@cdklabs/multi-az-observability 0.0.1-alpha.3 → 0.0.1-alpha.30

package/.jsii

@@ -3887,7 +3887,7 @@
     "stability": "experimental"
   },
   "homepage": "https://github.com/cdklabs/cdk-multi-az-observability",
-  "jsiiVersion": "5.5.21 (build db74302)",
+  "jsiiVersion": "5.5.25 (build c24bfd2)",
   "keywords": [
     "aws-cdk",
     "cdk",
@@ -3908,7 +3908,7 @@
   },
   "name": "@cdklabs/multi-az-observability",
   "readme": {
-    "markdown": "# multi-az-observability\nThis is a CDK construct for multi-AZ observability to help detect single-AZ impairments. This is currently an `alpha` version, but is being used in the AWS [Advanced Multi-AZ Resilience Patterns](https://catalog.workshops.aws/multi-az-gray-failures/en-US) workshop.\n\nThere is a lot of available information to think through and combine to provide signals about single-AZ impact. To simplify the setup and use reasonable defaults, this construct (available in TypeScript, Go, Python, and .NET [Java coming soon]) sets up the necessary observability. To use the CDK construct, you first define your service like this:\n\n```csharp\nvar wildRydesService = new Service(new ServiceProps(){\n    ServiceName = \"WildRydes\",\n    BaseUrl = \"http://www.example.com\",\n    FaultCountThreshold = 25,\n    AvailabilityZoneNames = vpc.AvailabilityZones,\n    Period = Duration.Seconds(60),\n    LoadBalancer = loadBalancer,\n    DefaultAvailabilityMetricDetails = new ServiceMetricDetails(new ServiceMetricDetailsProps() {\n        AlarmStatistic = \"Sum\",\n        DatapointsToAlarm = 3,\n        EvaluationPeriods = 5,\n        FaultAlarmThreshold = 1,\n        FaultMetricNames = new string[] { \"Fault\", \"Error\" },\n        GraphedFaultStatistics = new string[] { \"Sum\" },\n        GraphedSuccessStatistics = new string[] { \"Sum\" },\n        MetricNamespace = metricsNamespace,\n        Period = Duration.Seconds(60),\n        SuccessAlarmThreshold = 99,\n        SuccessMetricNames = new string[] {\"Success\"},\n        Unit = Unit.COUNT,\n    }),\n    DefaultLatencyMetricDetails = new ServiceMetricDetails(new ServiceMetricDetailsProps(){\n        AlarmStatistic = \"p99\",\n        DatapointsToAlarm = 3,\n        EvaluationPeriods = 5,\n        FaultAlarmThreshold = 1,\n        FaultMetricNames = new string[] { \"FaultLatency\" },\n        GraphedFaultStatistics = new string[] { \"p50\" },\n        GraphedSuccessStatistics = new string[] { \"p50\", \"p99\", \"tm50\", \"tm99\" },\n        MetricNamespace = metricsNamespace,\n        Period = Duration.Seconds(60),\n        SuccessAlarmThreshold = 100,\n        SuccessMetricNames = new string[] {\"SuccessLatency\"},\n        Unit = Unit.MILLISECONDS,\n    }),\n    DefaultContributorInsightRuleDetails =  new ContributorInsightRuleDetails(new ContributorInsightRuleDetailsProps() {\n        AvailabilityZoneIdJsonPath = azIdJsonPath,\n        FaultMetricJsonPath = faultMetricJsonPath,\n        InstanceIdJsonPath = instanceIdJsonPath,\n        LogGroups = serverLogGroups,\n        OperationNameJsonPath = operationNameJsonPath,\n        SuccessLatencyMetricJsonPath = successLatencyMetricJsonPath\n    }),\n    CanaryTestProps = new AddCanaryTestProps() {\n        RequestCount = 10,\n        LoadBalancer = loadBalancer,\n        Schedule = \"rate(1 minute)\",\n        NetworkConfiguration = new NetworkConfigurationProps() {\n            Vpc = vpc,\n            SubnetSelection = new SubnetSelection() { SubnetType = SubnetType.PRIVATE_ISOLATED }\n        }\n    }\n});\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Signin\",\n    Path = \"/signin\",\n    Service = wildRydesService,\n    Critical = true,\n    HttpMethods = new string[] { \"GET\" },\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Signin\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Signin\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Signin\",\n        SuccessAlarmThreshold = 150,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Signin\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 250\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Pay\",\n    Path = \"/pay\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Pay\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Pay\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Pay\",\n        