Lots of Nomad updates to support metrics

nomad/metrics/.terraform.lock.hcl

@@ -0,0 +1,20 @@
+# This file is maintained automatically by "terraform init".
+# Manual edits may be lost in future updates.
+
+provider "registry.terraform.io/hashicorp/nomad" {
+  version = "1.4.16"
+  hashes = [
+    "h1:tyfjD/maKzb0RxxD9KWgLnkJu9lnYziYsQgGw85Giz8=",
+    "zh:0d4fbb7030d9caac3b123e60afa44f50c83cc2a983e1866aec7f30414abe7b0e",
+    "zh:0db080228e07c72d6d8ca8c45249d6f97cd0189fce82a77abbdcd49a52e57572",
+    "zh:0df88393271078533a217654b96f0672c60eb59570d72e6aefcb839eea87a7a0",
+    "zh:2883b335bb6044b0db6a00e602d6926c047c7f330294a73a90d089f98b24d084",
+    "zh:390158d928009a041b3a182bdd82376b50530805ae92be2b84ed7c3b0fa902a0",
+    "zh:7169b8f8df4b8e9659c49043848fd5f7f8473d0471f67815e8b04980f827f5ef",
+    "zh:9417ee1383b1edd137024882d7035be4dca51fb4f725ca00ed87729086ec1755",
+    "zh:a22910b5a29eeab5610350700b4899267c1b09b66cf21f7e4d06afc61d425800",
+    "zh:a6185c9cd7aa458cd81861058ba568b6411fbac344373a20155e20256f4a7557",
+    "zh:b6260ca9f034df1b47905b4e2a9c33b67dbf77224a694d5b10fb09ae92ffad4c",
+    "zh:d87c12a6a7768f2b6c2a59495c7dc00f9ecc52b1b868331d4c284f791e278a1e",
+  ]
+}

