Summary: Fixed escluster to run as docker

Changesets:
- Simplify config with Docker; had to up the memory as JVM minimum memory is 2Gb!
- Adjust listening ip to all interface so that cluster discovery works
- Tried to handle authentication for ES endpoints but fail
TODO: Once hashicorp/nomad#2924 is fixed; need to send in the

java/go-elasticsearch/docker-cluster-elasticsearch.nomad

@@ -0,0 +1,204 @@
+job "search" {
+  datacenters = [
+    "dc1"
+  ]
+  type = "service"
+
+  update {
+    stagger = "10s"
+    max_parallel = 1
+  }
+
+  group "simple" {
+    count = 1
+    restart {
+      attempts = 2
+      interval = "5m"
+      delay = "25s"
+      mode = "delay"
+    }
+
+    task "elasticsearch" {
+      driver = "docker"
+
+      config {
+        image = "docker.elastic.co/elasticsearch/elasticsearch:5.5.1"
+        port_map {
+          eshttp = 9200
+          estransport = 9300
+        }
+        volumes = [
+          "local/elasticsearch.yml:/usr/share/elasticsearch/config/elasticsearch.yml"
+        ]
+      }
+
+      env {
+        ES_JAVA_OPTS = "-Xmx500m -Xms500m"
+      }
+
+      resources {
+        # 100 MHz; burstable
+        cpu = 100
+        disk = 500
+        # 768 MB
+        memory = 2048
+        network {
+          mbits = 10
+          port "eshttp" {}
+          port "estransport" {}
+        }
+      }
+
+      service {
+        tags = [
+          "simple",
+          "search"
+        ]
+        port = "eshttp"
+        check {
+          name = "alive"
+          type = "tcp"
+          interval = "10s"
+          timeout = "2s"
+        }
+      }
+
+      template {
+        data = <<EOH
+          # Ref: https://www.elastic.co/guide/en/elasticsearch/reference/current/modules-network.html#network-interface-values
+          # network.host: [ "0.0.0.0", "_eth0_", "_site_", "_global_" ]
+          # discovery.zen.ping.unicast.hosts: ["127.0.0.1", "[::1]"]
+          # http.port: {{ env "NOMAD_HOST_PORT_eshttp" }}
+          # transport.tcp.port: {{ env "NOMAD_HOST_PORT_estransport" }}
+          # Below special cases needed to address proxy
+          # network.bind_host (incoming), network.publish_host (node-to-node communications)
+          # http.publish_port: {{ env "NOMAD_HOST_PORT_eshttp" }}
+          # transport.publish_port: {{ env "NOMAD_HOST_PORT_estransport" }}
+
+          network.host: 0.0.0.0
+          # network.publish_host: "{{ env "attr.unique.network.ip-address" }}"
+          # network.bind_host: 0.0.0.0
+          # http.port: "{{ env "NOMAD_HOST_PORT_eshttp" }}"
+          # transport.tcp.port: "{{ env "NOMAD_HOST_PORT_estransport" }}"
+
+          # Specific port from example elasticsearch.yml (memory should be 1/2 of available memory)
+          # bootstrap.memory_lock: true
+          # node.name: node-1
+          # node.attr.rack: r1
+          # path.data: /path/to/data,/another/path/to/data_for_perf_raid0
+          # path.logs: /path/to/logs
+
+          # Below are tweaks which may only be suitable in dev environments
+          cluster.routing.allocation.disk.threshold_enabled: false
+      EOH
+        destination = "/local/elasticsearch.yml"
+        change_mode = "noop"
+      }
+
+    }
+  }
+
+  group "complex" {
+    count = 3
+    restart {
+      attempts = 3
+      interval = "1m"
+      delay = "10s"
+      mode = "delay"
+    }
+
+    ephemeral_disk {
+      migrate = true
+      size = "1500"
+      sticky = true
+    }
+
+    // Main leader job in pod; the ElasticSearch Cluster
+    task "elasticsearch" {
+
+      driver = "docker"
+
+      config {
+        image = "docker.elastic.co/elasticsearch/elasticsearch:5.5.1"
+        port_map {
+          eshttp = 9200
+          estransport = 9300
+        }
+        volumes = [
+          "local/elasticsearch.yml:/usr/share/elasticsearch/config/elasticsearch.yml"
+        ]
+      }
+
+      env {
+        ES_CLUSTER_NAME = "escluster"
+        ES_HOME = "local/elasticsearch-5.5.0"
+      }
+
+      resources {
+        # 150 MHz; burstable
+        cpu = 150
+        # 500 MB
+        disk = 500
+        # 1024 MB
+        memory = 3068
+        network {
+          mbits = 100
+          port "eshttp" {}
+          port "estransport" {}
+        }
+      }
+
+      service {
+        name = "escluster-transport"
+        port = "estransport"
+      }
+
+      service {
+        name = "escluster"
+        tags = [
+          "lolcats",
+          "cluster"
+        ]
+        port = "eshttp"
+        check {
+          name = "available"
+          type = "tcp"
+          interval = "10s"
+          timeout = "2s"
+        }
+        check {
+          name = "ready"
+          type = "http"
+          port = "eshttp"
+          path = "/_cluster/health?wait_for_status=yellow"
+          interval = "30s"
+          timeout = "10s"
+        }
+      }
+
+      # Great refrenece from:
+      # https://groups.google.com/forum/#!searchin/nomad-tool/elasticsearch%7Csort:relevance/nomad-tool/0qtyyor1ORs/uN9uIyN7BAAJ
+      template {
+        data = <<EOH
+          cluster.name: {{ env "ES_CLUSTER_NAME" }}
+          network.host: 0.0.0.0
+          network.publish_host: 0.0.0.0
+          network.bind_host: 0.0.0.0
+
+          discovery.zen.minimum_master_nodes: 2
+          {{ if service "escluster-transport"}}discovery.zen.ping.unicast.hosts:{{ range service "escluster-transport" }}
+            - {{ if eq .Address "::1" }}localhost{{ else }}{{ .Address }}{{ end }}:{{ .Port }}{{ end }}{{ end }}
+
+          # Below are tweaks which may only be suitable in dev environments
+          cluster.routing.allocation.disk.threshold_enabled: false
+      EOH
+        destination = "local/elasticsearch.yml"
+        change_mode = "noop"
+      }
+
+    }
+
+    // Sidecar for monitoring/admin etc..
+    // task "cerebro" {}
+  }
+}