chore: Modernize microsoft-iis-mixin

.gitignore

@@ -2,3 +2,4 @@
 vendor
 jsonnetfile.lock.json
 *.zip
+.worktrees

microsoft-iis-mixin/.lint

@@ -1,4 +1,6 @@
 exclusions:
+  panel-datasource-rule:
+    reason: "Modern mixins use signal-based architecture where datasource references are handled by the framework"
   panel-units-rule:
     reason: "Custom units are used for better user experience in these panels"
     entries:
@@ -16,6 +18,7 @@ exclusions:
       - panel: "Worker process startup failures"
       - panel: "Worker process shutdown failures"
       - panel: "Worker process ping failures"
+      - panel: "Worker output cache hit ratio"
   template-datasource-rule:
     reason: "Based on new convention we are using variable names prometheus_datasource and loki_datasource where as linter expects 'datasource'"
   template-instance-rule:

microsoft-iis-mixin/alerts/alerts.libsonnet → microsoft-iis-mixin/alerts.libsonnet

@@ -1,14 +1,14 @@
 {
-  prometheusAlerts+:: {
-    groups+: [
+  new(this): {
+    groups: [
       {
-        name: 'microsoft-iis',
+        name: this.config.uid + '-alerts',
         rules: [
           {
-            alert: 'MicrosoftIISHighNumberOfRejectedAsyncIORequests',
+            alert: 'IISHighRejectedAsyncIORequests',
             expr: |||
               increase(windows_iis_rejected_async_io_requests_total[5m]) > %(alertsWarningHighRejectedAsyncIORequests)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -17,14 +17,14 @@
               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.
-              ||| % $._config,
+              ||| % this.config,
             },
           },
           {
-            alert: 'MicrosoftIISHighNumberOf5xxRequestErrors',
+            alert: 'IISHigh5xxRequestErrors',
             expr: |||
               sum without (pid, status_code)(increase(windows_iis_worker_request_errors_total{status_code=~"5.*"}[5m])) > %(alertsCriticalHigh5xxRequests)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'critical',
@@ -33,14 +33,14 @@
               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.
-              ||| % $._config,
+              ||| % this.config,
             },
           },
           {
-            alert: 'MicrosoftIISLowSuccessRateForWebsocketConnections',
+            alert: 'IISLowWebsocketConnectionSuccess',
             expr: |||
               sum without (pid)  (increase(windows_iis_worker_websocket_connection_accepted_total[5m]) / clamp_min(increase(windows_iis_worker_websocket_connection_attempts_total[5m]),1)) * 100 > %(alertsCriticalLowWebsocketConnectionSuccessRate)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'critical',
@@ -49,14 +49,14 @@
               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 above the threshold of %(alertsCriticalLowWebsocketConnectionSuccessRate)s.
-              ||| % $._config,
+              ||| % this.config,
             },
           },
           {
-            alert: 'MicrosoftIISThreadpoolUtilizationNearingMax',
+            alert: 'IISThreadpoolUtilizationHigh',
             expr: |||
               sum without (pid, state)(windows_iis_worker_threads / windows_iis_worker_max_threads) * 100 > %(alertsCriticalHighThreadPoolUtilization)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'critical',
@@ -65,14 +65,14 @@
               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.
-              ||| % $._config,
+              ||| % this.config,
             },
           },
           {
-            alert: 'MicrosoftIISHighNumberOfWorkerProcessFailures',
+            alert: 'IISHighWorkerProcessFailures',
             expr: |||
               increase(windows_iis_total_worker_process_failures[5m]) > %(alertsWarningHighWorkerProcessFailures)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -81,7 +81,7 @@
               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.
-              ||| % $._config,
+              ||| % this.config,
             },
           },
         ],

microsoft-iis-mixin/config.libsonnet

@@ -1,17 +1,49 @@
 {
-  _config+:: {
-    dashboardTags: ['microsoft-iis-mixin'],
-    dashboardPeriod: 'now-1h',
-    dashboardTimezone: 'default',
-    dashboardRefresh: '1m',
+  local this = self,
 
-    // alerts thresholds
-    alertsWarningHighRejectedAsyncIORequests: 20,
-    alertsCriticalHigh5xxRequests: 5,
-    alertsCriticalLowWebsocketConnectionSuccessRate: 80,
-    alertsCriticalHighThreadPoolUtilization: 90,
-    alertsWarningHighWorkerProcessFailures: 10,
+  // Basic filtering
+  filteringSelector: 'job=~"$job", instance=~"$instance"',
+  queriesSelector: 'job=~"$job", instance=~"$instance"',
+  groupLabels: ['job'],
+  instanceLabels: ['instance'],
 
