update_azure_apim_detectors

docs/severity.md

@@ -429,6 +429,7 @@
 |Azure API Management Service capacity|X|X|-|-|-|
 |Azure API Management Service duration of gateway request|X|X|-|-|-|
 |Azure API Management Service duration of backend request|X|X|-|-|-|
+|Azure API Management Service cpu percentage of gateway|X|X|-|-|-|
 
 
 ## integration_azure-app-service-plan

modules/integration_azure-api-management-service/README.md

@@ -79,6 +79,7 @@ This module creates the following SignalFx detectors which could contain one or
 |Azure API Management Service capacity|X|X|-|-|-|
 |Azure API Management Service duration of gateway request|X|X|-|-|-|
 |Azure API Management Service duration of backend request|X|X|-|-|-|
+|Azure API Management Service cpu percentage of gateway|X|X|-|-|-|
 
 ## How to collect required metrics?
 
@@ -98,6 +99,7 @@ Here is the list of required metrics for detectors in this module.
 
 * `BackendDuration`
 * `Capacity`
+* `CpuPercent_Gateway`
 * `Duration`
 * `NetworkConnectivity`
 

modules/integration_azure-api-management-service/conf/04-cpu-percent-gateway.yaml

@@ -0,0 +1,21 @@
+---
+module: "Azure API Management Service"
+name: "CPU Percentage of Gateway"
+filtering: "filter('resource_type', 'Microsoft.ApiManagement/service') and filter('primary_aggregation_type', 'true')"
+aggregation: ".mean(by=['azure_resource_name', 'azure_resource_group_name', 'azure_region'])"
+value_unit: "%"
+transformation: true
+signals:
+  signal:
+    metric: "CpuPercent_Gateway"
+rules:
+  critical:
+    threshold: 95
+    comparator: ">"
+    lasting_duration: '5m'
+  major:
+    threshold: 90
+    comparator: ">"
+    lasting_duration: '5m'
+    dependency: critical
+...

modules/integration_azure-api-management-service/detectors-gen.tf

@@ -165,3 +165,49 @@ EOF
   max_delay = var.duration_of_backend_request_max_delay
 }
 
+resource "signalfx_detector" "cpu_percentage_of_gateway" {
+  name = format("%s %s", local.detector_name_prefix, "Azure API Management Service cpu percentage of gateway")
+
+  authorized_writer_teams = var.authorized_writer_teams
+  teams                   = try(coalescelist(var.teams, var.authorized_writer_teams), null)
+  tags                    = compact(concat(local.common_tags, local.tags, var.extra_tags))
+
+  viz_options {
+    label        = "signal"
+    value_suffix = "%"
+  }
+
+  program_text = <<-EOF
+    base_filtering = filter('resource_type', 'Microsoft.ApiManagement/service') and filter('primary_aggregation_type', 'true')
+    signal = data('CpuPercent_Gateway', filter=base_filtering and ${module.filtering.signalflow})${var.cpu_percentage_of_gateway_aggregation_function}${var.cpu_percentage_of_gateway_transformation_function}.publish('signal')
+    detect(when(signal > ${var.cpu_percentage_of_gateway_threshold_critical}%{if var.cpu_percentage_of_gateway_lasting_duration_critical != null}, lasting='${var.cpu_percentage_of_gateway_lasting_duration_critical}', at_least=${var.cpu_percentage_of_gateway_at_least_percentage_critical}%{endif})).publish('CRIT')
+    detect(when(signal > ${var.cpu_percentage_of_gateway_threshold_major}%{if var.cpu_percentage_of_gateway_lasting_duration_major != null}, lasting='${var.cpu_percentage_of_gateway_lasting_duration_major}', at_least=${var.cpu_percentage_of_gateway_at_least_percentage_major}%{endif}) and (not when(signal > ${var.cpu_percentage_of_gateway_threshold_critical}%{if var.cpu_percentage_of_gateway_lasting_duration_critical != null}, lasting='${var.cpu_percentage_of_gateway_lasting_duration_critical}', at_least=${var.cpu_percentage_of_gateway_at_least_percentage_critical}%{endif}))).publish('MAJOR')
+EOF
+
+  rule {
+    description           = "is too high > ${var.cpu_percentage_of_gateway_threshold_critical}%"
+    severity              = "Critical"
+    detect_label          = "CRIT"
+    disabled              = coalesce(var.cpu_percentage_of_gateway_disabled_critical, var.cpu_percentage_of_gateway_disabled, var.detectors_disabled)
+    notifications         = try(coalescelist(lookup(var.cpu_percentage_of_gateway_notifications, "critical", []), var.notifications.critical), null)
+    runbook_url           = try(coalesce(var.cpu_percentage_of_gateway_runbook_url, var.runbook_url), "")
+    tip                   = var.cpu_percentage_of_gateway_tip
+    parameterized_subject = var.message_subject == "" ? local.rule_subject : var.message_subject
+    parameterized_body    = var.message_body == "" ? local.rule_body : var.message_body
+  }
+
+  rule {
+    description           = "is too high > ${var.cpu_percentage_of_gateway_threshold_major}%"
+    severity              = "Major"
+    detect_label          = "MAJOR"
+    disabled              = coalesce(var.cpu_percentage_of_gateway_disabled_major, var.cpu_percentage_of_gateway_disabled, var.detectors_disabled)
+    notifications         = try(coalescelist(lookup(var.cpu_percentage_of_gateway_notifications, "major", []), var.notifications.major), null)
+    runbook_url           = try(coalesce(var.cpu_percentage_of_gateway_runbook_url, var.runbook_url), "")
+    tip                   = var.cpu_percentage_of_gateway_tip
+    parameterized_subject = var.message_subject == "" ? local.rule_subject : var.message_subject
+    parameterized_body    = var.message_body == "" ? local.rule_body : var.message_body
+  }
+
+  max_delay = var.cpu_percentage_of_gateway_max_delay
+}
+

