fix(monitoring): scrape node-dankclaw via host.containers.internal

node-exporter is net=host, so its 192.168.0.143:9100 bind is a host bind,
not a published port — the bridge-networked Prometheus container cannot
reach its own host's LAN IP there (netavark DNATs published ports only).
The node-dankclaw target has been red by construction since the
observability cutover while the container reported healthy. Point it at
host.containers.internal:9100 (netavark delivers that to the host LAN IP,
where node-exporter listens) — the same rule frank-large and frank-small
already use. node-dankserver is a different machine and keeps its LAN
address. Update the cross-file contract test and the Targets table.

serverAppConfigs-docs/Infrastructure/Prometheus.md

@@ -25,9 +25,11 @@ Authentik; post-auth the Caddy vhost routes straight to `prometheus:9090`) —
 | Job | Target | Why |
 |-----|--------|-----|
 | `prometheus` | `localhost:9090` | itself |
-| `node-dankclaw`, `node-dankserver` | `192.168.0.143:9100`, `192.168.0.235:9100` | host metrics from each machine's [[Node-Exporter]] |
+| `node-dankclaw` | `host.containers.internal:9100` | host metrics from [[Node-Exporter]] — a net=host bind, so via the gateway (the LAN address is unreachable from this bridge container) |
+| `node-dankserver` | `192.168.0.235:9100` | host metrics from [[Node-Exporter]] — a different machine, so the LAN address works |
 | `grafana`, `loki` | `grafana:3000`, `loki:3100` | container names on the `dankclaw_default` network; both serve real `/metrics` |
-| `frank-medium`, `frank-large` | `192.168.0.235:7770`, `192.168.0.143:7770` | the LLM fleet — `vllm:` and `llamacpp:` series (LAN addresses, not the guarded vhosts) |
+| `frank-medium` | `192.168.0.235:7770` | the LLM fleet — `vllm:` series; a different machine, so the LAN address works |
+| `frank-large`, `frank-small` | `host.containers.internal:7770`, `:7771` | the LLM fleet — `llamacpp:` series; same-host net=host binds, so via the gateway |
 
 **Three jobs read down, and they are not outages** (audited 2026-09-16):
 

shared/prometheus/prometheus.yml

@@ -27,10 +27,16 @@ scrape_configs:
     static_configs:
       - targets: ["localhost:9090"]
 
-  # Host metrics — both hosts
+  # Host metrics — both hosts. node-dankclaw is scraped over
+  # host.containers.internal, not its own LAN address: this Prometheus is a
+  # bridge container, and on this netavark host a container can only reach the
+  # host's LAN IP at a PUBLISHED port — node-exporter's net=host bind on
+  # 192.168.0.143:9100 is refused by construction (the same rule that sends
+  # frank-large and frank-small to the gateway below). node-dankserver is a
+  # different machine, so its LAN address works.
   - job_name: "node-dankclaw"
     static_configs:
-      - targets: ["192.168.0.143:9100"]
+      - targets: ["host.containers.internal:9100"]
 
   - job_name: "node-dankserver"
     static_configs:

tests/observability_quadlet.bats

@@ -108,12 +108,27 @@ service_section() {
 @test "prometheus scrapes node-exporter where node-exporter actually listens" {
     # Cross-file contract: if either side's address changes alone, host metrics
     # silently stop with no error anywhere.
+    #
+    # This Prometheus is a bridge container (Network=dankclaw_default), and on
+    # this netavark host a container can only reach the host's LAN IP at a
+    # PUBLISHED port. node-exporter is net=host, so its 192.168.0.143:9100 bind
+    # is a host bind, not a published port — unreachable at its own LAN address.
+    # The node-dankclaw job therefore scrapes host.containers.internal:9100,
+    # which netavark delivers to the host's LAN IP, where node-exporter listens.
+    # (node-dankserver is a different machine, so its LAN address is reached
+    # over the real network and is scraped directly.)
     local unit="$UNIT_DIR/node-exporter.container"
     local cfg="$REPO_ROOT/shared/prometheus/prometheus.yml"
-    local addr
+    local lan
-    addr=$(grep -oE '192\.168\.0\.[0-9]+:9100' "$unit" | head -1)
+    lan=$(grep -oE '192\.168\.0\.[0-9]+:9100' "$unit" | head -1)
-    [ -n "$addr" ] || { echo "node-exporter has no LAN listen address"; return 1; }
+    [ -n "$lan" ] || { echo "node-exporter has no LAN listen address"; return 1; }
-    grep -q "\"$addr\"" "$cfg" || { echo "prometheus does not scrape $addr"; return 1; }
+    # The exporter must still bind that LAN address — it is where the
+    # host.containers.internal delivery lands.
+    grep -q -- "--web.listen-address=$lan" "$unit" \
+        || { echo "node-exporter no longer binds $lan"; return 1; }
+    # The local job scrapes via the gateway, not the (unreachable) LAN address.
+    grep -q '"host.containers.internal:9100"' "$cfg" \
+        || { echo "node-dankclaw does not scrape host.containers.internal:9100"; return 1; }
 }
 
 @test "container names still match what Caddy and Prometheus address by name" {