chore: IIS Modernization

microsoft-iis-mixin/README.md

@@ -9,11 +9,11 @@ The Microsoft IIS mixin contains the following dashboards:
 
 and the following alerts:
 
-- MicrosoftIISHighNumberOfRejectedAsyncIORequests
-- MicrosoftIISHighNumberOf5xxRequestErrors
-- MicrosoftIISLowSuccessRateForWebsocketConnections
-- MicrosoftIISThreadpoolUtilizationNearingMax
-- MicrosoftIISHighNumberOfWorkerProcessFailures
+- MicrosoftIISRejectedAsyncIORequests
+- MicrosoftIIS5xxRequestErrors
+- MicrosoftIISSuccessRateForWebsocket
+- MicrosoftIISThreadpoolUtilization
+- MicrosoftIISWorkerProcessFailures
 
 Default thresholds can be configured in `config,libsonnet`
 
@@ -57,11 +57,11 @@ The Microsoft IIS applications dashboard provides details on worker requests, we
 ![Screenshot3 of the applications dashboard](https://storage.googleapis.com/grafanalabs-integration-assets/iis/screenshots/application-3.png)
 ## Alerts overview
 
-MicrosoftIISHighNumberOfRejectedAsyncIORequests: There are a high number of rejected async I/O requests for a site.
-MicrosoftIISHighNumberOf5xxRequestErrors: There are a high number of 5xx request errors for an application.
-MicrosoftIISLowSuccessRateForWebsocketConnections: There is a low success rate for websocket connections for an application.
-MicrosoftIISThreadpoolUtilizationNearingMax: The thread pool utilization is nearing max capacity.
-MicrosoftIISHighNumberOfWorkerProcessFailures: There are a high number of worker process failures for an application.
+- `MicrosoftIISRejectedAsyncIORequests`: There are a high number of rejected async I/O requests for a site.
+- `MicrosoftIIS5xxRequestErrors`: There are a high number of 5xx request errors for an application.
+- `MicrosoftIISSuccessRateForWebsocket`: There is a low success rate for websocket connections for an application.
+- `MicrosoftIISThreadpoolUtilization`: The thread pool utilization is nearing max capacity.
+- `MicrosoftIISWorkerProcessFailures`: There are a high number of worker process failures for an application.
 
