fix(monitoring): use bluejay_v2 auth for snmp-nas (not public_v2)

Synology NAS is configured with community bluejay_monitor
(→ snmp.yml auth 'bluejay_v2'), not public. public_v2 was returning
HTTP 500 from snmp-exporter for this target. Verified bluejay_v2
returns metrics.

Keeps printer (10.0.58.107) on public_v2 — Epson ET-3750 uses
community "public" as documented in its SNMP settings.

apps/monitoring/noc-monitoring.yaml

@@ -143,7 +143,7 @@ data:
         metrics_path: /snmp
         params:
           module: [synology]
-          auth: [public_v2]
+          auth: [bluejay_v2]
         relabel_configs:
           - source_labels: [__address__]
             target_label: __param_target
@@ -247,7 +247,7 @@ data:
               device_type: "printer"
               printer_model: "NuPrint 210"
 
-      # Print.Web OTEL metrics (counters: jobs enqueued/completed/failed, bytes, duration histograms)
+      # Print.Web OTEL metrics (print counters/histograms plus Ollama runner gauges)
       - job_name: "printweb-otel"
         scrape_interval: 30s
         metrics_path: /metrics/prometheus
@@ -433,6 +433,8 @@ data:
             expr: rate(print_jobs_completed_total[5m]) / rate(print_jobs_enqueued_total[5m])
           - record: print:job_duration_p95:5m
             expr: histogram_quantile(0.95, rate(print_job_duration_ms_bucket[5m]))
+          - record: print:ollama_runner_keepalive_remaining_seconds:max
+            expr: max by(instance, model) (ai_ollama_runner_keepalive_remaining_seconds{job="printweb-otel"})
       - name: relay-rates
         interval: 15s
         rules:
@@ -528,6 +530,16 @@ data:
               summary: "High print volume on edge2 ({{ $value | printf \"%.0f\" }} jobs/min)"
               description: "Print rate exceeds 30 jobs/min for 5 minutes. Possible runaway print loop."
 
+          - alert: PrintOllamaRunnerLongKeepAlive
+            expr: ai_ollama_runner_keepalive_remaining_seconds{job="printweb-otel"} > 600
+            for: 2m
+            labels:
+              severity: warning
+              alert_channel: thermal_print
+            annotations:
+              summary: "Print.Web Ollama runner held for >10m ({{ $labels.model }})"
+              description: "Print.Web reports model {{ $labels.model }} with {{ $value | printf \"%.0f\" }}s of keep-alive remaining. Check concurrent requests before the Pi 5 Ollama lane thrashes."
+
       - name: pi-fleet
         rules:
           - alert: PiManagerDown
@@ -2896,6 +2908,33 @@ data:
                 relativeTimeRange: {from: 120, to: 0}
                 datasourceUid: __expr__
                 model: {type: threshold, expression: B, conditions: [{evaluator: {params: [1], type: lt}}], refId: C}
+          - uid: print-ollama-runner-long-keepalive
+            title: Print.Web Ollama runner keep-alive >10m
+            condition: C
+            for: 2m
+            noDataState: NoData
+            execErrState: OK
+            annotations:
+              summary: Print.Web Ollama runner held too long
+              description: "A Print.Web Ollama runner has more than 10 minutes of keep-alive remaining. Check active AI requests before the Pi 5 Ollama lane thrashes."
+              runbook: "1. Open https://print.iamworkin.lan/admin 2. Check Ollama Fleet model + remaining keep-alive 3. Query Prometheus: ai_ollama_runner_keepalive_remaining_seconds{job=\"printweb-otel\"} 4. Stop duplicate model callers before restarting Ollama."
+            labels:
+              severity: warning
+              service: print-web
+              alert_channel: thermal_print
+            data:
+              - refId: A
+                relativeTimeRange: {from: 120, to: 0}
+                datasourceUid: prometheus
+                model: {expr: 'max by(model) (ai_ollama_runner_keepalive_remaining_seconds{job="printweb-otel"})', instant: true, refId: A}
+              - refId: B
+                relativeTimeRange: {from: 120, to: 0}
+                datasourceUid: __expr__
+                model: {type: reduce, expression: A, reducer: last, refId: B}
+              - refId: C
+                relativeTimeRange: {from: 120, to: 0}
+                datasourceUid: __expr__
+                model: {type: threshold, expression: B, conditions: [{evaluator: {params: [600], type: gt}}], refId: C}
       - orgId: 1
         name: Infrastructure
         folder: AI Stack Alerts