SuccessAlarmThreshold = 200,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Pay\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 300\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Ride\",\n    Path = \"/ride\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Ride\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Ride\",\n        SuccessAlarmThreshold = 350,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 550\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Home\",\n    Path = \"/home\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Home\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Home\",\n        SuccessAlarmThreshold = 100,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 200\n    })\n}));\n```\n\nThen you provide that service definition to the CDK construct.\n\n```csharp\nInstrumentedServiceMultiAZObservability multiAvailabilityZoneObservability = new InstrumentedServiceMultiAZObservability(this, \"MultiAZObservability\", new InstrumentedServiceMultiAZObservabilityProps() {\n    Service = wildRydesService,\n    CreateDashboards = true,\n    Interval = Duration.Minutes(60), // The interval for the dashboard\n    OutlierDetectionAlgorithm = OutlierDetectionAlgorithm.STATIC\n});\n```\n\nYou define some characteristics of the service, default values for metrics and alarms, and then add operations as well as any overrides for default values that you need. The construct can also automatically create synthetic canaries that test each operation with a very simple HTTP check, or you can configure your own synthetics and just tell the construct about the metric details and optionally log files. This creates metrics, alarms, and dashboards that can be used to detect single-AZ impact.\n\nIf you don't have service specific logs and custom metrics with per-AZ dimensions, you can still use the construct to evaluate ALB and NAT Gateway metrics to find single AZ faults.\n\n```csharp\nBasicServiceMultiAZObservability multiAvailabilityZoneObservability = new BasicServiceMultiAZObservability(this, \"MultiAZObservability\", new BasicServiceMultiAZObservabilityProps() {\n    ApplicationLoadBalancers = new IApplicationLoadBalancer[] { loadBalancer },\n    NatGateways = new Dictionary<string, CfnNatGateway>() {\n        { \"us-east-1a\", natGateway1},\n        { \"us-east-1b\", natGateway2},\n        { \"us-east-1c\", natGateway3},\n    },\n    CreateDashboard = true,\n    OutlierDetectionAlgorithm = OutlierDetectionAlgorithm.STATIC,\n    FaultCountPercentageThreshold = 1.0, // The fault rate to alarm on for errors seen from the ALBs in the same AZ\n    PacketLossImpactPercentageThreshold = 0.01, // The percentage of packet loss to alarm on for the NAT Gateways in the same AZ\n    ServiceName = \"WildRydes\",\n    Period = Duration.Seconds(60), // The period for metric evaluation\n    Interval = Duration.Minutes(60) // The interval for the dashboards\n    EvaluationPeriods = 5,\n    DatapointsToAlarm = 3\n});\n```\n\nIf you provide a load balancer, the construct assumes it is deployed in each AZ of the VPC the load balancer is associated with and will look for HTTP metrics using those AZs as dimensions.\n\nBoth options support running workloads on EC2, ECS, Lambda, and EKS.\n"
+    "markdown": "![Build Workflow](https://github.com/cdklabs/cdk-multi-az-observability/actions/workflows/build.yml/badge.svg) ![Release Workflow](https://github.com/cdklabs/cdk-multi-az-observability/actions/workflows/release.yml/badge.svg)\n\n# multi-az-observability\nThis is a CDK construct for multi-AZ observability to help detect single-AZ impairments. This is currently an `alpha` version, but is being used in the AWS [Advanced Multi-AZ Resilience Patterns](https://catalog.workshops.aws/multi-az-gray-failures/en-US) workshop.