observability: the LLM surfaces follow the two-endpoint split

frank-large has no /metrics to scrape any more, and scraping it would be red by
construction: the halogen engine publishes /health, /cache and /v1/* only. That
job is removed with the reason written down (the remedy is the blackbox
exporter this fleet already wants for the comfyui/services probes), and
frank-small's llama.cpp endpoint takes the slot at 192.168.0.143:7771.

Homepage: frank-large's card still described the ik runtime, and the new
endpoint had no card at all. Both fixed, with a loopback probe -- the checker
runs on Dankclaw, which is also what makes the card green-able.

Dashboard: the endpoint picker was sourced from a vLLM-only metric, so a
llama.cpp job could never be selected, and the prose named frank-large as the
llama.cpp endpoint. The picker now matches both runtimes' series and the
llama.cpp targets carry the job filter every other panel already used.

The seams assert the new contract including the ABSENCE of a 7770 target, so a
well-meaning edit cannot re-add a permanently-red probe.

shared/grafana/provisioning/dashboards/llm-fleet.json

@@ -1,7 +1,7 @@
 {
   "uid": "llm-fleet",
   "title": "LLM fleet — capacity and reuse",
-  "description": "The signals that explain frank-medium under load, from the 2026-09-16 capacity work: per-card KV fill, queue depth, prefix-cache reuse, and request latency. frank-large (llama.cpp) has throughput counters only. See serverAppConfigs-docs/LLM/frank-medium-radiance-mxfp4.md for the numbers these panels are read against.",
+  "description": "The signals that explain frank-medium under load, from the 2026-09-16 capacity work: per-card KV fill, queue depth, prefix-cache reuse, and request latency. frank-small (llama.cpp on Dankclaw's RTX 3090) exports throughput counters and appears in the same endpoint picker; frank-large runs the halogen engine, which publishes /health, /cache and /v1/* and no metrics endpoint at all, so it cannot appear here. See serverAppConfigs-docs/LLM/frank-medium-radiance-mxfp4.md for the numbers these panels are read against.",
   "tags": [
     "llm",
     "frank-medium",
@@ -26,7 +26,7 @@
           "uid": "prometheus"
         },
         "query": {
-          "query": "label_values(vllm:kv_cache_usage_perc, job)",
+          "query": "label_values({__name__=~\"vllm:kv_cache_usage_perc|llamacpp:.*\"}, job)",
           "refId": "A"
         },
         "includeAll": true,
@@ -185,7 +185,7 @@
       "id": 5,
       "type": "timeseries",
       "title": "Decode and prefill throughput",
-      "description": "vLLM generation tokens, and llama.cpp's counters for frank-large (llamacpp:prompt_tokens_seconds is prompt throughput).",
+      "description": "vLLM generation tokens, plus llama.cpp's counters for frank-small: llamacpp:predicted_tokens_seconds is decode and llamacpp:prompt_tokens_seconds is prompt throughput. The llama.cpp series carry the job label of their scrape, so the endpoint picker selects between the runtimes.",
       "gridPos": {
         "h": 8,
         "w": 12,
@@ -219,8 +219,13 @@
         },
         {
           "refId": "C",
-          "expr": "llamacpp:predicted_tokens_seconds",
-          "legendFormat": "{{job}} llama.cpp decode"
+          "expr": "llamacpp:predicted_tokens_seconds{job=~\"$job\"}",
+          "legendFormat": "{{job}} llama.cpp decode tok/s"
+        },
+        {
+          "refId": "D",
+          "expr": "llamacpp:prompt_tokens_seconds{job=~\"$job\"}",
+          "legendFormat": "{{job}} llama.cpp prefill tok/s"
         }
       ]
     },

shared/homepage/services.json

@@ -222,7 +222,7 @@
   {
     "name": "frank-large",
     "category": "admin",
-    "description": "The large LLM (Qwen3.8-Flash-Next Uncensored, ik_llama.cpp) on Dankclaw.",
+    "description": "The large LLM (Qwen3.8-Flash-Next w4b, halogen-flash-server on the Strix Halo iGPU). Two 200k-token conversations resident.",
     "url": "",
     "docs": "",
     "probe": "127.0.0.1:7770",
@@ -241,6 +241,17 @@
     "wip": false,
     "showLink": false
   },
+  {
+    "name": "frank-small",
+    "category": "admin",
+    "description": "The small LLM (Qwen3.8-27B GSQ-RCO IQ3_S with MTP and vision, llama.cpp) on Dankclaw's RTX 3090. The fleet's image path.",
+    "url": "",
+    "docs": "",
+    "probe": "127.0.0.1:7771",
+    "method": "http",
+    "wip": false,
+    "showLink": false
+  },
   {
     "name": "mariadb",
     "category": "admin",

shared/prometheus/prometheus.yml

@@ -68,7 +68,15 @@ scrape_configs:
   # and the prefix-cache pair vllm:prefix_cache_hits_total over
   # vllm:prefix_cache_queries_total (their rates give the hit rate, which says
   # whether pinned conversations are actually being reused).
-  # frank-large is llama.cpp and exposes throughput counters under llamacpp:.
+  # frank-large is deliberately NOT scraped: the halogen engine that serves it
+  # publishes /health, /cache and /v1/* and no Prometheus text endpoint, so a
+  # target here would be red by construction and would fire LlmEndpointDown
+  # forever. Its liveness is on the homepage (a loopback probe) and in the
+  # engine's own boot ledger. The remedy, when the fleet wants it, is the
+  # blackbox exporter already wanted for the comfyui/services probes.
+  #
+  # frank-small is llama.cpp (its serve script passes --metrics) and does
+  # export text-format counters under llamacpp: on the port its unit publishes.
   - job_name: "frank-medium"
     metrics_path: "/metrics"
     static_configs:
@@ -76,9 +84,9 @@ scrape_configs:
         labels:
           service: frank-medium
 
-  - job_name: "frank-large"
+  - job_name: "frank-small"
     metrics_path: "/metrics"
     static_configs:
-      - targets: ["192.168.0.143:7770"]
+      - targets: ["192.168.0.143:7771"]
         labels:
-          service: frank-large
+          service: frank-small

shared/prometheus/rules/llm-fleet.yml

@@ -75,9 +75,13 @@ groups:
         annotations:
           summary: "{{ $labels.job }}: scrape target down for 2 minutes"
           description: >-
-            Prometheus cannot reach the LLM endpoint. Check the host's units:
-            systemctl --user is-active frank-medium-{a,b,pro} on Dankclaw's
-            peers (frank-large) or Dankserver (frank-medium).
+            Prometheus cannot reach the LLM endpoint. Check the units:
+            `systemctl --user is-active frank-medium-{a,b,proxy}` on Dankserver,
+            `systemctl is-active frank-small` (the llama.cpp root unit) or
+            `systemctl --user is-active frank-large` (the halogen quadlet) on
+            Dankclaw. frank-large is not scraped at all — the halogen engine
+            exports no metrics — so this alert can only ever name the endpoints
+            that do export one.
 
       # What an operator actually feels: wall-clock per request, prefill
       # included. A cold 50k-token prefill alone is ~22 s on one card.