AP-565 enable monitoring in soak

terraform/examples/multi-node-cluster/main.tf

@@ -276,3 +276,20 @@ module "dist" {
   pop_id                   = var.edge01_pop_id
   server_id                = var.edge01_host
 }
+
+module "monitoring_edge" {
+  source                   = "../../modules/monitoring_edge"
+  docker_host              = "${var.docker_protocol}${var.edge01_host}"
+  docker_network           = docker_network.edge01.name
+  docker_image_username = var.docker_image_username
+  docker_image_repository = "${var.docker_image_repository}_monitoring_edge"
+  docker_image_tag = var.docker_image_tag
+  docker_registry_address  = var.docker_registry_address
+  docker_registry_username = var.docker_registry_username
+  docker_registry_password = var.docker_registry_password
+  docker_log_driver     = var.docker_log_driver
+  hostname    = var.monitoring_edge_hostname
+  pop_id                   = var.edge01_pop_id
+  server_id                = var.edge01_host
+  core_hosts = [element(split("@", var.control01_host), 1)]
+}

terraform/examples/multi-node-cluster/variables.tf

@@ -149,3 +149,9 @@ variable "docker_log_driver" {
   default     = "json-file"
   description = "Docker log driver to use, see https://docs.docker.com/config/containers/logging/configure/"
 }
+
+variable "monitoring_edge_hostname" {
+  default     = "monitoring_edge"
+  description = "Hostmaster email address used in SOA records"
+}
+

terraform/modules/monitoring_edge/main.tf

@@ -0,0 +1,95 @@
+terraform {
+  required_version = ">= 0.12"
+}
+
+locals {
+  docker_image_name = "${var.docker_image_username}/${var.docker_image_repository}:${var.docker_image_tag}"
+}
+
+provider "docker" {
+  host = var.docker_host
+
+  # If registry address is provided, configure registry_auth
+  dynamic "registry_auth" {
+    for_each = var.docker_registry_address != null ? list(var.docker_registry_address) : []
+    iterator = address
+    content {
+      address  = address.value
+      username = var.docker_registry_username
+      password = var.docker_registry_password
+    }
+  }
+}
+
+data "docker_registry_image" "monitoring_edge" {
+  count = var.docker_registry_address != null ? 1 : 0
+  name  = local.docker_image_name
+}
+
+resource "docker_image" "monitoring_edge" {
+  count         = var.docker_registry_address != null ? 1 : 0
+  name          = data.docker_registry_image.monitoring_edge[count.index].name
+  pull_triggers = [data.docker_registry_image.monitoring_edge[count.index].sha256_digest]
+  keep_locally  = true
+}
+
+resource "docker_volume" "monitoring_edge" {
+  name = "ns1monitoring_edge"
+}
+
+resource "docker_container" "monitoring_edge" {
+  name = "monitoring_edge"
+  # If using registry, use sha of found image, otherwise use name that should be found on docker host
+  image = var.docker_registry_address != null ? docker_image.monitoring_edge[0].latest : local.docker_image_name
+
+  env = [
+    # "CONFIG_PORT=3305",
+    "CONTAINER_NAME=${var.container_name}",
+  ]
+
+  restart = "unless-stopped"
+
+  hostname = var.hostname
+
+  log_driver = var.docker_log_driver
+
+  healthcheck {
+    test     = ["CMD", "supd", "health"]
+    interval = "15s"
+    timeout  = "10s"
+    retries  = 3
+  }
+
+  volumes {
+    volume_name    = docker_volume.monitoring_edge.name
+    container_path = "/ns1/data"
+  }
+
+
+  command = [
+    "--pop_id",
+    var.pop_id,
+    "--server_id",
+    var.server_id,
+    "--core_host",
+    join(",", var.core_hosts),
+    "--monitoring_region",
+    var.monitoring_region,
+    "--digest_service_def_id",
+    var.digest_service_def_id,
+    "--log_level",
+    var.log_level,
+    "--metrics_addr_base",
+    var.metrics_addr_base,
+    "--inst_id",
+    var.inst_id,
+    "--use_privileged_ping",
+    var.use_privileged_ping,
+    "--jitter_seconds",
+    var.jitter_seconds,
+]
+
+  network_mode = "host"
+
+  privileged = true
+}

terraform/modules/monitoring_edge/variables.tf

@@ -0,0 +1,103 @@
+variable "docker_image_tag" {
+  default     = "2.5.0"
+  description = "The image tag of the Docker image. Defaults to the latest GA version number."
+}
+
+variable "docker_image_username" {
+  default     = "ns1inc"
+  description = "The username used in the Docker image name. This should not need to be changed."
+}
+
+variable "docker_image_repository" {
+  default     = "privatedns_monitoring_edge"
+  description = "The repository name used in the Docker image name. This should not need to be changed."
+}
+
+variable "docker_registry_address" {
+  default     = null
+  description = "The absolute URL of the Docker registry (i.e. 'https://registry.hub.docker.com') to pull the container images from.  If not provided, it's assumed the container image is already loaded on the Docker host."
+}
+
+variable "docker_registry_username" {
+  default     = null
+  description = "Username for authentication to Docker registry.  Only required if 'docker_registry_address' is provided."
+}
+
+variable "docker_registry_password" {
+  default     = null
+  description = "Password for authentication to Docker registry.   Only required if 'docker_registry_address' is provided."
+}
+
+variable "docker_host" {
+  description = "The address of the Docker host to deploy the container on. Both ssh:// and tcp:// are supported.  See https://www.terraform.io/docs/providers/docker/index.html for more details."
+}
+
+variable "docker_network" {
+  description = "The name of the Docker network to connect to the container.  Must already exist on the Docker host."
+}
+
+variable "container_name" {
+  default     = "monitoring_edge"
+  description = "The name of the Docker container."
+}
+
+variable "pop_id" {
+  default     = "mypop"
+  description = "Specifies the location (datacenter/pop) of the server where the data container is running"
+}
+
+variable "server_id" {
+  default     = "myserver"
+  description = "Identifies a specific server in a location where the data container is running"
+}
+
+variable "inst_id" {
+  default     = "1"
+  description = "Identifies a specific instance of the service"
+}
+
+variable "core_hosts" {
+  type        = list(string)
+  default     = ["core"]
+  description = "List of upstream core containers.  If core containers are on same Docker host, container name can be used.  If core containers are on a seperate Docker host, IP or FQDN of host should be used."
+}
+
+variable "hostname" {
+  default     = "monitoring_edge"
+  description = "Hostname to give the running container"
+}
+
+variable "docker_log_driver" {
+  default     = "json-file"
+  description = "Docker log driver to use, see https://docs.docker.com/config/containers/logging/configure/"
+}
+
+variable "monitoring_region" {
+  default = "lga"
+  description = "Monitoring region code of the service definition"
+}
+
+variable "digest_service_def_id" {
+  default = "4"
+  description = "ID of the monitoring service definition for reading the monitoring jobs digest"
+}
+
+variable "log_level" {
+  default = "info"
+  description = "Logs level of monprobed service"
+}
+
+variable "metrics_addr_base" {
+  default = "unix:///var/run/ns1daemons/"
+  description = "Unix socket that monprobed serves metrics over"
+}
+
+variable "use_privileged_ping" {
+  default = "true"
+  description = "Use privileged to create the necessary raw ICMP sockets for ping probes"
+}
+
+variable "jitter_seconds" {
+  default = "5"
+  description = "A probe will take between 0 and jitter_seconds seconds to start running after it is created"
+}