nomad/metrics/exporters.nomad

@@ -0,0 +1,225 @@
+job "metrics" {
+  datacenters = ["dc1"]
+  type = "system"
+
+  group "exporters" {
+
+    network {
+      mode = "bridge"
+
+      port "cadvisor" {
+        host_network = "nomad-bridge"
+        to = 8080
+      }
+
+      port "node_exporter" {
+        host_network = "nomad-bridge"
+        to = 9100
+      }
+
+      port "promtail" {
+        host_network = "nomad-bridge"
+        to = 9080
+      }
+
+      port "expose" {
+        host_network = "nomad-bridge"
+      }
+
+      port "cadvisor_envoy_metrics" {
+        host_network = "nomad-bridge"
+        to = 9102
+      }
+    }
+
+    service {
+      name = "cadvisor"
+      port = "cadvisor"
+
+      meta {
+        metrics_addr = "${NOMAD_ADDR_expose}"
+        envoy_metrics_addr = "${NOMAD_ADDR_cadvisor_envoy_metrics}"
+        nomad_dc = "${NOMAD_DC}"
+        nomad_node_name = "${node.unique.name}"
+      }
+
+      connect {
+        sidecar_service {
+          proxy {
+            local_service_port = 8080
+
+            expose {
+              path {
+                path = "/metrics"
+                protocol = "http"
+                local_path_port = 8080
+                listener_port = "expose"
+              }
+            }
+
+            config {
+              envoy_prometheus_bind_addr = "0.0.0.0:9102"
+            }
+          }
+        }
+
+        sidecar_task {
+          resources {
+            cpu    = 50
+            memory = 50
+          }
+        }
+      }
+
+      check {
+        type = "http"
+        path = "/metrics"
+        port = "cadvisor"
+        interval = "10s"
+        timeout = "10s"
+      }
+
+      // TODO: Remove traefik tags
+      tags = [
+        "traefik.enable=true",
+        "traefik.http.routers.cadvisor.entrypoints=web,websecure",
+        "traefik.http.routers.cadvisor.rule=Host(`cadvisor.dev.homelab`)",
+        "traefik.http.routers.cadvisor.tls=true",
+      ]
+    }
+
+    task "cadvisor" {
+      driver = "docker"
+      
+      config {
+        # image = "iamthefij/cadvisor:0.37.5"
+        image = "gcr.io/cadvisor/cadvisor:v0.39.3"
+        args = ["--docker_only=true"]
+
+        ports = ["cadvisor"]
+
+        # volumes = [
+        #   "/:/rootfs:ro",
+        #   "/var/run:/var/run:rw",
+        #   "/sys:/sys:ro",
+        #   "/var/lib/docker/:/var/lib/docker:ro",
+        #   "/cgroup:/cgroup:ro",
+        #   "/etc/machine-id:/etc/machine-id:ro",
+        # ]
+
+        mount {
+          type = "bind"
+          source = "/"
+          target = "/rootfs"
+          readonly = true
+        }
+
+        mount {
+          type = "bind"
+          source = "/var/run"
+          target = "/var/run"
+          readonly = false
+        }
+
+        mount {
+          type = "bind"
+          source = "/sys"
+          target = "/sys"
+          readonly = true
+        }
+
+        mount {
+          type = "bind"
+          source = "/var/lib/docker"
+          target = "/var/lib/docker"
+          readonly = true
+        }
+
+        # mount {
+        #   type = "bind"
+        #   source = "/cgroup"
+        #   target = "/cgroup"
+        #   readonly = true
+        # }
+
+        mount {
+          type = "bind"
+          source = "/etc/machine-id"
+          target = "/etc/machine-id"
+          readonly = true
+        }
+      }
+
+      resources {
+        cpu = 50
+        memory = 100
+      }
+    }
+
+    service {
+      name = "nodeexporter"
+      port = "node_exporter"
+
+      meta {
+        metrics_addr = "${NOMAD_ADDR_node_exporter}"
+        nomad_dc = "${NOMAD_DC}"
+        nomad_node_name = "${node.unique.name}"
+      }
+
+      connect {
+        sidecar_service {
+          proxy {
+            local_service_port = 9100
+          }
+        }
+
+        sidecar_task {
+          resources {
+            cpu    = 50
+            memory = 50
+          }
+        }
+      }
+
+      check {
+        type = "http"
+        path = "/metrics"
+        port = "node_exporter"
+        interval = "10s"
+        timeout = "10s"
+      }
+
+      // TODO: Remove traefik tags
+      tags = [
+        "traefik.enable=true",
+        "traefik.http.routers.node_exporter.entrypoints=web,websecure",
+        "traefik.http.routers.node_exporter.rule=Host(`node_exporter.dev.homelab`)",
+        "traefik.http.routers.node_exporter.tls=true",
+      ]
+    }
+
+    task "node_exporter" {
+      driver = "docker"
+      
+      config {
+        image = "prom/node-exporter:v1.0.1"
+        args = ["--path.rootfs", "/host"]
+
+        ports = ["node_exporter"]
+
+        mount {
+          type = "bind"
+          source = "/"
+          target = "/host"
+          readonly = true
+        }
+      }
+
+      resources {
+        cpu = 50
+        memory = 50
+      }
+    }
+  }
+}
+

nomad/metrics/metrics.tf

@@ -0,0 +1,46 @@
+variable "base_hostname" {
+  type = string
+  description = "Base hostname to serve content from"
+  default = "dev.homelab"
+}
+
+variable "consul_address" {
+  type    = string
+  description = "address of consul server for dynamic scraping"
+}
+
+resource "nomad_job" "exporters" {
+  hcl2 {
+    enabled = true
+  }
+
+  jobspec = file("${path.module}/exporters.nomad")
+}
+
+resource "nomad_job" "prometheus" {
+  hcl2 {
+    enabled = true
+    vars = {
+      "consul_address" = "${var.consul_address}",
+    }
+  }
+
+  jobspec = file("${path.module}/prometheus.nomad")
+}
+#
+# resource "nomad_job" "prometheus" {
+#   hcl2 {
+#     enabled = true
+#   }
+#
+# }
+#
+# resource "nomad_job" "grafana" {
+#   hcl2 {
+#     enabled = true
+#     vars = {
+#       "base_hostname" = "${var.base_hostname}",
+#     }
+#   }
+#
+# }

