Fix: LiteLLM on app1 (not Core), Ollama on both Core+app1

README.md

@@ -13,7 +13,7 @@
 - **Provider:** netcup RS 2000 G12
 - **Specs:** 8 vCPU EPYC 9645, 15 GB DDR5 ECC, 512 GB NVMe
 - **OS:** Debian 13
-- **Role:** Hermes + LiteLLM + Ollama fallback + Portals
+- Role: Hermes + Ollama fallback + Portals
 - **Key Services:**
   - Hermes Agent (Telegram + cron, 22 cron jobs)
   - Caddy reverse proxy (11 domains, auto-TLS)
@@ -28,6 +28,7 @@
   - Komodo (Docker, port 9120)
   - Tailscale, StrongSwan, WireGuard (home CCR tunnel)
   - Redis cache
+  - Ollama (systemd, port 11434) — lightweight fallback LLM
 
 ### App1 Server
 - **Hostname:** app1
@@ -41,8 +42,8 @@
   - Open WebUI (Docker, port 3000) — ai.itpropartner.com
   - n8n + Postgres (Docker, port 5678) — n8n.itpropartner.com
   - Caddy (systemd, 80/443)
+  - LiteLLM (Docker) + Postgres — admin-ai.itpropartner.com
   - 5 MCP servers: Super Search (:8899), Browser (:8901), Filesystem (:8900), Email (:8902), Git (:8903)
-- **Planned:** LiteLLM + Postgres (migrating from old-ai 178.156.167.181)
 
 ### App2 Server
 - **Hostname:** app2