Merge pull request #76 from dasmeta/DMVP-2778-kube-state-metrics-refa…

…ctoring

fix(DMVP-2778): Kube State Metrics refactoring

README.md

@@ -231,7 +231,7 @@ worker_groups = {
 | Name | Description | Type | Default | Required |
 |------|-------------|------|---------|:--------:|
 | <a name="input_account_id"></a> [account\_id](#input\_account\_id) | AWS Account Id to apply changes into | `string` | `null` | no |
-| <a name="input_adot_config"></a> [adot\_config](#input\_adot\_config) | n/a | `any` | <pre>{<br>  "accepte_namespace_regex": "(default|kube-system)",<br>  "additional_metrics": {}<br>}</pre> | no |
+| <a name="input_adot_config"></a> [adot\_config](#input\_adot\_config) | n/a | `any` | <pre>{<br>  "accept_namespace_regex": "(default|kube-system)",<br>  "additional_metrics": {},<br>  "log_group_name": "adot_log_group"<br>}</pre> | no |
 | <a name="input_adot_version"></a> [adot\_version](#input\_adot\_version) | The version of the AWS Distro for OpenTelemetry addon to use. | `string` | `"v0.78.0-eksbuild.1"` | no |
 | <a name="input_alb_log_bucket_name"></a> [alb\_log\_bucket\_name](#input\_alb\_log\_bucket\_name) | n/a | `string` | `""` | no |
 | <a name="input_alb_log_bucket_path"></a> [alb\_log\_bucket\_path](#input\_alb\_log\_bucket\_path) | ALB-INGRESS-CONTROLLER | `string` | `""` | no |
@@ -268,6 +268,7 @@ worker_groups = {
 | <a name="input_node_groups"></a> [node\_groups](#input\_node\_groups) | Map of EKS managed node group definitions to create | `any` | <pre>{<br>  "default": {<br>    "desired_size": 2,<br>    "iam_role_additional_policies": [<br>      "arn:aws:iam::aws:policy/CloudWatchAgentServerPolicy"<br>    ],<br>    "instance_types": [<br>      "t3.medium"<br>    ],<br>    "max_size": 4,<br>    "min_size": 2<br>  }<br>}</pre> | no |
 | <a name="input_node_groups_default"></a> [node\_groups\_default](#input\_node\_groups\_default) | Map of EKS managed node group default configurations | `any` | <pre>{<br>  "disk_size": 50,<br>  "iam_role_additional_policies": [<br>    "arn:aws:iam::aws:policy/CloudWatchAgentServerPolicy"<br>  ],<br>  "instance_types": [<br>    "t3.medium"<br>  ]<br>}</pre> | no |
 | <a name="input_node_security_group_additional_rules"></a> [node\_security\_group\_additional\_rules](#input\_node\_security\_group\_additional\_rules) | n/a | `any` | <pre>{<br>  "ingress_cluster_10250": {<br>    "description": "Metric server to node groups",<br>    "from_port": 10250,<br>    "protocol": "tcp",<br>    "self": true,<br>    "to_port": 10250,<br>    "type": "ingress"<br>  },<br>  "ingress_cluster_8443": {<br>    "description": "Metric server to node groups",<br>    "from_port": 8443,<br>    "protocol": "tcp",<br>    "source_cluster_security_group": true,<br>    "to_port": 8443,<br>    "type": "ingress"<br>  }<br>}</pre> | no |
+| <a name="input_prometheus_metrics"></a> [prometheus\_metrics](#input\_prometheus\_metrics) | Prometheus Metrics | `any` | `[]` | no |
 | <a name="input_region"></a> [region](#input\_region) | AWS Region name. | `string` | `null` | no |
 | <a name="input_roles"></a> [roles](#input\_roles) | Variable describes which role will user have K8s | <pre>list(object({<br>    actions   = list(string)<br>    resources = list(string)<br>  }))</pre> | `[]` | no |
 | <a name="input_scale_down_unneeded_time"></a> [scale\_down\_unneeded\_time](#input\_scale\_down\_unneeded\_time) | Scale down unneeded in minutes | `number` | `2` | no |

examples/spot-instance/README.md

@@ -12,7 +12,7 @@
 
 | Name | Version |
 |------|---------|
-| <a name="provider_aws"></a> [aws](#provider\_aws) | 4.67.0 |
+| <a name="provider_aws"></a> [aws](#provider\_aws) | >= 3.41 |
 
 ## Modules
 

main.tf

@@ -322,6 +322,7 @@ module "adot" {
   oidc_provider_arn           = module.eks-cluster[0].oidc_provider_arn
   adot_config                 = var.adot_config
   adot_version                = var.adot_version
+  prometheus_metrics          = var.prometheus_metrics
   region                      = local.region
   depends_on = [
     module.eks-cluster,
@@ -358,6 +359,10 @@ resource "helm_release" "kube-state-metrics" {
   version          = "4.22.3"
   create_namespace = false
   atomic           = true
+  set_list {
+    name  = "metricAllowlist"
+    value = var.prometheus_metrics
+  }
 }
 