\n\nThere is a lot of available information to think through and combine to provide signals about single-AZ impact. To simplify the setup and use reasonable defaults, this construct (available in TypeScript, Go, Python, and .NET [Java coming soon]) sets up the necessary observability. To use the CDK construct, you first define your service like this:\n\n```csharp\nvar wildRydesService = new Service(new ServiceProps(){\n    ServiceName = \"WildRydes\",\n    BaseUrl = \"http://www.example.com\",\n    FaultCountThreshold = 25,\n    AvailabilityZoneNames = vpc.AvailabilityZones,\n    Period = Duration.Seconds(60),\n    LoadBalancer = loadBalancer,\n    DefaultAvailabilityMetricDetails = new ServiceMetricDetails(new ServiceMetricDetailsProps() {\n        AlarmStatistic = \"Sum\",\n        DatapointsToAlarm = 3,\n        EvaluationPeriods = 5,\n        FaultAlarmThreshold = 1,\n        FaultMetricNames = new string[] { \"Fault\", \"Error\" },\n        GraphedFaultStatistics = new string[] { \"Sum\" },\n        GraphedSuccessStatistics = new string[] { \"Sum\" },\n        MetricNamespace = metricsNamespace,\n        Period = Duration.Seconds(60),\n        SuccessAlarmThreshold = 99,\n        SuccessMetricNames = new string[] {\"Success\"},\n        Unit = Unit.COUNT,\n    }),\n    DefaultLatencyMetricDetails = new ServiceMetricDetails(new ServiceMetricDetailsProps(){\n        AlarmStatistic = \"p99\",\n        DatapointsToAlarm = 3,\n        EvaluationPeriods = 5,\n        FaultAlarmThreshold = 1,\n        FaultMetricNames = new string[] { \"FaultLatency\" },\n        GraphedFaultStatistics = new string[] { \"p50\" },\n        GraphedSuccessStatistics = new string[] { \"p50\", \"p99\", \"tm50\", \"tm99\" },\n        MetricNamespace = metricsNamespace,\n        Period = Duration.Seconds(60),\n        SuccessAlarmThreshold = 100,\n        SuccessMetricNames = new string[] {\"SuccessLatency\"},\n        Unit = Unit.MILLISECONDS,\n    }),\n    DefaultContributorInsightRuleDetails =  new ContributorInsightRuleDetails(new ContributorInsightRuleDetailsProps() {\n        AvailabilityZoneIdJsonPath = azIdJsonPath,\n        FaultMetricJsonPath = faultMetricJsonPath,\n        InstanceIdJsonPath = instanceIdJsonPath,\n        LogGroups = serverLogGroups,\n        OperationNameJsonPath = operationNameJsonPath,\n        SuccessLatencyMetricJsonPath = successLatencyMetricJsonPath\n    }),\n    CanaryTestProps = new AddCanaryTestProps() {\n        RequestCount = 10,\n        LoadBalancer = loadBalancer,\n        Schedule = \"rate(1 minute)\",\n        NetworkConfiguration = new NetworkConfigurationProps() {\n            Vpc = vpc,\n            SubnetSelection = new SubnetSelection() { SubnetType = SubnetType.PRIVATE_ISOLATED }\n        }\n    }\n});\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Signin\",\n    Path = \"/signin\",\n    Service = wildRydesService,\n    Critical = true,\n    HttpMethods = new string[] { \"GET\" },\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Signin\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Signin\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Signin\",\n        SuccessAlarmThreshold = 150,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Signin\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 250\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Pay\",\n    Path = \"/pay\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Pay\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Pay\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Pay\",\n        SuccessAlarmThreshold = 200,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Pay\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 300\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Ride\",\n    Path = \"/ride\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Ride\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Ride\",\n        SuccessAlarmThreshold = 350,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 550\n    })\n}));\nwildRydesService.AddOperation(new Operation(new OperationProps() {\n    OperationName = \"Home\",\n    Path = \"/home\",\n    Service = wildRydesService,\n    HttpMethods = new string[] { \"GET\" },\n    Critical = true,\n    ServerSideAvailabilityMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Home\",\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultAvailabilityMetricDetails),\n    ServerSideLatencyMetricDetails = new OperationMetricDetails(new OperationMetricDetailsProps() {\n        OperationName = \"Home\",\n        SuccessAlarmThreshold = 100,\n        MetricDimensions = new MetricDimensions(new Dictionary<string, string> {{ \"Operation\", \"Ride\"}}, \"AZ-ID\", \"Region\")\n    }, wildRydesService.DefaultLatencyMetricDetails),\n    CanaryTestLatencyMetricsOverride = new CanaryTestMetricsOverride(new CanaryTestMetricsOverrideProps() {\n        SuccessAlarmThreshold = 200\n    })\n}));\n```\n\nThen you provide that service definition to the CDK construct.