nomad/metrics/prometheus.nomad

@@ -0,0 +1,182 @@
+variable "consul_address" {
+  type = string
+  description = "Full address of Consul instance to get catalog from"
+  default = "http://127.0.0.1:5400"
+}
+
+job "prometheus" {
+  datacenters = ["dc1"]
+
+  group "prometheus" {
+    count = 1
+
+    network {
+      mode = "bridge"
+
+      port "web" {
+        host_network = "nomad-bridge"
+        to = 9090
+      }
+    }
+
+    service {
+      port = "web"
+
+      connect {
+        sidecar_service {
+          proxy {
+            local_service_port = 9090
+          }
+        }
+
+        sidecar_task {
+          resources {
+            cpu    = 50
+            memory = 50
+          }
+        }
+      }
+
+      check {
+        type = "http"
+        path = "/"
+        port = "web"
+        interval = "10s"
+        timeout = "10s"
+      }
+
+      // TODO: Remove traefik tags
+      tags = [
+        "traefik.enable=true",
+        "traefik.http.routers.prometheus.entrypoints=web,websecure",
+        "traefik.http.routers.prometheus.rule=Host(`prometheus.dev.homelab`)",
+        "traefik.http.routers.prometheus.tls=true",
+      ]
+    }
+
+    task "prometheus" {
+      driver = "docker"
+
+      config {
+        image = "prom/prometheus:v2.30.2"
+        ports = ["web"]
+        args = [
+          "--config.file=/etc/prometheus/config/prometheus.yml",
+          "--storage.tsdb.path=/prometheus",
+          "--web.listen-address=0.0.0.0:9090",
+          "--web.console.libraries=/usr/share/prometheus/console_libraries",
+          "--web.console.templates=/usr/share/prometheus/consoles",
+        ]
+
+        mount {
+          type = "bind"
+          target = "/etc/prometheus/config"
+          source = "local/config"
+        }
+      }
+
+      template {
+        data = <<EOF
+---
+global:
+  scrape_interval: 30s
+  evaluation_interval: 3s
+
+scrape_configs:
+  - job_name: prometheus
+    static_configs:
+      - targets:
+        - 0.0.0.0:9090
+
+  - job_name: "nomad_server"
+    metrics_path: "/v1/metrics"
+    params:
+      format:
+        - "prometheus"
+    consul_sd_configs:
+      - server: "${var.consul_address}"
+        services:
+          - "nomad"
+        tags:
+          - "http"
+
+  - job_name: "nomad_client"
+    metrics_path: "/v1/metrics"
+    params:
+      format:
+        - "prometheus"
+    consul_sd_configs:
+      - server: "${var.consul_address}"
+        services:
+          - "nomad-client"
+
+  - job_name: "consul"
+    metrics_path: "/v1/agent/metrics"
+    params:
+      format:
+        - "prometheus"
+    consul_sd_configs:
+      - server: "${var.consul_address}"
+        services:
+          - "consul"
+    relabel_configs:
+      - source_labels: [__meta_consul_address]
+        replacement: $1:8500
+        target_label: __address__
+        
+  - job_name: "exporters"
+    metrics_path: "/metrics"
+    consul_sd_configs:
+      - server: "${var.consul_address}"
+        services:
+          - "cadvisor"
+          - "nodeexporter"
+    relabel_configs:
+      - source_labels: [__meta_consul_service_metadata_metrics_addr]
+        action: keep
+        regex: (.+)
+      - source_labels: [__meta_consul_service_metadata_metrics_addr]
+        target_label: __address__
+      - source_labels: [__meta_consul_service]
+        target_label: consul_service
+      - source_labels: [__meta_consul_node]
+        target_label: consul_node
+      - source_labels: [__meta_consul_service_nomad_dc]
+        target_label: nomad_dc
+      - source_labels: [__meta_consul_service_nomad_node_name]
+        target_label: nomad_node_name
+
+  - job_name: "envoy"
+    metrics_path: "/metrics"
+    consul_sd_configs:
+      - server: "${var.consul_address}"
+    relabel_configs:
+      - source_labels: [__meta_consul_service]
+        action: drop
+        regex: (.+)-sidecar-proxy
+      - source_labels: [__meta_consul_service_metadata_envoy_metrics_addr]
+        action: keep
+        regex: (.+)
+      - source_labels: [__meta_consul_service_metadata_envoy_metrics_addr]
+        target_label: __address__
+      - source_labels: [__meta_consul_service]
+        target_label: consul_service
+      - source_labels: [__meta_consul_node]
+        target_label: consul_node
+      - source_labels: [__meta_consul_service_nomad_dc]
+        target_label: nomad_dc
+      - source_labels: [__meta_consul_service_nomad_node_name]
+        target_label: nomad_node_name
+        EOF
+        change_mode = "signal"
+        change_signal = "SIGHUP"
+        destination = "local/config/prometheus.yml"
+      }
+
+      resources {
+        cpu = 100
+        memory = 200
+      }
+    }
+  }
+}

