initial commit

2026-02-13 15:44:24 +08:00
parent 66be8224f4
commit 09ce4c881a
570 changed files with 61807 additions and 0 deletions
@@ -0,0 +1,116 @@
+resource "aws_cloudwatch_metric_alarm" "Kafka-ZooKeeperRequestLatencyMsMean" {
+  alarm_name                = "${var.settings.ZooKeeperRequestLatencyMsMean.ecccode}-Kafka_${var.cluster-name}-ZooKeeperRequestLatencyMsMean"
+  comparison_operator       = var.settings.ZooKeeperRequestLatencyMsMean.comparison_operator
+  evaluation_periods        = var.settings.ZooKeeperRequestLatencyMsMean.evaluation_periods
+  metric_name               = "ZooKeeperRequestLatencyMsMean"
+  period                    = var.settings.ZooKeeperRequestLatencyMsMean.period
+  statistic                 = var.settings.ZooKeeperRequestLatencyMsMean.statistic
+  threshold                 = var.settings.ZooKeeperRequestLatencyMsMean.threshold
+  alarm_description         = "Kafka:ZooKeeperRequestLatencyMsMean"
+  namespace                 = "AWS/Kafka"
+  insufficient_data_actions = []
+  actions_enabled           = var.actions-enabled
+  alarm_actions             = [var.settings.ZooKeeperRequestLatencyMsMean.action]
+  ok_actions                = [var.settings.ZooKeeperRequestLatencyMsMean.action]
+  dimensions = {
+    "Cluster Name" = var.cluster-name
+  }
+}
+
+data "aws_msk_cluster" "msk-cluster" {
+  cluster_name = var.cluster-name
+}
+
+data "aws_msk_broker_nodes" "msk-broker" {
+  cluster_arn = data.aws_msk_cluster.msk-cluster.arn
+}
+
+resource "aws_cloudwatch_metric_alarm" "Kafka-CpuUserSystem" {
+  for_each                  = toset([for i in data.aws_msk_broker_nodes.msk-broker.node_info_list[*].broker_id : tostring(i)])
+  alarm_name                = "${var.settings.CpuUserSystem.ecccode}-Kafka_${var.cluster-name}-${each.value}-CpuUsage"
+  comparison_operator       = var.settings.CpuUserSystem.comparison_operator
+  evaluation_periods        = var.settings.CpuUserSystem.evaluation_periods
+  threshold                 = var.settings.CpuUserSystem.threshold
+  alarm_description         = "Kafka:ZooKeeperRequestLatencyMsMean"
+  insufficient_data_actions = []
+  actions_enabled           = var.actions-enabled
+  alarm_actions             = [var.settings.CpuUserSystem.action]
+  ok_actions                = [var.settings.CpuUserSystem.action]
+  metric_query {
+    id = "m1"
+    metric {
+      metric_name = "CpuUser"
+      namespace   = "AWS/Kafka"
+      period      = var.settings.CpuUserSystem.period
+      stat        = var.settings.CpuUserSystem.statistic
+      dimensions = {
+        "Cluster Name" = var.cluster-name
+        "Broker ID"    = each.value
+      }
+    }
+  }
+
+  metric_query {
+    id = "m2"
+    metric {
+      metric_name = "CpuSystem"
+      namespace   = "AWS/Kafka"
+      period      = var.settings.CpuUserSystem.period
+      stat        = var.settings.CpuUserSystem.statistic
+      dimensions = {
+        "Cluster Name" = var.cluster-name
+        "Broker ID"    = each.value
+      }
+    }
+  }
+
+  metric_query {
+    id          = "e1"
+    expression  = "m1 + m2"
+    label       = "CpuUserSystem"
+    return_data = "true"
+  }
+}
+
+resource "aws_cloudwatch_metric_alarm" "Kafka-KafkaDataLogsDiskUsed" {
+  for_each                  = toset([for i in data.aws_msk_broker_nodes.msk-broker.node_info_list[*].broker_id : tostring(i)])
+  alarm_name                = "${var.settings.KafkaDataLogsDiskUsed.ecccode}-Kafka_${var.cluster-name}-${each.value}-KafkaDataLogsDiskUsed"
+  comparison_operator       = var.settings.KafkaDataLogsDiskUsed.comparison_operator
+  evaluation_periods        = var.settings.KafkaDataLogsDiskUsed.evaluation_periods
+  metric_name               = "KafkaDataLogsDiskUsed"
+  period                    = var.settings.KafkaDataLogsDiskUsed.period
+  statistic                 = var.settings.KafkaDataLogsDiskUsed.statistic
+  threshold                 = var.settings.KafkaDataLogsDiskUsed.threshold
+  alarm_description         = "Kafka:KafkaDataLogsDiskUsed"
+  namespace                 = "AWS/Kafka"
+  insufficient_data_actions = []
+  actions_enabled           = var.actions-enabled
+  alarm_actions             = [var.settings.KafkaDataLogsDiskUsed.action]
+  ok_actions                = [var.settings.KafkaDataLogsDiskUsed.action]
+  dimensions = {
+    "Cluster Name" = var.cluster-name
+    "Broker ID"    = each.value
+  }
+}
+
+resource "aws_cloudwatch_metric_alarm" "Kafka-HeapMemoryAfterGC" {
+  for_each                  = toset([for i in data.aws_msk_broker_nodes.msk-broker.node_info_list[*].broker_id : tostring(i)])
+  alarm_name                = "${var.settings.HeapMemoryAfterGC.ecccode}-Kafka_${var.cluster-name}-${each.value}-HeapMemoryAfterGC"
+  comparison_operator       = var.settings.HeapMemoryAfterGC.comparison_operator
+  evaluation_periods        = var.settings.HeapMemoryAfterGC.evaluation_periods
+  metric_name               = "HeapMemoryAfterGC"
+  period                    = var.settings.HeapMemoryAfterGC.period
+  statistic                 = var.settings.HeapMemoryAfterGC.statistic
+  threshold                 = var.settings.HeapMemoryAfterGC.threshold
+  alarm_description         = "Kafka:HeapMemoryAfterGC"
+  namespace                 = "AWS/Kafka"
+  insufficient_data_actions = []
+  actions_enabled           = var.actions-enabled
+  alarm_actions             = [var.settings.HeapMemoryAfterGC.action]
+  ok_actions                = [var.settings.HeapMemoryAfterGC.action]
+  dimensions = {
+    "Cluster Name" = var.cluster-name
+    "Broker ID"    = each.value
+  }
+}
+