\n\n```csharp\nInstrumentedServiceMultiAZObservability multiAvailabilityZoneObservability = new InstrumentedServiceMultiAZObservability(this, \"MultiAZObservability\", new InstrumentedServiceMultiAZObservabilityProps() {\n    Service = wildRydesService,\n    CreateDashboards = true,\n    Interval = Duration.Minutes(60), // The interval for the dashboard\n    OutlierDetectionAlgorithm = OutlierDetectionAlgorithm.STATIC\n});\n```\n\nYou define some characteristics of the service, default values for metrics and alarms, and then add operations as well as any overrides for default values that you need. The construct can also automatically create synthetic canaries that test each operation with a very simple HTTP check, or you can configure your own synthetics and just tell the construct about the metric details and optionally log files. This creates metrics, alarms, and dashboards that can be used to detect single-AZ impact.\n\nIf you don't have service specific logs and custom metrics with per-AZ dimensions, you can still use the construct to evaluate ALB and NAT Gateway metrics to find single AZ faults.\n\n```csharp\nBasicServiceMultiAZObservability multiAZObservability = new BasicServiceMultiAZObservability(this, \"basic-service-\", new BasicServiceMultiAZObservabilityProps() {\n    ApplicationLoadBalancerProps = new ApplicationLoadBalancerDetectionProps() {\n        ApplicationLoadBalancers = [ myALB ],\n        LatencyStatistic = Stats.Percentile(99),\n        FaultCountPercentThreshold = 1,\n        LatencyThreshold = 500\n    },\n    NatGatewayProps = new NatGatewayDetectionProps() {\n        PacketLossPercentThreshold = 0.01,\n        NatGateways = {\n           { \"us-east-1a\", [ natGateway1 ] },\n           { \"us-east-1b\", [ natGateway2 ] },\n           { \"us-east-1c\", [ natGateway3 ] }\n        },\n    },\n    CreateDashboard = true,\n    DatapointsToAlarm = 2,\n    EvaluationPeriods = 3,\n    ServiceName = \"WildRydes\",\n    Period = Duration.Seconds(60),\n    Interval = Duration.Minutes(60),\n});\n```\n\nIf you provide a load balancer, the construct assumes it is deployed in each AZ of the VPC the load balancer is associated with and will look for HTTP metrics using those AZs as dimensions.\n\nBoth options support running workloads on EC2, ECS, Lambda, and EKS."
   },
   "repository": {
     "type": "git",
@@ -4142,6 +4142,223 @@
       ],
       "symbolId": "src/canaries/props/AddCanaryTestProps:AddCanaryTestProps"
     },
+    "@cdklabs/multi-az-observability.ApplicationLoadBalancerAvailabilityOutlierAlgorithm": {
+      "assembly": "@cdklabs/multi-az-observability",
+      "docs": {
+        "stability": "experimental",
+        "summary": "The options for calculating if an ALB is an outlier for availability."
+      },
+      "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerAvailabilityOutlierAlgorithm",
+      "kind": "enum",
+      "locationInModule": {
+        "filename": "src/outlier-detection/ApplicationLoadBalancerAvailabilityOutlierAlgorithm.ts",
+        "line": 4
+      },
+      "members": [
+        {
+          "docs": {
+            "stability": "experimental",
+            "summary": "This will take the availability threshold and calculate if one AZ is responsible for that percentage of errors."
+          },
+          "name": "STATIC"
+        }
+      ],
+      "name": "ApplicationLoadBalancerAvailabilityOutlierAlgorithm",
+      "symbolId": "src/outlier-detection/ApplicationLoadBalancerAvailabilityOutlierAlgorithm:ApplicationLoadBalancerAvailabilityOutlierAlgorithm"
+    },
+    "@cdklabs/multi-az-observability.ApplicationLoadBalancerDetectionProps": {
+      "assembly": "@cdklabs/multi-az-observability",
+      "datatype": true,
+      "docs": {
+        "stability": "experimental",
+        "summary": "The properties for performing zonal impact detection with ALB(s)."