nomad/services.tf

@@ -32,11 +32,11 @@ provider "nomad" {
 
 # Define services as modules
 
-module "mysql-server" {
-  source = "./mysql"
-
-  base_hostname = var.base_hostname
-}
+# module "mysql-server" {
+#   source = "./mysql"
+#
+#   base_hostname = var.base_hostname
+# }
 
 module "blocky" {
   source = "./blocky"
@@ -51,14 +51,21 @@ module "traefik" {
   base_hostname = var.base_hostname
 }
 
-module "nextcloud" {
-  source = "./nextcloud"
+module "metrics" {
+  source = "./metrics"
 
+  consul_address = var.consul_address
   base_hostname = var.base_hostname
-
-  depends_on = [module.mysql-server]
 }
 
+# module "nextcloud" {
+#   source = "./nextcloud"
+#
+#   base_hostname = var.base_hostname
+#
+#   depends_on = [module.mysql-server]
+# }
+
 resource "nomad_job" "whoami" {
   hcl2 {
     enabled = true

nomad/setup-cluster.yml

@@ -19,6 +19,12 @@
       consul_ports_grpc: 8502
       consul_client_address: "0.0.0.0"
 
+      # Enable metrics
+      consul_config_custom:
+        telemetry:
+          prometheus_retention_time: "2h"
+
+
   tasks:
     - name: Start Consul
       systemd:
@@ -26,6 +32,7 @@
         name: consul
 
     - name: Add values
+      # TODO: This can be run from localhost by using an address
       block:
         - name: Install pip
           pip:
@@ -35,6 +42,7 @@
           consul_kv:
             key: ansible_test
             value: Hello from Ansible!
+
       run_once: true
 
 - name: Build Nomad cluster
@@ -49,28 +57,55 @@
       nomad_install_upgrade: true
       nomad_allow_purge_config: true
 
-      nomad_encrypt_enable: true
-      # nomad_use_consul: true
-
-      nomad_cni_enable: true
-      nomad_cni_version: 1.0.1
-      nomad_docker_enable: true
-      nomad_docker_dmsetup: false
-
-      nomad_bind_address: 0.0.0.0
+      # Properly map install arch
       nomad_architecture_map:
         x86_64: amd64
         armhfv6: arm
         armv7l: arm
 
+      nomad_encrypt_enable: true
+      # nomad_use_consul: true
+
+      # Metrics
+      nomad_telemetry: true
+      nomad_telemetry_prometheus_metrics: true
+      nomad_telemetry_publish_allocation_metrics: true
+      nomad_telemetry_publish_node_metrics: true
+
+      # Enable container plugins
+      nomad_cni_enable: true
+      nomad_cni_version: 1.0.1
+      nomad_docker_enable: true
+      nomad_docker_dmsetup: false
+
+      # Customize docker plugin
+      nomad_plugins:
+        docker:
+          config:
+            volumes:
+              enabled: true
+              selinuxlabel: "z"
+            extra_labels:
+              - "job_name"
+              - "job_id"
+              - "task_group_name"
+              - "task_name"
+              - "namespace"
+              - "node_name"
+              - "node_id"
+
+      # Bind nomad
+      nomad_bind_address: 0.0.0.0
+
+      # Create networks for binding task ports
       nomad_host_networks:
         - name: nomad-bridge
-          # cidr: 172.26.64.0/20
           interface: nomad
           reserved_ports: "22"
         - name: loopback
           interface: lo
           reserved_ports: "22"
+
   tasks:
     - name: Start Nomad
       systemd:

nomad/traefik/traefik.tf

@@ -5,7 +5,8 @@ variable "base_hostname" {
 }
 
 variable "consul_address" {
-  type    = string
+  type = string
+  description = "address of consul server for dynamic routes"
 }
 
 resource "nomad_job" "traefik" {