docs: add Loki/Promtail, LiteLLM, agentmemory, Raspberry Pi; drop Elastic

- Replace Elastic Stack with Loki + Promtail throughout
- Add LiteLLM (LLM proxy), agentmemory, athome (personal homepage) to applications
- Add Raspberry Pi (naruto) to hardware table — Zigbee2MQTT + Mosquitto via Ansible
- Update CloudNativePG note to include LiteLLM as consumer
- Add raspberry_pi role to repo layout

README.md

@@ -43,7 +43,7 @@ graph TB
             Longhorn
             ArgoCD
             Prometheus
-            ECK["Elastic Stack"]
+            Loki
             Istio["Istio Ambient"]
         end
         subgraph apps["Applications"]
@@ -83,8 +83,9 @@ graph TB
 | VM | `docker-host11` | Docker host w/ GPU passthrough | 2 vCPU · 4 GB RAM · 192 GB · Intel QuickSync |
 | VM | `docker-lb` | Caddy reverse proxy (LAN only) | 1 vCPU · 2 GB RAM |
 | VPS | `mii` | Edge node (Netcup) | WireGuard · Traefik · Pangolin |
+| SBC | `naruto` | Raspberry Pi | Zigbee2MQTT · Mosquitto MQTT broker |
 
-All VMs run Debian 12 on `virtio` network bridges, provisioned from cloud-init templates via Terraform + Ansible.
+All VMs run Debian 12 on `virtio` network bridges, provisioned from cloud-init templates via Terraform + Ansible. The Raspberry Pi runs Docker Compose, managed by the same Ansible setup.
 
 ---
 
@@ -100,9 +101,9 @@ All VMs run Debian 12 on `virtio` network bridges, provisioned from cloud-init t
 | **Cert-Manager** | Helm (ArgoCD) | Automated TLS via Let's Encrypt DNS-01 (Cloudflare API) |
 | **Sealed Secrets** | Helm (ArgoCD) | Encrypts secrets for safe storage in Git |
 | **Longhorn** | Helm (ArgoCD) | Distributed block storage (RWO + RWX) across all 14 agents |
-| **CloudNativePG** | Operator (ArgoCD) | HA PostgreSQL — used by Immich |
+| **CloudNativePG** | Operator (ArgoCD) | HA PostgreSQL — used by Immich and LiteLLM |
-| **Elastic Stack (ECK)** | Operator (ArgoCD) | Elasticsearch + Kibana + Fleet + Elastic Agents for observability |
 | **Kube-Prometheus-Stack** | Helm (ArgoCD) | Prometheus + Grafana monitoring |
+| **Loki + Promtail** | Helm (ArgoCD) | Log aggregation — Promtail DaemonSet ships pod logs to Loki |
 | **Goldilocks + VPA** | Helm (ArgoCD) | Resource usage analysis and request/limit rightsizing |
 | **Istio (Ambient)** | Helm (ArgoCD) | Service mesh — ztunnel DaemonSet on all nodes (L4); no Waypoint proxies yet |
 | **K3s Upgrade Controller** | Operator (ArgoCD) | Automated rolling K3s version upgrades |
@@ -130,6 +131,9 @@ All VMs run Debian 12 on `virtio` network bridges, provisioned from cloud-init t
 | **Arr Stack** | Media automation suite | Prowlarr · Sonarr · Radarr · Unpackarr |
 | **Download clients** | VPN-isolated download clients (×2) | Gluetun sidecar |
 | **Jellyfin** | Media server with hardware transcoding | Docker · Intel QuickSync |
+| **LiteLLM** | OpenAI-compatible LLM proxy — routes to Gemini, DeepSeek, and other providers | CloudNativePG |
+| **agentmemory** | Persistent memory server for AI agents | Longhorn PVC · private Gitea registry |
+| **athome** | Personal homepage | – |
 
 ---
 
@@ -143,7 +147,7 @@ No ports are open on the home router. Internal load balancing goes through Metal
 
 Longhorn handles block storage by replicating volumes across all 14 agent nodes. The media library lives on a dedicated NFS host instead — latency matters when Jellyfin is reading large video files, and NFS is simpler for that.
 
-Metrics go to Prometheus + Grafana. Logs and fleet management go to Elastic Stack via the ECK operator, with Elastic Agents running as a DaemonSet so every node is covered.
+Metrics go to Prometheus + Grafana. Logs go to Loki, also surfaced in Grafana — Promtail runs as a DaemonSet and ships pod logs from every node.
 
 All VMs are provisioned with Terraform and configured by Ansible. Rebuilding from scratch doesn't require remembering anything.
 
@@ -160,7 +164,8 @@ ansible-homelab/
 │   ├── kube_vip/         # kube-vip DaemonSet + TLS SAN config
 │   ├── docker_host/      # Docker + Intel QuickSync GPU passthrough
 │   ├── proxmox/          # Proxmox node setup
-│   └── edge_vps/         # VPS: WireGuard, Traefik, Pangolin, Elastic Agent
+│   ├── edge_vps/         # VPS: WireGuard, Traefik, Pangolin, Elastic Agent
+│   └── raspberry_pi/     # Raspberry Pi: Zigbee2MQTT + Mosquitto via Docker Compose
 └── playbooks/
 
 argocd-homelab/

docs/observability.md

@@ -1,6 +1,6 @@
 # Observability
 
-Two parallel stacks: Prometheus for metrics, Elastic for logs.
+Metrics and logs both surface in Grafana, keeping dashboards and log queries in one place.
 
 ---
 
@@ -16,21 +16,9 @@ Alertmanager routes alerts to Ntfy via a custom webhook bridge.
 
 ---
 
-## Logs and fleet management
+## Logs
 
-The ECK operator (Elastic Cloud on Kubernetes) manages the Elastic stack in the `elastic-system` namespace:
+Loki runs in the `loki` namespace (single-node, filesystem storage). Promtail runs as a DaemonSet and ships pod logs from every cluster node to Loki. Logs are queryable in Grafana alongside metrics.
-
-| Component | Purpose |
-|-----------|---------|
-| Elasticsearch | Log storage and search (single-node, 15 Gi heap) |
-| Kibana | Log exploration and dashboards |
-| Fleet Server | Manages Elastic Agent enrollment and policies |
-| Elastic Agent (DaemonSet) | Ships logs and metrics from every cluster node |
-| Elastic Agent (standalone) | Runs on docker-host11 and the edge VPS |
-
-The DaemonSet tolerates the control-plane `NoSchedule` taint so server nodes are covered too.
-
-Elastic alert rules are bridged to Ntfy via `elastic-ntfy-bridge`, a small CronJob that polls the Elasticsearch alerts API and forwards new alerts as push notifications.
 
 ---
 
@@ -38,8 +26,4 @@ Elastic alert rules are bridged to Ntfy via `elastic-ntfy-bridge`, a small CronJ
 
 ```
 Prometheus Alertmanager ──► Ntfy (push notification)
-                                      ▲
-Elasticsearch alert rule ──► elastic-ntfy-bridge CronJob ─┘
 ```
-
-Both sources land in the same Ntfy topic.