add Grafana Alloy log shipper + data capture suite logging gaps

- Add Alloy service to local observability docker-compose stack
  to tail plugin-structured.jsonl and push to Loki
- Add config.alloy with JSONL parsing and label extraction
  (level, component, event, domain)
- Data capture suite dashboard and plugin logging improvements
- Deploy script and dashboard bridge updates

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: wgutmann

.env.example

@@ -74,6 +74,16 @@ SIMSTEWARD_LOG_DEBUG=              # set to 1 for local debug mode only
 # For OTLP HTTP on 4318: also set OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
 # Prometheus UI (host): http://localhost:9090 — Grafana datasource UID: prometheus_local
 
+# --- Sentry (error tracking + release tracking) ---
+# 3 separate Sentry projects — each component has its own DSN:
+#   simhub-plugin:    C# plugin (DSN used in plugin Init, configurable via env var)
+#   index-dashboard:  Main dashboard (DSN hardcoded in index.html)
+#   test-dashboard:   Data capture suite (DSN hardcoded in data-capture-suite.html)
+# Plugin DSN (overrides hardcoded default):
+# SIMSTEWARD_SENTRY_DSN=https://ab2d0a6f7cd97033a46f4fa7d90dabab@o4511097126780928.ingest.us.sentry.io/4511102961319936
+# Auth token for deploy.ps1 release/deploy tracking (Organization:Read, Release:Admin scopes):
+# SENTRY_AUTH_TOKEN=sntrys_...
+
 # --- Local observability stack (observability/local/docker-compose.yml) ---
 # GRAFANA_STORAGE_PATH=           # e.g. S:/sim-steward-grafana-storage (Loki, Grafana, Prometheus TSDB, local stack data)
 # LOKI_PUSH_TOKEN=                # optional; if Loki gateway requires auth

deploy.ps1

@@ -107,7 +107,8 @@ $PluginDlls = @(
     "Fleck.dll",
     "Newtonsoft.Json.dll",
     "IRSDKSharper.dll",
-    "YamlDotNet.dll"
+    "YamlDotNet.dll",
+    "Sentry.dll"
 )
 
 function Read-PluginDllProductVersion {
@@ -411,6 +412,62 @@ if (-not [string]::IsNullOrWhiteSpace($script:SimStewardPluginVersionDeployed))
 }
 Write-Host ""
 
+# ── Sentry release + deploy tracking ────────────────────────────────────────
+$sentryOrg = 'sim-steward'
+$sentryProjects = @('simhub-plugin', 'web-dashboards')
+$sentryAuthToken = if (-not [string]::IsNullOrWhiteSpace($env:SENTRY_AUTH_TOKEN)) { $env:SENTRY_AUTH_TOKEN }
+                   elseif (-not [string]::IsNullOrWhiteSpace($env:SENTRY_ELEVATED_API_KEY)) { $env:SENTRY_ELEVATED_API_KEY }
+                   else { $null }
+$sentryRelease = if (-not [string]::IsNullOrWhiteSpace($script:SimStewardPluginVersionDeployed)) { $script:SimStewardPluginVersionDeployed } else { $null }
+
+function Push-SentryApi {
+    param([string]$Path, [hashtable]$Body)
+    if ([string]::IsNullOrWhiteSpace($sentryAuthToken) -or [string]::IsNullOrWhiteSpace($sentryRelease)) { return }
+    try {
+        $url = "https://sentry.io/api/0/organizations/$sentryOrg/$Path"
+        $json = $Body | ConvertTo-Json -Compress -Depth 5
+        $headers = @{ Authorization = "Bearer $sentryAuthToken"; 'Content-Type' = 'application/json' }
+        Invoke-RestMethod -Uri $url -Method Post -Headers $headers -Body $json -ErrorAction Stop | Out-Null
+    } catch {
+        # Non-fatal: deploy must not fail because Sentry API is down
+        Write-Warning "Sentry API ($Path): $($_.Exception.Message)"
+    }
+}
+
+if (-not [string]::IsNullOrWhiteSpace($sentryAuthToken) -and -not [string]::IsNullOrWhiteSpace($sentryRelease)) {
+    Write-Host "Registering Sentry release: $sentryRelease (3 projects)"
+
+    # Create release across all 3 projects with commits
+    $fullSha = ''
+    try { $fullSha = (& git -C $PluginRoot rev-parse HEAD 2>$null) } catch {}
+    Push-SentryApi "releases/" @{
+        version  = $sentryRelease
+        projects = $sentryProjects
+        refs     = @(@{ repository = "simsteward/simhub-plugin"; commit = $fullSha })
+    }
+
+    # Deploy: simhub-plugin (C# DLLs)
+    Push-SentryApi "releases/$sentryRelease/deploys/" @{
+        environment = 'local'
+        name        = 'simhub-plugin'
+    }
+
+    # Deploy: web-dashboards (all HTML/JS dashboards)
+    Push-SentryApi "releases/$sentryRelease/deploys/" @{
+        environment = 'local'
+        name        = 'web-dashboards'
+    }
+
+    Write-Host "Sentry release + 2 deploys registered (simhub-plugin, web-dashboards)."
+    Push-LokiEvent 'deploy_sentry_release' 'INFO' "Sentry release registered: $sentryRelease" @{
+        sentry_release  = $sentryRelease
+        sentry_org      = $sentryOrg
+        sentry_projects = ($sentryProjects -join ',')
+    }
+} elseif ([string]::IsNullOrWhiteSpace($sentryAuthToken)) {
+    Write-Host "Skipping Sentry release tracking (SENTRY_AUTH_TOKEN not set)."
+}
+
 # ── 5. Re-launch SimHub ─────────────────────────────────────────────────────
 $skipLaunch = $env:SIMHUB_SKIP_LAUNCH -eq "1"
 if ($skipLaunch) {