 ## Install Tools
 

microsoft-iis-mixin/alerts.libsonnet

@@ -0,0 +1,91 @@
+{
+  new(this): {
+    groups: [
+      {
+        name: 'MicrosoftIISAlerts',
+        rules: [
+          {
+            alert: 'MicrosoftIISRejectedAsyncIORequests',
+            expr: |||
+              increase(windows_iis_rejected_async_io_requests_total{%(filteringSelector)s}[5m]) > %(alertsWarningHighRejectedAsyncIORequests)s
+            ||| % this.config,
+            'for': '5m',
+            labels: {
+              severity: 'warning',
+            },
+            annotations: {
+              summary: 'There are a high number of rejected async I/O requests for a site.',
+              description:
+                ('The number of rejected async IO requests is {{ printf "%%.0f" $value }} over the last 5m on {{ $labels.instance }} - {{ $labels.site }}, ' +
+                 'which is above the threshold of %(alertsWarningHighRejectedAsyncIORequests)s.') % this.config,
+            },
+          },
+          {
+            alert: 'MicrosoftIIS5xxRequestErrors',
+            expr: |||
+              sum without (pid, status_code)(increase(windows_iis_worker_request_errors_total{status_code=~"5.."%(filteringSelector)s}[5m])) > %(alertsCriticalHigh5xxRequests)s
+            ||| % (this.config { filteringSelector: if this.config.filteringSelector != '' then ',' + this.config.filteringSelector else '' }),
+            'for': '5m',
+            labels: {
+              severity: 'critical',
+            },
+            annotations: {
+              summary: 'There are a high number of 5xx request errors for an application.',
+              description:
+                ('The number of 5xx request errors is {{ printf "%%.0f" $value }} over the last 5m on {{ $labels.instance }} - {{ $labels.app }}, ' +
+                 'which is above the threshold of %(alertsCriticalHigh5xxRequests)s.') % this.config,
+            },
+          },
+          {
+            alert: 'MicrosoftIISSuccessRateForWebsocket',
+            expr: |||
+              sum without (pid)  (increase(windows_iis_worker_websocket_connection_accepted_total{%(filteringSelector)s}[5m]) / clamp_min(increase(windows_iis_worker_websocket_connection_attempts_total{%(filteringSelector)s}[5m]),1)) * 100 < %(alertsCriticalLowWebsocketConnectionSuccessRate)s
+            ||| % this.config,
+            'for': '5m',
+            labels: {
+              severity: 'critical',
+            },
+            annotations: {
+              summary: 'There is a low success rate for websocket connections for an application.',
+              description:
+                ('The success rate for websocket connections is {{ printf "%%.0f" $value }} over the last 5m on {{ $labels.instance }} - {{ $labels.app }}, ' +
+                 'which is below the threshold of %(alertsCriticalLowWebsocketConnectionSuccessRate)s.') % this.config,
+            },
+          },
+          {
+            alert: 'MicrosoftIISThreadpoolUtilization',
+            expr: |||
+              sum without (pid, state)(windows_iis_worker_threads{%(filteringSelector)s} / windows_iis_worker_max_threads{%(filteringSelector)s}) * 100 > %(alertsCriticalHighThreadPoolUtilization)s
+            ||| % this.config,
+            'for': '5m',
+            labels: {
+              severity: 'critical',
+            },
+            annotations: {
+              summary: 'The thread pool utilization is nearing max capacity.',
+              description:
+                ('The threadpool utilization is at {{ printf "%%.0f" $value }} over the last 5m on {{ $labels.instance }} - {{ $labels.app }}, ' +
+                 'which is above the threshold of %(alertsCriticalHighThreadPoolUtilization)s.') % this.config,
+            },
+          },
+          {
+            alert: 'MicrosoftIISWorkerProcessFailures',
+            expr: |||
+              increase(windows_iis_total_worker_process_failures{%(filteringSelector)s}[5m]) > %(alertsWarningHighWorkerProcessFailures)s
+            ||| % this.config,
+            'for': '5m',
+            labels: {
+              severity: 'warning',
+            },
+            annotations: {
+              summary: 'There are a high number of worker process failures for an application.',
+              description:
+                ('The number of worker process failures is at {{ printf "%%.0f" $value }} over the last 5m on {{ $labels.instance }} - {{ $labels.app }}, ' +
+                 'which is above the threshold of %(alertsWarningHighWorkerProcessFailures)s.') % this.config,
+            },
+          },
+        ],
+      },
+    ],
+  },
+}

microsoft-iis-mixin/config.libsonnet

@@ -1,17 +1,38 @@
 {
-  _config+:: {
-    dashboardTags: ['microsoft-iis-mixin'],
-    dashboardPeriod: 'now-1h',
-    dashboardTimezone: 'default',
-    dashboardRefresh: '1m',
+  local this = self,
+  filteringSelector: '',  // set to apply static filters to all queries and alerts, i.e. job="bar"
+  groupLabels: ['job'],
+  instanceLabels: ['instance'],
+  logLabels: ['job', 'instance'],
 
-    // alerts thresholds
-    alertsWarningHighRejectedAsyncIORequests: 20,
-    alertsCriticalHigh5xxRequests: 5,
-    alertsCriticalLowWebsocketConnectionSuccessRate: 80,
-    alertsCriticalHighThreadPoolUtilization: 90,
-    alertsWarningHighWorkerProcessFailures: 10,
 
-    enableLokiLogs: true,
+  // Dashboard settings
+  dashboardTags: [this.uid + '-mixin'],
+  uid: 'microsoft-iis',
+  dashboardNamePrefix: 'Microsoft IIS',
+  dashboardPeriod: 'now-30m',
+  dashboardTimezone: 'default',
+  dashboardRefresh: '1m',
+
+  // Logs configuration
+  enableLokiLogs: true,
+  extraLogLabels: ['level'],  // Required by logs-lib
+  logsVolumeGroupBy: 'level',
+  showLogsVolume: true,
+
+  // Alert thresholds
+  alertsWarningHighRejectedAsyncIORequests: 20,  // count
+  alertsCriticalHigh5xxRequests: 5,  // %
+  alertsCriticalLowWebsocketConnectionSuccessRate: 80,  // %
+  alertsCriticalHighThreadPoolUtilization: 90,  // %
+  alertsWarningHighWorkerProcessFailures: 10,  // count
+
+  // Metrics source
+  metricsSource: ['prometheus'],
+
+  // Signal definitions grouped by dashboard
+  signals+: {
+    overview: (import './signals/overview.libsonnet')(this),
+    applications: (import './signals/applications.libsonnet')(this),
   },
 }