 module "autoscaler" {

modules/adot/README.md

@@ -74,14 +74,14 @@ No modules.
 | Name | Description | Type | Default | Required |
 |------|-------------|------|---------|:--------:|
 | <a name="input_adot_collector_policy_arns"></a> [adot\_collector\_policy\_arns](#input\_adot\_collector\_policy\_arns) | List of IAM policy ARNs to attach to the ADOT collector service account. | `list(string)` | <pre>[<br>  "arn:aws:iam::aws:policy/AmazonPrometheusRemoteWriteAccess",<br>  "arn:aws:iam::aws:policy/CloudWatchAgentServerPolicy",<br>  "arn:aws:iam::aws:policy/AWSXrayWriteOnlyAccess"<br>]</pre> | no |
-| <a name="input_adot_config"></a> [adot\_config](#input\_adot\_config) | accept\_namespace\_regex defines the list of namespaces from which metrics will be exported, and additional\_metrics defines additional metrics to export. | `any` | <pre>{<br>  "accept_namespace_regex": "(default|kube-system)",<br>  "additional_metrics": [],<br>  "helm_values": null<br>}</pre> | no |
+| <a name="input_adot_config"></a> [adot\_config](#input\_adot\_config) | accept\_namespace\_regex defines the list of namespaces from which metrics will be exported, and additional\_metrics defines additional metrics to export. | `any` | <pre>{<br>  "accept_namespace_regex": "(default|kube-system)",<br>  "additional_metrics": [],<br>  "helm_values": null,<br>  "log_group_name": "adot_log_group"<br>}</pre> | no |
 | <a name="input_adot_log_group_name"></a> [adot\_log\_group\_name](#input\_adot\_log\_group\_name) | ADOT log group name | `string` | `"adot_log_group_name"` | no |
 | <a name="input_adot_version"></a> [adot\_version](#input\_adot\_version) | The version of the AWS Distro for OpenTelemetry addon to use. | `string` | `"v0.78.0-eksbuild.1"` | no |
 | <a name="input_cluster_name"></a> [cluster\_name](#input\_cluster\_name) | K8s cluster name. | `string` | n/a | yes |
 | <a name="input_eks_oidc_root_ca_thumbprint"></a> [eks\_oidc\_root\_ca\_thumbprint](#input\_eks\_oidc\_root\_ca\_thumbprint) | EKS oidc root ca thumbprint. | `string` | n/a | yes |
 | <a name="input_namespace"></a> [namespace](#input\_namespace) | The namespace to install the AWS Distro for OpenTelemetry addon. | `string` | `"adot"` | no |
 | <a name="input_oidc_provider_arn"></a> [oidc\_provider\_arn](#input\_oidc\_provider\_arn) | EKC oidc provider arn. | `string` | n/a | yes |
-| <a name="input_prometheus_metrics"></a> [prometheus\_metrics](#input\_prometheus\_metrics) | Prometheus Metrics | `any` | `{}` | no |
+| <a name="input_prometheus_metrics"></a> [prometheus\_metrics](#input\_prometheus\_metrics) | Prometheus Metrics | `any` | `[]` | no |
 | <a name="input_region"></a> [region](#input\_region) | AWS Region | `string` | n/a | yes |
 
 ## Outputs

modules/adot/main.tf

@@ -24,7 +24,7 @@ resource "helm_release" "adot-collector" {
       region                     = local.region
       cluster_name               = var.cluster_name
       accept_namespace_regex     = var.adot_config.accept_namespace_regex
-      log_group_name             = var.adot_log_group_name
+      log_group_name             = var.adot_config.log_group_name
       metrics                    = local.merged_metrics
       metrics_namespace_specific = local.merged_namespace_specific
       prometheus_metrics         = var.prometheus_metrics

modules/adot/templates/adot-values.yaml.tpl

@@ -133,9 +133,9 @@ adotCollector:
               - Deployment
             metric_name_selectors:
             - kube_deployment_spec_replicas
-%{ for key,value in prometheus_metrics }
-          - dimensions: ${key}
-            metric_name_selectors: ${jsonencode(value)}
+            - kube_deployment_status_replicas_available
+%{ for value in prometheus_metrics }
+            - ${value}
 %{ endfor ~}
           namespace: ContainerInsights
           parse_json_encoded_attr_values:

modules/adot/tests/template_file/README.md

@@ -12,7 +12,7 @@
 
 | Name | Version |
 |------|---------|
-| <a name="provider_helm"></a> [helm](#provider\_helm) | 2.9.0 |
+| <a name="provider_helm"></a> [helm](#provider\_helm) | n/a |
 | <a name="provider_test"></a> [test](#provider\_test) | n/a |
 
 ## Modules

modules/adot/variables.tf

@@ -41,6 +41,7 @@ variable "adot_config" {
   default = {
     accept_namespace_regex = "(default|kube-system)"
     additional_metrics     = []
+    log_group_name         = "adot_log_group"
     # ADOT helm chart values.yaml, if you don't use variable adot will be deployed with module default values file
     helm_values = null
   }
@@ -54,7 +55,7 @@ variable "region" {
 variable "prometheus_metrics" {
   description = "Prometheus Metrics"
   type        = any
-  default     = {}
+  default     = []
 }
 
 variable "adot_log_group_name" {

variables.tf

@@ -266,8 +266,9 @@ variable "metrics_exporter" {
 variable "adot_config" {
   type = any
   default = {
-    accepte_namespace_regex = "(default|kube-system)"
-    additional_metrics      = {}
+    accept_namespace_regex = "(default|kube-system)"
+    additional_metrics     = {}
+    log_group_name         = "adot_log_group"
   }
 }
 
@@ -377,3 +378,9 @@ variable "enable_olm" {
   default     = false
   description = "To install OLM controller (experimental)."
 }
+
+variable "prometheus_metrics" {
+  description = "Prometheus Metrics"
+  type        = any
+  default     = []
+}