observability/local/config.alloy

@@ -0,0 +1,43 @@
+// Grafana Alloy — tail plugin-structured.jsonl → Loki
+// Docs: https://grafana.com/docs/alloy/latest/
+
+local.file_match "simsteward_structured" {
+  path_targets = [{"__path__" = "/var/log/simsteward/plugin-structured.jsonl"}]
+  sync_period  = "5s"
+}
+
+loki.source.file "simsteward_structured" {
+  targets    = local.file_match.simsteward_structured.targets
+  forward_to = [loki.process.simsteward.receiver]
+
+  tail_from_end = true
+}
+
+loki.process "simsteward" {
+  forward_to = [loki.write.local.receiver]
+
+  // Extract low-cardinality labels from JSON; everything else stays in the log line.
+  stage.json {
+    expressions = {
+      level     = "level",
+      component = "component",
+      event     = "event",
+      domain    = "domain",
+    }
+  }
+
+  stage.labels {
+    values = {
+      level     = "",
+      component = "",
+      event     = "",
+      domain    = "",
+    }
+  }
+}
+
+loki.write "local" {
+  endpoint {
+    url = "http://loki:3100/loki/api/v1/push"
+  }
+}

observability/local/docker-compose.yml

@@ -81,6 +81,17 @@ services:
       - ${GRAFANA_STORAGE_PATH:-S:/sim-steward-grafana-storage}/grafana:/var/lib/grafana
       - ./grafana/provisioning:/etc/grafana/provisioning:ro
 
+  alloy:
+    image: grafana/alloy:v1.5.1
+    depends_on:
+      loki:
+        condition: service_healthy
+    volumes:
+      - ./config.alloy:/etc/alloy/config.alloy:ro
+      - ${SIMSTEWARD_DATA_PATH}:/var/log/simsteward:ro
+      - ${GRAFANA_STORAGE_PATH:-S:/sim-steward-grafana-storage}/alloy:/tmp/positions
+    command: ["run", "/etc/alloy/config.alloy", "--storage.path=/tmp/positions"]
+
   data-api:
     build: ./data-api
     ports:

src/SimSteward.Dashboard/data-capture-suite.html

@@ -244,6 +244,15 @@
 .filter-clear { background: none; border: none; color: var(--muted); cursor: pointer; font-size: 0.82rem; padding: 0 4px; margin-left: -28px; margin-right: 4px; transition: color 0.15s; z-index: 1; }
 .filter-clear:hover { color: var(--text); }
 </style>
+<script src="https://browser.sentry-cdn.com/9.27.0/bundle.min.js" crossorigin="anonymous"></script>
+<script>
+Sentry.init({
+  dsn: 'https://46ad94b35a51979ba4223eeffe38f5c8@o4511097126780928.ingest.us.sentry.io/4511103122210816',
+  environment: 'local',
+  tracesSampleRate: 1.0,
+});
+Sentry.setTag('dashboard', 'data-capture-suite');
+</script>
 </head>
 <body>
 <div class="wrap">
@@ -368,6 +377,9 @@
 // ── Microinteraction helpers ─────────────────────────────────────────
 function logUI(elementId, eventType, message) {
   send({ action: 'log', event: 'dashboard_ui_event', element_id: elementId, event_type: eventType, message: message });
+  if (typeof Sentry !== 'undefined') {
+    Sentry.addBreadcrumb({ category: 'ui.' + eventType, message: message, data: { element_id: elementId }, level: 'info' });
+  }
 }
 