-    enableLokiLogs: true,
+  // Dashboard settings
+  dashboardTags: ['microsoft-iis-mixin'],
+  uid: 'microsoft-iis',
+  dashboardNamePrefix: 'Microsoft IIS',
+  dashboardRefresh: '1m',
+  dashboardPeriod: 'now-1h',
+  dashboardTimezone: 'default',
+
+  // Logs configuration
+  enableLokiLogs: true,
+  logLabels: ['job', 'instance', 'level'],
+  extraLogLabels: [],
+  logsVolumeGroupBy: 'level',
+  showLogsVolume: true,
+
+  // Alert thresholds
+  alertsWarningHighRejectedAsyncIORequests: 20,  // count
+  alertsCriticalHigh5xxRequests: 5,  // %
+  alertsCriticalLowWebsocketConnectionSuccessRate: 80,  // %
+  alertsCriticalHighThreadPoolUtilization: 90,  // %
+  alertsWarningHighWorkerProcessFailures: 10,  // count
+
+  // Metrics source
+  metricsSource: 'prometheus',
+
+  // Signal definitions
+  signals+: {
+    requests: (import './signals/requests.libsonnet')(this),
+    connections: (import './signals/connections.libsonnet')(this),
+    data_transfer: (import './signals/data_transfer.libsonnet')(this),
+    async_io: (import './signals/async_io.libsonnet')(this),
+    server_cache: (import './signals/server_cache.libsonnet')(this),
+    app_pools: (import './signals/app_pools.libsonnet')(this),
+    worker_processes: (import './signals/worker_processes.libsonnet')(this),
+    worker_requests: (import './signals/worker_requests.libsonnet')(this),
+    worker_cache: (import './signals/worker_cache.libsonnet')(this),
+    worker_threads: (import './signals/worker_threads.libsonnet')(this),
+    worker_websocket: (import './signals/worker_websocket.libsonnet')(this),
   },
 }

microsoft-iis-mixin/dashboards.libsonnet

@@ -0,0 +1,139 @@
+local g = import './g.libsonnet';
+local logslib = import 'logs-lib/logs/main.libsonnet';
+
+{
+  local root = self,
+  new(this)::
+    local prefix = this.config.dashboardNamePrefix;
+    local links = this.grafana.links;
+    local tags = this.config.dashboardTags;
+    local uid = g.util.string.slugify(this.config.uid);
+    local vars = this.grafana.variables;
+    local annotations = this.grafana.annotations;
+    local refresh = this.config.dashboardRefresh;
+    local period = this.config.dashboardPeriod;
+    local timezone = this.config.dashboardTimezone;
+
+    {
+      // Microsoft IIS overview dashboard
+      'microsoft-iis-overview.json':
+        g.dashboard.new(prefix + ' overview')
+        + g.dashboard.withDescription('Dashboard providing an overview of Microsoft IIS performance.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.overviewRequests,
+                this.grafana.rows.overviewAsyncIO,
+                this.grafana.rows.overviewTraffic,
+                this.grafana.rows.overviewConnections,
+              ]
+              +
+              if this.config.enableLokiLogs then
+                [this.grafana.rows.overviewLogs] else []
+                                                      +
+                                                      [
+                                                        this.grafana.rows.overviewCache,
+                                                      ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.query.new('site')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('site', 'windows_iis_requests_total{job=~"$job", instance=~"$instance"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('Site')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '_overview',
+          tags,
+          links { microsoftIISOverview+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+
+      // Microsoft IIS applications dashboard
+      'microsoft-iis-applications.json':
+        g.dashboard.new(prefix + ' applications')
+        + g.dashboard.withDescription('Dashboard providing detailed application performance metrics for Microsoft IIS.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.applicationsRequests,
+                this.grafana.rows.applicationsWebsocket,
+                this.grafana.rows.applicationsWorkerProcesses,
+                this.grafana.rows.applicationsCache,
+              ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.query.new('application')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('app', 'windows_iis_current_application_pool_state{job=~"$job", instance=~"$instance"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('Application')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '_applications',
+          tags,
+          links { microsoftIISApplications+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+    }
+    +
+    if this.config.enableLokiLogs then
+      {
+        'microsoft-iis-logs.json':
+          logslib.new(
+            prefix + ' logs',
+            datasourceName=vars.datasources.loki.name,
+            datasourceRegex=vars.datasources.loki.regex,
+            filterSelector=this.config.filteringSelector,
+            labels=this.config.logLabels + this.config.extraLogLabels,
+            formatParser=null,
+            showLogsVolume=this.config.showLogsVolume,
+            logsVolumeGroupBy=this.config.logsVolumeGroupBy,
+            extraFilters=[]
+          )
+          {
+            dashboards+:
+              {
+                logs+:
+                  root.applyCommon(
+                    [],
+                    uid + '_logs',
+                    tags,
+                    links { logs+:: {} },
+                    annotations,
+                    timezone,
+                    refresh,
+                    period
+                  ),
+              },
+          }.dashboards.logs,
+      } else {},
+
+  applyCommon(vars, uid, tags, links, annotations, timezone, refresh, period):
+    g.dashboard.withTags(tags)
+    + g.dashboard.withUid(uid)
+    + g.dashboard.withLinks(std.objectValues(links))
+    + g.dashboard.withTimezone(timezone)
+    + g.dashboard.withRefresh(refresh)
+    + g.dashboard.time.withFrom(period)
+    + g.dashboard.withVariables(vars)
+    + g.dashboard.withAnnotations(std.objectValues(annotations)),
+}