+      },
+      "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerDetectionProps",
+      "kind": "interface",
+      "locationInModule": {
+        "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+        "line": 8
+      },
+      "name": "ApplicationLoadBalancerDetectionProps",
+      "properties": [
+        {
+          "abstract": true,
+          "docs": {
+            "stability": "experimental",
+            "summary": "The application load balancers to collect metrics from."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 13
+          },
+          "name": "applicationLoadBalancers",
+          "type": {
+            "collection": {
+              "elementtype": {
+                "fqn": "aws-cdk-lib.aws_elasticloadbalancingv2.IApplicationLoadBalancer"
+              },
+              "kind": "array"
+            }
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "remarks": "This should align with your availability goal. For example\n1% or 5%, provided as 1 or 5.",
+            "stability": "experimental",
+            "summary": "The percentage of faults for a single ALB to consider an AZ to be unhealthy, a number between 0 and 100."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 20
+          },
+          "name": "faultCountPercentThreshold",
+          "type": {
+            "primitive": "number"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "stability": "experimental",
+            "summary": "The statistic used to measure target response latency, like p99,  which can be specified using Stats.percentile(99) or \"p99\"."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 32
+          },
+          "name": "latencyStatistic",
+          "type": {
+            "primitive": "string"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "stability": "experimental",
+            "summary": "The threshold in milliseconds for ALB targets whose responses are slower than this value at the specified percentile statistic."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 26
+          },
+          "name": "latencyThreshold",
+          "type": {
+            "primitive": "number"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "STATIC",
+            "stability": "experimental",
+            "summary": "The method used to determine if an AZ is an outlier for availability for Application Load Balancer metrics."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 44
+          },
+          "name": "availabilityOutlierAlgorithm",
+          "optional": true,
+          "type": {
+            "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerAvailabilityOutlierAlgorithm"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "\"This depends on the algorithm used. STATIC: 66\"",
+            "stability": "experimental",
+            "summary": "The threshold for the outlier detection algorithm."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 51
+          },
+          "name": "availabilityOutlierThreshold",
+          "optional": true,
+          "type": {
+            "primitive": "number"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "Z_SCORE",
+            "stability": "experimental",
+            "summary": "The method used to determine if an AZ is an outlier for latency for Application Load Balancer metrics."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 38
+          },
+          "name": "latencyOutlierAlgorithm",
+          "optional": true,
+          "type": {
+            "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerLatencyOutlierAlgorithm"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "\"This depends on the algorithm used. STATIC: 66. Z_SCORE: 3.\"",
+            "stability": "experimental",
+            "summary": "The threshold for the outlier detection algorithm."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps.ts",
+            "line": 58
+          },
+          "name": "latencyOutlierThreshold",
+          "optional": true,
+          "type": {
+            "primitive": "number"
+          }
+        }
+      ],
+      "symbolId": "src/basic_observability/props/ApplicationLoadBalancerDetectionProps:ApplicationLoadBalancerDetectionProps"
+    },
+    "@cdklabs/multi-az-observability.ApplicationLoadBalancerLatencyOutlierAlgorithm": {
+      "assembly": "@cdklabs/multi-az-observability",
+      "docs": {
+        "stability": "experimental",
+        "summary": "The options for calculating if an AZ is an outlier for latency for ALBs."
+      },
+      "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerLatencyOutlierAlgorithm",
+      "kind": "enum",
+      "locationInModule": {
+        "filename": "src/outlier-detection/ApplicationLoadBalancerLatencyOutlierAlgorithm.ts",
+        "line": 4
+      },
+      "members": [
+        {
+          "docs": {
+            "remarks": "This provides a static comparison\nof the number of high latency requests in one AZ versus the others",
+            "stability": "experimental",
+            "summary": "This will take the latency threshold and count the number of requests per AZ  that exceed this threshold and then calculate the percentage of requests exceeding this threshold belong to each AZ."
+          },
+          "name": "STATIC"
+        },
+        {
+          "docs": {
+            "remarks": "It uses\nthe target response time of all requests to calculate the standard deviation and\naverage for all AZs. This is the default.",
+            "stability": "experimental",
+            "summary": "This calculates the z score of latency in one AZ against the other AZs."