 function flashEl(el, cls, ms) {
@@ -877,18 +889,26 @@
     }
     // Log after open so send() works
     logUI('ws-pill', 'state_change', 'WebSocket connected');
+    if (typeof Sentry !== 'undefined') {
+      Sentry.addBreadcrumb({ category: 'ws', message: 'WebSocket connected', level: 'info' });
+      Sentry.setTag('ws_connected', 'true');
+    }
   });
   ws.addEventListener('close', () => {
     _wsOk = false;
     _pluginSeen = false;
     setPill(false);
+    if (typeof Sentry !== 'undefined') {
+      Sentry.addBreadcrumb({ category: 'ws', message: 'WebSocket disconnected', level: 'warning' });
+      Sentry.setTag('ws_connected', 'false');
+    }
     setTimeout(connect, 2000);
   });
   ws.addEventListener('message', ev => {
     try {
       const msg = JSON.parse(ev.data);
       if (msg.type === 'state') { _pluginSeen = true; handleState(msg); }
-    } catch {}
+    } catch (e) { if (typeof Sentry !== 'undefined') Sentry.captureException(e); }
   });
 }
 
@@ -1009,6 +1029,9 @@
   // Version display
   var verEl = document.getElementById('lbl-version');
   if (verEl && msg.pluginVersion) verEl.textContent = 'v' + msg.pluginVersion;
+  if (msg.pluginVersion && typeof Sentry !== 'undefined') {
+    Sentry.setTag('plugin_version', msg.pluginVersion);
+  }
 
   // Signals
   setSig('sig-ws', _wsOk);

src/SimSteward.Dashboard/index.html

@@ -475,6 +475,16 @@
 ::-webkit-scrollbar-track { background: transparent; }
 ::-webkit-scrollbar-thumb { background: var(--border); border-radius: 2px; }
 </style>
+<script src="https://browser.sentry-cdn.com/9.27.0/bundle.min.js" crossorigin="anonymous"></script>
+<script>
+Sentry.init({
+  dsn: 'https://46ad94b35a51979ba4223eeffe38f5c8@o4511097126780928.ingest.us.sentry.io/4511103122210816',
+  environment: 'local',
+  release: '',
+  tracesSampleRate: 1.0,
+  initialScope: { tags: { dashboard: 'index' } },
+});
+</script>
 </head>
 <body>
 <div class="app">
@@ -766,15 +776,26 @@
       renderInc();
       renderStd();
       log('health','INFO ','ws_connected','WebSocket connected to plugin at ' + url);
+      if (typeof Sentry !== 'undefined') {
+        Sentry.addBreadcrumb({ category: 'ws', message: 'WebSocket connected', level: 'info' });
+        Sentry.setTag('ws_connected', true);
+      }
     };
     ws.onclose = () => {
       ws = null; mockPaused = false;
       setWs('disconnected');
       log('health','WARN ','ws_closed','Connection lost — retrying in 3 s');
+      if (typeof Sentry !== 'undefined') {
+        Sentry.addBreadcrumb({ category: 'ws', message: 'WebSocket disconnected', level: 'warning' });
+        Sentry.setTag('ws_connected', false);
+      }
       setTimeout(connectWs, 3000);
     };
     ws.onerror = () => { ws = null; mockPaused = false; };
-    ws.onmessage = e => { try { onMsg(JSON.parse(e.data)); } catch {} };
+    ws.onmessage = e => {
+      try { onMsg(JSON.parse(e.data)); }
+      catch (err) { if (typeof Sentry !== 'undefined') Sentry.captureException(err); }
+    };
   } catch {
     setWs('disconnected');
     setTimeout(connectWs, 3000);
@@ -789,6 +810,14 @@
 
 /** Structured UI telemetry for Loki (plugin → plugin-structured.jsonl; ingestion outside plugin). */
 function sendDashboardUiEvent({ element_id, event_type, message, value }) {
+  if (typeof Sentry !== 'undefined') {
+    Sentry.addBreadcrumb({
+      category: 'ui.' + event_type,
+      message: message,
+      data: { element_id },
+      level: 'info',
+    });
+  }
   if (!ws || ws.readyState !== WebSocket.OPEN) return false;
   const o = { action: 'log', event: 'dashboard_ui_event', element_id, event_type, message };
   if (value !== undefined && value !== null) o.value = value;
@@ -829,6 +858,11 @@
     const pv = m.pluginVersion ? String(m.pluginVersion) : '';
     pvEl.textContent = pv ? ('v' + pv) : 'v—';
     pvEl.title = pv ? ('Sim Steward plugin ' + pv) : 'Sim Steward plugin build';
+    if (pv && typeof Sentry !== 'undefined') {
+      const client = Sentry.getClient();
+      if (client && client.getOptions) client.getOptions().release = pv;
+      Sentry.setTag('plugin_version', pv);
+    }
   }
   const pill = document.getElementById('mode-pill');
   pill.className = 'mode-pill ' + (mode==='Replay'?'replay':'waiting');