modules/integration_azure-api-management-service/outputs.tf

@@ -3,6 +3,11 @@ output "capacity" {
   value       = signalfx_detector.capacity
 }
 
+output "cpu_percentage_of_gateway" {
+  description = "Detector resource for cpu_percentage_of_gateway"
+  value       = signalfx_detector.cpu_percentage_of_gateway
+}
+
 output "duration_of_backend_request" {
   description = "Detector resource for duration_of_backend_request"
   value       = signalfx_detector.duration_of_backend_request

modules/integration_azure-api-management-service/variables-gen.tf

@@ -318,3 +318,93 @@ variable "duration_of_backend_request_at_least_percentage_major" {
   type        = number
   default     = 1
 }
+# cpu_percentage_of_gateway detector
+
+variable "cpu_percentage_of_gateway_notifications" {
+  description = "Notification recipients list per severity overridden for cpu_percentage_of_gateway detector"
+  type        = map(list(string))
+  default     = {}
+}
+
+variable "cpu_percentage_of_gateway_aggregation_function" {
+  description = "Aggregation function and group by for cpu_percentage_of_gateway detector (i.e. \".mean(by=['host'])\")"
+  type        = string
+  default     = ".mean(by=['azure_resource_name', 'azure_resource_group_name', 'azure_region'])"
+}
+
+variable "cpu_percentage_of_gateway_transformation_function" {
+  description = "Transformation function for cpu_percentage_of_gateway detector (i.e. \".mean(over='5m')\")"
+  type        = string
+  default     = ""
+}
+
+variable "cpu_percentage_of_gateway_max_delay" {
+  description = "Enforce max delay for cpu_percentage_of_gateway detector (use \"0\" or \"null\" for \"Auto\")"
+  type        = number
+  default     = null
+}
+
+variable "cpu_percentage_of_gateway_tip" {
+  description = "Suggested first course of action or any note useful for incident handling"
+  type        = string
+  default     = ""
+}
+
+variable "cpu_percentage_of_gateway_runbook_url" {
+  description = "URL like SignalFx dashboard or wiki page which can help to troubleshoot the incident cause"
+  type        = string
+  default     = ""
+}
+
+variable "cpu_percentage_of_gateway_disabled" {
+  description = "Disable all alerting rules for cpu_percentage_of_gateway detector"
+  type        = bool
+  default     = null
+}
+
+variable "cpu_percentage_of_gateway_disabled_critical" {
+  description = "Disable critical alerting rule for cpu_percentage_of_gateway detector"
+  type        = bool
+  default     = null
+}
+
+variable "cpu_percentage_of_gateway_disabled_major" {
+  description = "Disable major alerting rule for cpu_percentage_of_gateway detector"
+  type        = bool
+  default     = null
+}
+
+variable "cpu_percentage_of_gateway_threshold_critical" {
+  description = "Critical threshold for cpu_percentage_of_gateway detector in %"
+  type        = number
+  default     = 95
+}
+
+variable "cpu_percentage_of_gateway_lasting_duration_critical" {
+  description = "Minimum duration that conditions must be true before raising alert"
+  type        = string
+  default     = "5m"
+}
+
+variable "cpu_percentage_of_gateway_at_least_percentage_critical" {
+  description = "Percentage of lasting that conditions must be true before raising alert (>= 0.0 and <= 1.0)"
+  type        = number
+  default     = 1
+}
+variable "cpu_percentage_of_gateway_threshold_major" {
+  description = "Major threshold for cpu_percentage_of_gateway detector in %"
+  type        = number
+  default     = 90
+}
+
+variable "cpu_percentage_of_gateway_lasting_duration_major" {
+  description = "Minimum duration that conditions must be true before raising alert"
+  type        = string
+  default     = "5m"
+}
+
+variable "cpu_percentage_of_gateway_at_least_percentage_major" {
+  description = "Percentage of lasting that conditions must be true before raising alert (>= 0.0 and <= 1.0)"
+  type        = number
+  default     = 1
+}