+          },
+          "name": "Z_SCORE"
+        }
+      ],
+      "name": "ApplicationLoadBalancerLatencyOutlierAlgorithm",
+      "symbolId": "src/outlier-detection/ApplicationLoadBalancerLatencyOutlierAlgorithm:ApplicationLoadBalancerLatencyOutlierAlgorithm"
+    },
     "@cdklabs/multi-az-observability.AvailabilityZoneMapper": {
       "assembly": "@cdklabs/multi-az-observability",
       "base": "constructs.Construct",
@@ -4515,7 +4732,7 @@
         },
         "locationInModule": {
           "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-          "line": 86
+          "line": 81
         },
         "parameters": [
           {
@@ -4544,7 +4761,7 @@
       "kind": "class",
       "locationInModule": {
         "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-        "line": 39
+        "line": 34
       },
       "name": "BasicServiceMultiAZObservability",
       "properties": [
@@ -4555,7 +4772,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 74
+            "line": 69
           },
           "name": "aggregateZonalIsolatedImpactAlarms",
           "overrides": "@cdklabs/multi-az-observability.IBasicServiceMultiAZObservability",
@@ -4575,7 +4792,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 56
+            "line": 51
           },
           "name": "serviceName",
           "overrides": "@cdklabs/multi-az-observability.IBasicServiceMultiAZObservability",
@@ -4590,7 +4807,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 68
+            "line": 63
           },
           "name": "albZonalIsolatedImpactAlarms",
           "optional": true,
@@ -4611,7 +4828,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 51
+            "line": 46
           },
           "name": "applicationLoadBalancers",
           "optional": true,
@@ -4632,7 +4849,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 79
+            "line": 74
           },
           "name": "dashboard",
           "optional": true,
@@ -4647,7 +4864,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 46
+            "line": 41
           },
           "name": "natGateways",
           "optional": true,
@@ -4673,7 +4890,7 @@
           },
           "locationInModule": {
             "filename": "src/basic_observability/BasicServiceMultiAZObservability.ts",
-            "line": 62
+            "line": 57
           },
           "name": "natGWZonalIsolatedImpactAlarms",
           "optional": true,
@@ -4714,7 +4931,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 123
+            "line": 93
           },
           "name": "datapointsToAlarm",
           "type": {
@@ -4730,62 +4947,13 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 118
+            "line": 88
           },
           "name": "evaluationPeriods",
           "type": {
             "primitive": "number"
           }
         },
-        {
-          "abstract": true,
-          "docs": {
-            "remarks": "For example\n1% or 5%, specify as 1 or 5.",
-            "stability": "experimental",
-            "summary": "The percentage of faults for a single ALB to consider an AZ to be unhealthy, this should align with your availability goal."
-          },
-          "immutable": true,
-          "locationInModule": {
-            "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 49
-          },
-          "name": "faultCountPercentageThreshold",
-          "type": {
-            "primitive": "number"
-          }
-        },
-        {
-          "abstract": true,
-          "docs": {
-            "stability": "experimental",
-            "summary": "The statistic used to measure target response latency, like p99,  which can be specified using Stats.percentile(99) or \"p99\"."
-          },
-          "immutable": true,
-          "locationInModule": {
-            "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 61
-          },
-          "name": "latencyStatistic",
-          "type": {
-            "primitive": "string"
-          }
-        },
-        {
-          "abstract": true,
-          "docs": {
-            "stability": "experimental",
-            "summary": "The threshold in seconds for ALB targets whose responses are slower than this value at the specified percentile statistic."
-          },
-          "immutable": true,
-          "locationInModule": {
-            "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 55
-          },
-          "name": "latencyThreshold",
-          "type": {
-            "primitive": "number"
-          }
-        },
         {
           "abstract": true,
           "docs": {
@@ -4795,7 +4963,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 34
+            "line": 31
           },
           "name": "serviceName",
           "type": {
@@ -4805,25 +4973,20 @@
         {
           "abstract": true,
           "docs": {
-            "default": "\"No alarms for ALBs will be created\"",
-            "remarks": "There will be an alarm created for \neach AZ for each ALB. Then, there will be a composite alarm for AZ created from the input\nof all ALBs. You must either specify an ALB or a NAT GW.",
+            "default": "\"No ALBs will be used to calculate impact.\"",
+            "remarks": "You must specify this\nand/or natGatewayProps.",
             "stability": "experimental",
-            "summary": "The application load balancers being used by the service."
+            "summary": "Properties for ALBs to detect single AZ impact."
           },
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 29
+            "line": 26
           },
-          "name": "applicationLoadBalancers",
+          "name": "applicationLoadBalancerProps",
           "optional": true,
           "type": {
-            "collection": {
-              "elementtype": {
-                "fqn": "aws-cdk-lib.aws_elasticloadbalancingv2.IApplicationLoadBalancer"
-              },
-              "kind": "array"
-            }
+            "fqn": "@cdklabs/multi-az-observability.ApplicationLoadBalancerDetectionProps"
           }
         },
         {
@@ -4837,7 +5000,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 97
+            "line": 67
           },
           "name": "assetsBucketParameterName",
           "optional": true,
@@ -4856,7 +5019,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 113
+            "line": 83
           },
           "name": "assetsBucketPrefixParameterName",
           "optional": true,
@@ -4874,7 +5037,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 75
+            "line": 45
           },
           "name": "createDashboard",
           "optional": true,
@@ -4892,7 +5055,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 82
+            "line": 52
           },
           "name": "interval",
           "optional": true,
@@ -4903,48 +5066,20 @@
         {
           "abstract": true,
           "docs": {
-            "default": "\"No alarms for NAT Gateways will be created\"",
-            "remarks": "One alarm per NAT GW will be created. If multiple NAT GWs\nare provided for a single AZ, those alarms will be aggregated into\na composite alarm for the AZ. You must either specify an ALB or a NAT GW.",
-            "stability": "experimental",
-            "summary": "(Optional) A map of Availability Zone name to the NAT Gateways in that AZ."
