fix: correct alb metrics to use gauge promql functions (#639)

* fix: correct alb metrics to use gauge promql functions

* fix: correct sli lib references

* feat: add ci item to burn rate alerts and add yaxis to alb graphs

* fix: correction to saturation override naming

monitoring-as-code/src/alerts/burn-rate-alerts.libsonnet

@@ -136,6 +136,8 @@ local createBurnRateAlerts(config, sliSpec, sliKey, journeyKey) =
         },
         labels: {
           ci_type: 'CMDB_CI_Service_Auto',
+          configuration_item: config.configurationItem,
+          primary_impact_service: config.applicationServiceName,
           title: alertTitle,
           wait_for: '%(for)s' % errorBudgetBurnWindow,
           factor: std.toString(errorBudgetBurnWindow.factor),

monitoring-as-code/src/metric-types.libsonnet

@@ -283,7 +283,7 @@
     },
     sliTypesConfig: {
       availability: {
-        library: (import 'sli-value-libraries/availability-counter-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
+        library: (import 'sli-value-libraries/availability-gauge-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
         description: 'The error rate for %(sliDescription)s should be below %(metric_target_percent)0.1f%%',
         targetMetrics: {
           code4xx: 'count4xx',
@@ -530,7 +530,7 @@
     },
     sliTypesConfig: {
       availability: {
-        library: (import 'sli-value-libraries/availability-counter-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
+        library: (import 'sli-value-libraries/availability-gauge-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
         description: 'The error rate for %(sliDescription)s should be below %(metric_target_percent)0.1f%%',
         targetMetrics: {
           code4xx: 'sum4xx',
@@ -567,7 +567,7 @@
     },
     sliTypesConfig: {
       availability: {
-        library: (import 'sli-value-libraries/availability-counter-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
+        library: (import 'sli-value-libraries/availability-gauge-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet'),
         description: 'The error rate for %(sliDescription)s should be below %(metric_target_percent)0.1f%%',
         targetMetrics: {
           code4xx: 'sum4xx',

monitoring-as-code/src/sli-value-libraries/availability-counter-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet → monitoring-as-code/src/sli-value-libraries/availability-gauge-using-2-failure-metrics-for-num-and-1-total-metric-for-dem.libsonnet

@@ -32,20 +32,16 @@ local createSliValueRule(sliSpec, sliMetadata, config) =
         sum without (%(selectorLabels)s) (label_replace(label_replace(
           (
             (
-              sum by(%(selectorLabels)s) (
-                rate(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s])
-                or
-                0 * %(codeAllMetric)s{%(selectors)s}
-              )
-              + 
-              sum by(%(selectorLabels)s) (
-                rate(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s])
-                or
-                0 * %(codeAllMetric)s{%(selectors)s}
-              )
-            )
+            sum by(%(selectorLabels)s) (avg_over_time(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s])
+            or 0 * %(codeAllMetric)s{%(selectors)s})
+            +
+            sum by(%(selectorLabels)s) (avg_over_time(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s])
+            or 0 * %(codeAllMetric)s{%(selectors)s})
+            >=0)
             /
-            sum by(%(selectorLabels)s) (rate(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]))
+            (
+            sum by(%(selectorLabels)s) (avg_over_time(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]))
+            >=0)
           ),
         "sli_environment", "$1", "%(environmentSelectorLabel)s", "(.*)"), "sli_product", "$1", "%(productSelectorLabel)s", "(.*)"))
       ||| % {
@@ -97,38 +93,40 @@ local createGraphPanel(sliSpec) =
   ).addTarget(
     prometheus.target(
       |||
-        sum(rate(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]) or vector(0))
+        sum(avg_over_time(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
       ||| % {
         codeAllMetric: targetMetrics.codeAll,
         selectors: std.join(',', dashboardSelectors),
         evalInterval: sliSpec.evalInterval,
       },
-      legendFormat='requests per second',
+      legendFormat='average requests',
     ),
   ).addTarget(
     prometheus.target(
       |||
-        sum(rate(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s]) or vector(0))
-        + 
-        sum(rate(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s]) or vector(0))
+        sum(avg_over_time(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
+        +
+        sum(avg_over_time(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
       ||| % {
         code4xxMetric: targetMetrics.code4xx,
         code5xxMetric: targetMetrics.code5xx,
         selectors: std.join(',', dashboardSelectors),
         evalInterval: sliSpec.evalInterval,
       },
-      legendFormat='errors per second',
+      legendFormat='average errors',
     )
   ).addTarget(
     prometheus.target(
       |||
         (
-          sum(rate(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s]) or vector(0))
-          + 
-          sum(rate(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s]) or vector(0))
+        sum(avg_over_time(%(code4xxMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
+        +
+        sum(avg_over_time(%(code5xxMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
         )
         /
-        sum(rate(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]))
+        (
+        sum(avg_over_time(%(codeAllMetric)s{%(selectors)s}[%(evalInterval)s]) >=0 or vector(0))
+        )
       ||| % {
         code4xxMetric: targetMetrics.code4xx,
         code5xxMetric: targetMetrics.code5xx,

monitoring-as-code/src/sli-value-libraries/saturation-gauge-with-useage-metric-and-percent-target.libsonnet

@@ -70,6 +70,17 @@ local createGraphPanel(sliSpec) =
     },
     min=0,
     fill=4,
+    formatY2='percentunit',
+    thresholds=[
+      {
+        value: sliSpec.metricTarget,
+        colorMode: 'critical',
+        op: 'gt',
+        line: true,
+        fill: false,
+        yaxis: 'right',
+      },
+    ],
   ).addTarget(
     prometheus.target(
       |||
@@ -94,12 +105,14 @@ local createGraphPanel(sliSpec) =
         selectors: std.join(',', dashboardSelectors),
         evalInterval: sliSpec.evalInterval,
       },
-      legendFormat='avg period where maximum saturation > %s percentage' % sliSpec.counterPercentTarget,
+      legendFormat='avg period where saturation > %s percent' % sliSpec.counterPercentTarget,
     )
   ).addSeriesOverride(
     {
-      alias: '/avg period where maximum saturation > %s percentage/' % sliSpec.counterPercentTarget,
+      alias: '/avg period where saturation > %s percent/' % sliSpec.counterPercentTarget,
+      yaxis: 2,
       color: 'red',
+
     },
   ).addSeriesOverride(
     {