **Containers:** n8n, n8n-postgres
- **Services:** Docker, MariaDB, firewalld, fail2ban
- **Runs:** n8n automation workflows

### 10. tony-vps (87.99.159.142) — CPX21
- **Uptime:** 3d 6h
- **CPU:** 3 cores
- **RAM:** 3.7G total, 1.1G used
- **Disk:** 75G (11% used)
- **Docker:** No
- **Services:** llama-server (Llama 3.2 1B fallback), qemu-guest-agent
- **Cron:** hermes-heartbeat.sh (every minute)
- **Runs:** Tony's standalone Hermes, local LLM fallback
- **Notes:** Minimal services — dedicated to Tony's Hermes agent

---

## Key Findings

### 🟢 Ready for migration to netcup:
- unms (UISP/UNMS) — Docker stack, straightforward
- n8n — Simple Docker stack
- docker box (RunCloud staging, NPM, RustDesk, Uptime Kuma)
- fleettracker360 — Apache+MariaDB, simple LAMP app

### 🟡 Needs care:
- **hudu** — Docker stack but sensitive data migration
- **unifi** — Recently rebuilt, needs verification
- **tony-vps** — Tony's separate Hermes, coordinate with him

### 🔴 Critical attention:
- **ai.itpropartner.com** — **92% disk full** (18G left on 226G). Runs LiteLLM, Ollama, OpenWebUI. **Primary model serving for all Hermes instances.** Cannot reboot without risk.
- **Disk cleanup needed:** 199G of 226G used. Docker images/logs likely the culprit.

### 🔵 Warm standby:
- **app1-bu** — Hermes clone, WG tunnel active, ready for failover