-          },
-          "immutable": true,
-          "locationInModule": {
-            "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 20
-          },
-          "name": "natGateways",
-          "optional": true,
-          "type": {
-            "collection": {
-              "elementtype": {
-                "collection": {
-                  "elementtype": {
-                    "fqn": "aws-cdk-lib.aws_ec2.CfnNatGateway"
-                  },
-                  "kind": "array"
-                }
-              },
-              "kind": "map"
-            }
-          }
-        },
-        {
-          "abstract": true,
-          "docs": {
-            "default": "\"0.01 (as in 0.01%)\"",
+            "default": "\"No NAT Gateways will be used to calculate impact.\"",
+            "remarks": "You must specify\nthis and/or applicationLoadBalancerProps.",
             "stability": "experimental",
-            "summary": "The amount of packet loss in a NAT GW to determine if an AZ is actually impacted, recommendation is 0.01%."
+            "summary": "Properties for NAT Gateways to detect single AZ impact."
           },
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 42
+            "line": 18
           },
-          "name": "packetLossImpactPercentageThreshold",
+          "name": "natGatewayProps",
           "optional": true,
           "type": {
-            "primitive": "number"
+            "fqn": "@cdklabs/multi-az-observability.NatGatewayDetectionProps"
           }
         },
         {
@@ -4957,7 +5092,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "src/basic_observability/props/BasicServiceMultiAZObservabilityProps.ts",
-            "line": 68
+            "line": 38
           },
           "name": "period",
           "optional": true,
@@ -7203,36 +7338,6 @@
       },
       "name": "IServiceAlarmsAndRules",
       "properties": [
-        {
-          "abstract": true,
-          "docs": {
-            "stability": "experimental",
-            "summary": "An alarm for regional availability or latency impact of any critical operation as measured by the server-side."
-          },
-          "locationInModule": {
-            "filename": "src/alarmsandrules/IServiceAlarmsAndRules.ts",
-            "line": 43
-          },
-          "name": "regionalAvailabilityOrLatencyServerSideAlarm",
-          "type": {
-            "fqn": "aws-cdk-lib.aws_cloudwatch.IAlarm"
-          }
-        },
-        {
-          "abstract": true,
-          "docs": {
-            "stability": "experimental",
-            "summary": "An alarm for regional availability impact of any critical operation as measured by the server-side."
-          },
-          "locationInModule": {
-            "filename": "src/alarmsandrules/IServiceAlarmsAndRules.ts",
-            "line": 48
-          },
-          "name": "regionalAvailabilityServerSideAlarm",
-          "type": {
-            "fqn": "aws-cdk-lib.aws_cloudwatch.IAlarm"
-          }
-        },
         {
           "abstract": true,
           "docs": {
@@ -7241,7 +7346,7 @@
           },
           "locationInModule": {
             "filename": "src/alarmsandrules/IServiceAlarmsAndRules.ts",
-            "line": 53
+            "line": 43
           },
           "name": "regionalFaultCountServerSideAlarm",
           "type": {
@@ -7589,7 +7694,7 @@
         },
         "locationInModule": {
           "filename": "src/services/InstrumentedServiceMultiAZObservability.ts",
-          "line": 87
+          "line": 355
         },
         "parameters": [
           {
@@ -8035,6 +8140,104 @@
       ],
       "symbolId": "src/services/props/MetricDimensions:MetricDimensions"
     },
+    "@cdklabs/multi-az-observability.NatGatewayDetectionProps": {
+      "assembly": "@cdklabs/multi-az-observability",
+      "datatype": true,
+      "docs": {
+        "stability": "experimental",
+        "summary": "The properties for performing zonal impact detection with NAT Gateway(s)."