src/SimSteward.Plugin/DashboardBridge.cs

@@ -3,6 +3,7 @@
 using Fleck;
 using Newtonsoft.Json;
 using Newtonsoft.Json.Linq;
+using Sentry;
 
 namespace SimSteward.Plugin
 {
@@ -72,6 +73,7 @@ public void Start(string bindAddress, int port, string authToken)
                         }
                         catch (Exception ex)
                         {
+                            SentrySdk.CaptureException(ex);
                             _logger?.Warn($"DashboardBridge: getStateForNewClient failed: {ex.Message}");
                         }
                         try
@@ -82,6 +84,7 @@ public void Start(string bindAddress, int port, string authToken)
                         }
                         catch (Exception ex)
                         {
+                            SentrySdk.CaptureException(ex);
                             _logger?.Warn($"DashboardBridge: getLogTailForNewClient failed: {ex.Message}");
                         }
                     };
@@ -106,6 +109,7 @@ public void Start(string bindAddress, int port, string authToken)
             }
             catch (Exception ex)
             {
+                SentrySdk.CaptureException(ex);
                 _logger?.Error($"DashboardBridge: failed to start: {ex.Message}", ex);
                 throw;
             }
@@ -117,6 +121,7 @@ public void Stop()
             try { _server.Dispose(); }
             catch (Exception ex)
             {
+                SentrySdk.CaptureException(ex);
                 _logger?.Warn($"DashboardBridge: dispose error: {ex.Message}");
             }
             _server = null;

src/SimSteward.Plugin/SimSteward.Plugin.csproj

@@ -35,6 +35,7 @@
     <PackageReference Include="System.Net.Http" Version="4.3.4" />
     <PackageReference Include="OpenTelemetry" Version="1.9.0" />
     <PackageReference Include="OpenTelemetry.Exporter.OpenTelemetryProtocol" Version="1.9.0" />
+    <PackageReference Include="Sentry" Version="4.13.0" />
   </ItemGroup>
 
 </Project>

src/SimSteward.Plugin/SimStewardPlugin.DataCaptureSuite.cs

@@ -5,6 +5,7 @@
 using System.Diagnostics;
 using System.Linq;
 using IRSDKSharper;
+using Sentry;
 
 namespace SimSteward.Plugin
 {
@@ -151,6 +152,7 @@ private void ProcessDataCaptureSuiteTick()
                 }
                 catch (Exception ex)
                 {
+                    SentrySdk.CaptureException(ex);
                     _preflightSnapshot.Phase = "error";
                     _preflightSnapshot.Error = "BeginPreflight: " + ex.GetType().Name + ": " + ex.Message;
                     _preflightStep = PreflightStep.Complete;
@@ -161,6 +163,7 @@ private void ProcessDataCaptureSuiteTick()
                 try { TickPreflight(); }
                 catch (Exception ex)
                 {
+                    SentrySdk.CaptureException(ex);
                     _preflightSnapshot.Phase = "error";
                     _preflightSnapshot.Error = "TickPreflight@" + _preflightStep + ": " + ex.GetType().Name + ": " + ex.Message;
                     _preflightStep = PreflightStep.Complete;
@@ -411,6 +414,7 @@ private void TickPreflight()
                     }
                     catch (Exception ex)
                     {
+                        SentrySdk.CaptureException(ex);
                         SetPfTest("PC_CHECKERED", false, "seek_failed: " + ex.Message);
                         SetPfTest("PC_RESULTS", false, "seek_failed");
                         _preflightSnapshot.Error = "seek_failed: " + ex.Message;
@@ -681,6 +685,8 @@ private void BeginDataCaptureSuite()
 
             EmitSuiteLifecycleEvent(DataCaptureSuiteConstants.EventSuiteStarted,
                 $"Data capture suite started. test_run_id={_suiteTestRunId}", "T_start");
+            SentrySdk.AddBreadcrumb("Data capture suite started", "lifecycle",
+                data: new Dictionary<string, string> { ["test_run_id"] = _suiteTestRunId });
             _logger?.Info($"DataCaptureSuite started. test_run_id={_suiteTestRunId}");
         }