adding prometheus and deleted eks-cert-mgr-ns as it included in eks-c…

…ert-manager

lab/us-gov-east-1/vpc/cluster/eks-prometheus/README.md

@@ -0,0 +1,55 @@
+## eks-prometheus
+This module deploys EKS kubeenetes prometheus inside existing EKS cluster. Prometheus is an open-source systems monitoring and alerting tool.
+This module consisits of 4 components. It creates prometheus namespace and  copies image repositories for the following components from quay.io into local account ECR repository. It deploys these components using helm charts using the configured ECR repositories.
+    1. prometheus-alert-manager
+    2. prometheus-node-exporter
+    3. prometheus-pushgateway
+    4. prometheus-server
+
+### Dependencies
+This module is dependent on EKS module (eks). The cluster should exist already for this module to work.
+
+### Inputs
+    cluster_name
+    profile
+    prometheus_chart_version
+    prometheus_server_tag
+    prometheus_config_reloader_tag
+    alertmanager_tag
+    kube_state_metrics_tag
+    node_exporter_tag
+    pushgateway_tag
+    rwo_storage_class
+
+### Outputs
+    alertmanager_internal_endpoint
+    alertmanager_headless_internal_endpoint
+    pushgateway_internal_endpoint
+    prometheus_server_internal_endpoint
+
+### Issues observed/fixed
+    1. The rwo_storage_class value had to be updated from "gp3" to "gp3-encrypted"
+    2. The node_exporter_tag value had to be updated from "1.6.1" to "v1.8.1"
+    3. The kube_state_metrics_tag value had to be updated from "2.10.0" to "v2.6.0"
+    4. The alertmanager_tag value had to be updated from
+    5. The helm chart set config for the ecr image had to be split into 2 components, one for registry and other for repository as an example mentioned below:
+    ```
+    set {
+    name = "kube-state-metrics.image.registry"
+    value = module.images.images[local.ksm_key].dest_registry
+  }
+  set {
+    name = "kube-state-metrics.image.repository"
+    value = module.images.images[local.ksm_key].dest_repository
+  }
+  ```
+  6. In some other cases the image ecr repository had to be split by the colon separatory (:)
+    ```
+    set {
+    name = "alertmanager.configmapReload.image.repository"
+    value = split(":", module.images.images[local.prom_config_reload_key].dest_full_path)[0]
+  }
+  ```
+    
+
+

lab/us-gov-east-1/vpc/cluster/eks-cert-mgr-ns/provider.tf → lab/us-gov-east-1/vpc/cluster/eks-prometheus/provider.tf

@@ -1,3 +1,4 @@
+# provider.tf
 provider "aws" {
   region  = "us-gov-east-1"
   profile = "224384469011-lab-dev-gov.inf-admin-t3"

lab/us-gov-east-1/vpc/cluster/eks-prometheus/terragrunt.hcl

@@ -0,0 +1,31 @@
+terraform {
+  source      = "git@github.e.it.census.gov:SCT-Engineering/tfmod-prometheus.git?ref=platform-changes"
+  extra_arguments "retry_lock" {
+    commands  = get_terraform_commands_that_need_locking()
+    arguments = ["-lock-timeout=20m"]
+  }
+}
+
+remote_state {
+  backend     = "s3"
+  generate    = {
+    path      = "backend.tf"
+    if_exists = "overwrite_terragrunt"
+  }
+  config = {
+    bucket         = "tg-infrastructure-tf-state-lab-dev-ew-us-gov-east-1"
+    key            = "platform-eks-test-prometheus/terraform.tfstate"
+    region         = "us-gov-east-1"
+    encrypt        = true
+  }
+}
+
+dependency "eks" {
+  config_path = "../eks"
+}
+
+inputs = {
+  profile                                         = dependency.eks.inputs.profile
+  cluster_name                                    = dependency.eks.inputs.cluster_name
+  region                                          = "us-gov-east-1"
+}