+      },
+      "fqn": "@cdklabs/multi-az-observability.NatGatewayDetectionProps",
+      "kind": "interface",
+      "locationInModule": {
+        "filename": "src/basic_observability/props/NatGatewayDetectionProps.ts",
+        "line": 7
+      },
+      "name": "NatGatewayDetectionProps",
+      "properties": [
+        {
+          "abstract": true,
+          "docs": {
+            "stability": "experimental",
+            "summary": "A list of NAT Gateways per Availability Zone (using the AZ name as the key)."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/NatGatewayDetectionProps.ts",
+            "line": 12
+          },
+          "name": "natGateways",
+          "type": {
+            "collection": {
+              "elementtype": {
+                "collection": {
+                  "elementtype": {
+                    "fqn": "aws-cdk-lib.aws_ec2.CfnNatGateway"
+                  },
+                  "kind": "array"
+                }
+              },
+              "kind": "map"
+            }
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "PacketLossOutlierAlgorithm.STATIC",
+            "stability": "experimental",
+            "summary": "The algorithm to use to calculate an AZ as an outlier for packet loss."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/NatGatewayDetectionProps.ts",
+            "line": 26
+          },
+          "name": "packetLossOutlierAlgorithm",
+          "optional": true,
+          "type": {
+            "fqn": "@cdklabs/multi-az-observability.PacketLossOutlierAlgorithm"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "\"This depends on the outlier algorithm. STATIC: 66. Z-SCORE: 3.\"",
+            "stability": "experimental",
+            "summary": "The threshold used with the outlier calculation."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/NatGatewayDetectionProps.ts",
+            "line": 33
+          },
+          "name": "packetLossOutlierThreshold",
+          "optional": true,
+          "type": {
+            "primitive": "number"
+          }
+        },
+        {
+          "abstract": true,
+          "docs": {
+            "default": "0.01 (as in 0.01%)",
+            "stability": "experimental",
+            "summary": "The percentage of packet loss at which you consider there to be impact."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "src/basic_observability/props/NatGatewayDetectionProps.ts",
+            "line": 19
+          },
+          "name": "packetLossPercentThreshold",
+          "optional": true,
+          "type": {
+            "primitive": "number"
+          }
+        }
+      ],
+      "symbolId": "src/basic_observability/props/NatGatewayDetectionProps:NatGatewayDetectionProps"
+    },
     "@cdklabs/multi-az-observability.NetworkConfigurationProps": {
       "assembly": "@cdklabs/multi-az-observability",
       "datatype": true,
@@ -9224,6 +9427,30 @@
       "name": "OutlierDetectionAlgorithm",
       "symbolId": "src/utilities/OutlierDetectionAlgorithm:OutlierDetectionAlgorithm"
     },
+    "@cdklabs/multi-az-observability.PacketLossOutlierAlgorithm": {
+      "assembly": "@cdklabs/multi-az-observability",
+      "docs": {
+        "stability": "experimental",
+        "summary": "The options for calculating if a NAT Gateway is an outlier for packet loss."
+      },
+      "fqn": "@cdklabs/multi-az-observability.PacketLossOutlierAlgorithm",
+      "kind": "enum",
+      "locationInModule": {
+        "filename": "src/outlier-detection/PacketLossOutlierAlgorithm.ts",
+        "line": 4
+      },
+      "members": [
+        {
+          "docs": {
+            "stability": "experimental",
+            "summary": "This will take the availability threshold and calculate if one AZ is responsible for that percentage of packet loss."
+          },
+          "name": "STATIC"
+        }
+      ],
+      "name": "PacketLossOutlierAlgorithm",
+      "symbolId": "src/outlier-detection/PacketLossOutlierAlgorithm:PacketLossOutlierAlgorithm"
+    },
     "@cdklabs/multi-az-observability.Service": {
       "assembly": "@cdklabs/multi-az-observability",
       "docs": {
@@ -10160,6 +10387,6 @@
       "symbolId": "src/services/props/ServiceProps:ServiceProps"
     }
   },
-  "version": "0.0.1-alpha.3",
-  "fingerprint": "t/7gmiT5lsADp2kfhh/RTCx/Vw6I/IGnL93ALxhVEqk="
+  "version": "0.0.1-alpha.30",
+  "fingerprint": "ys9BtzAGnjsEfKYQ35UHnXiostwzMUwA0IV+FfWyLDs="
 }