Fix AWS permissions for Kubernetes Response Engine

integrations/kubernetes-response-engine/deployment/aws/.gitignore

@@ -1,4 +1,4 @@
 .terraform/*
 .terraform.*
 terraform.*
-*.yaml
+aws-auth-patch.yml

integrations/kubernetes-response-engine/deployment/aws/Makefile

@@ -1,11 +1,17 @@
-all: create configure
+deploy: rbac create configure
+
+rbac:
+	kubectl apply -f cluster-role.yaml
+	kubectl apply -f cluster-role-binding.yaml
 
 create:
-	terraform apply
+	terraform apply -auto-approve
 
 configure:
 	kubectl get -n kube-system configmap/aws-auth -o yaml | awk "/mapRoles: \|/{print;print \"$(shell terraform output patch_for_aws_auth)\";next}1" > aws-auth-patch.yml
 	kubectl -n kube-system replace -f aws-auth-patch.yml
 
 clean:
-	terraform destroy
+	terraform destroy -force
+	kubectl delete -f cluster-role-binding.yaml
+	kubectl delete -f cluster-role.yaml

integrations/kubernetes-response-engine/deployment/aws/cluster-role-binding.yaml

@@ -0,0 +1,12 @@
+apiVersion: rbac.authorization.k8s.io/v1beta1
+kind: ClusterRoleBinding
+metadata:
+  name: kubernetes-response-engine-cluster-role-binding
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: kubernetes-response-engine-cluster-role
+subjects:
+- kind: User
+  apiGroup: rbac.authorization.k8s.io
+  name: kubernetes-response-engine

integrations/kubernetes-response-engine/deployment/aws/cluster-role.yaml

@@ -0,0 +1,25 @@
+kind: ClusterRole
+apiVersion: rbac.authorization.k8s.io/v1
+metadata:
+  name: kubernetes-response-engine-cluster-role
+rules:
+  - apiGroups:
+    - ""
+    resources:
+      - pods
+    verbs:
+      - delete
+      - list
+      - patch
+  - apiGroups:
+    - ""
+    resources:
+      - nodes
+    verbs:
+      - patch
+  - apiGroups:
+      - batch
+    resources:
+      - jobs
+    verbs:
+      - create

integrations/kubernetes-response-engine/deployment/aws/lambda.tf

@@ -1,3 +1,7 @@
+resource "aws_iam_user" "kubernetes-response-engine-user" {
+  name = "kubernetes_response_engine"
+}
+
 resource "aws_iam_role" "iam-for-lambda" {
   name = "iam_for_lambda"
 
@@ -9,7 +13,7 @@ resource "aws_iam_role" "iam-for-lambda" {
       "Action": "sts:AssumeRole",
       "Principal": {
         "Service": "lambda.amazonaws.com",
-        "AWS": "${var.iam-user-arn}"
+        "AWS": "${aws_iam_user.kubernetes-response-engine-user.arn}"
       },
       "Effect": "Allow",
       "Sid": ""

integrations/kubernetes-response-engine/deployment/aws/outputs.tf

@@ -1,7 +1,7 @@
 locals {
   patch_for_aws_auth = <<CONFIGMAPAWSAUTH
     - rolearn: ${aws_iam_role.iam-for-lambda.arn}\n
-     username: kubernetes-admin\n
+     username: kubernetes-response-engine\n
      groups:\n
        - system:masters
 CONFIGMAPAWSAUTH

integrations/kubernetes-response-engine/deployment/cncf/Makefile

@@ -1,7 +1,6 @@
 deploy:
 	kubectl apply -f nats/
 	kubectl apply -f kubeless/
-	kubectl apply -f network-policy.yaml
 	kubectl apply -f .
 
 clean:

integrations/kubernetes-response-engine/deployment/cncf/rbac.yaml → integrations/kubernetes-response-engine/deployment/cncf/cluster-role-binding.yaml

@@ -1,12 +1,13 @@
 apiVersion: rbac.authorization.k8s.io/v1beta1
 kind: ClusterRoleBinding
 metadata:
-  name: sysdig-kubeless
+  name: kubernetes-response-engine-cluster-role-binding
 roleRef:
   apiGroup: rbac.authorization.k8s.io
   kind: ClusterRole
-  name: cluster-admin
+  name: kubernetes-response-engine-cluster-role
 subjects:
 - kind: ServiceAccount
   name: default
   namespace: default
+  apiGroup: rbac.authorization.k8s.io

integrations/kubernetes-response-engine/deployment/cncf/cluster-role.yaml

@@ -0,0 +1,25 @@
+kind: ClusterRole
+apiVersion: rbac.authorization.k8s.io/v1
+metadata:
+  name: kubernetes-response-engine-cluster-role
+rules:
+  - apiGroups:
+    - ""
+    resources:
+      - pods
+    verbs:
+      - delete
+      - list
+      - patch
+  - apiGroups:
+    - ""
+    resources:
+      - nodes
+    verbs:
+      - patch
+  - apiGroups:
+      - batch
+    resources:
+      - jobs
+    verbs:
+      - create

integrations/kubernetes-response-engine/playbooks/deploy_playbook_aws

@@ -16,7 +16,7 @@ You must pass the playbook and at least one topic to subscribe.
 
 Example:
 
-deploy_playbook -p slack -t "falco.error.*" -e SLACK_WEBHOOK_URL=http://foobar.com/... -k sysdig_eks
+deploy_playbook -p slack -e SLACK_WEBHOOK_URL=http://foobar.com/... -k sysdig_eks
 EOF
   exit 1
 }
@@ -27,7 +27,7 @@ playbook=""
 environment=("KUBECONFIG=kubeconfig" "KUBERNETES_LOAD_KUBE_CONFIG=1")
 eks_cluster="${EKS_CLUSTER}"
 
-while getopts "r:e:t:" arg; do
+while getopts "p:e:k:" arg; do
   case $arg in
     p)
       playbook="${OPTARG}"