Merge pull request #2904 from hashicorp/non-host-id

Adds a new -disable-host-node-id option to help when testing with containers.

command/agent/agent.go

@@ -630,6 +630,11 @@ func (a *Agent) makeRandomID() (string, error) {
 // high for us if this changes, so we will persist it either way. This will let
 // gopsutil change implementations without affecting in-place upgrades of nodes.
 func (a *Agent) makeNodeID() (string, error) {
+	// If they've disabled host-based IDs then just make a random one.
+	if a.config.DisableHostNodeID {
+		return a.makeRandomID()
+	}
+
 	// Try to get a stable ID associated with the host itself.
 	info, err := host.Info()
 	if err != nil {

command/agent/agent_test.go

@@ -318,7 +318,7 @@ func TestAgent_ReconnectConfigSettings(t *testing.T) {
 	}()
 }
 
-func TestAgent_NodeID(t *testing.T) {
+func TestAgent_setupNodeID(t *testing.T) {
 	c := nextConfig()
 	c.NodeID = ""
 	dir, agent := makeAgent(t, c)
@@ -384,6 +384,42 @@ func TestAgent_NodeID(t *testing.T) {
 	}
 }
 
+func TestAgent_makeNodeID(t *testing.T) {
+	c := nextConfig()
+	c.NodeID = ""
+	dir, agent := makeAgent(t, c)
+	defer os.RemoveAll(dir)
+	defer agent.Shutdown()
+
+	// We should get a valid host-based ID initially.
+	id, err := agent.makeNodeID()
+	if err != nil {
+		t.Fatalf("err: %v", err)
+	}
+	if _, err := uuid.ParseUUID(string(id)); err != nil {
+		t.Fatalf("err: %v", err)
+	}
+
+	// Calling again should yield the same ID since it's host-based.
+	another, err := agent.makeNodeID()
+	if err != nil {
+		t.Fatalf("err: %v", err)
+	}
+	if id != another {
+		t.Fatalf("bad: %s vs %s", id, another)
+	}
+
+	// Turn off host-based IDs and try again. We should get a random ID.
+	agent.config.DisableHostNodeID = true
+	another, err = agent.makeNodeID()
+	if err != nil {
+		t.Fatalf("err: %v", err)
+	}
+	if id == another {
+		t.Fatalf("bad: %s vs %s", id, another)
+	}
+}
+
 func TestAgent_AddService(t *testing.T) {
 	dir, agent := makeAgent(t, nextConfig())
 	defer os.RemoveAll(dir)

command/agent/command.go

@@ -100,6 +100,10 @@ func (c *Command) readConfig() *Config {
 	f.StringVar((*string)(&cmdConfig.NodeID), "node-id", "",
 		"A unique ID for this node across space and time. Defaults to a randomly-generated ID"+
 			" that persists in the data-dir.")
+	f.BoolVar(&cmdConfig.DisableHostNodeID, "disable-host-node-id", false,
+		"Setting this to true will prevent Consul from using information from the"+
+			" host to generate a node ID, and will cause Consul to generate a"+
+			" random node ID instead.")
 	f.StringVar(&dcDeprecated, "dc", "", "Datacenter of the agent (deprecated: use 'datacenter' instead).")
 	f.StringVar(&cmdConfig.Datacenter, "datacenter", "", "Datacenter of the agent.")
 	f.StringVar(&cmdConfig.DataDir, "data-dir", "", "Path to a data directory to store agent state.")

command/agent/config.go

@@ -358,6 +358,11 @@ type Config struct {
 	// to a randomly-generated ID that persists in the data-dir.
 	NodeID types.NodeID `mapstructure:"node_id"`
 
+	// DisableHostNodeID will prevent Consul from using information from the
+	// host to generate a node ID, and will cause Consul to generate a
+	// random ID instead.
+	DisableHostNodeID bool `mapstructure:"disable_host_node_id"`
+
 	// Node name is the name we use to advertise. Defaults to hostname.
 	NodeName string `mapstructure:"node_name"`
 
@@ -1371,6 +1376,9 @@ func MergeConfig(a, b *Config) *Config {
 	if b.NodeID != "" {
 		result.NodeID = b.NodeID
 	}
+	if b.DisableHostNodeID == true {
+		result.DisableHostNodeID = b.DisableHostNodeID
+	}
 	if b.NodeName != "" {
 		result.NodeName = b.NodeName
 	}

command/agent/config_test.go

@@ -59,8 +59,8 @@ func TestDecodeConfig(t *testing.T) {
 		t.Fatalf("bad: %#v", config)
 	}
 
-	// Without a protocol
+	// Node info
-	input = `{"node_id": "bar", "node_name": "foo", "datacenter": "dc2"}`
+	input = `{"node_id": "bar", "disable_host_node_id": true, "node_name": "foo", "datacenter": "dc2"}`
 	config, err = DecodeConfig(bytes.NewReader([]byte(input)))
 	if err != nil {
 		t.Fatalf("err: %s", err)
@@ -74,6 +74,10 @@ func TestDecodeConfig(t *testing.T) {
 		t.Fatalf("bad: %#v", config)
 	}
 
+	if config.DisableHostNodeID != true {
+		t.Fatalf("bad: %#v", config)
+	}
+
 	if config.Datacenter != "dc2" {
 		t.Fatalf("bad: %#v", config)
 	}
@@ -1652,14 +1656,15 @@ func TestMergeConfig(t *testing.T) {
 			UDPAnswerLimit:  4,
 			RecursorTimeout: 30 * time.Second,
 		},
-		Domain:           "other",
+		Domain:            "other",
-		LogLevel:         "info",
+		LogLevel:          "info",
-		NodeID:           "bar",
+		NodeID:            "bar",
-		NodeName:         "baz",
+		DisableHostNodeID: true,
-		ClientAddr:       "127.0.0.2",
+		NodeName:          "baz",
-		BindAddr:         "127.0.0.2",
+		ClientAddr:        "127.0.0.2",
-		AdvertiseAddr:    "127.0.0.2",
+		BindAddr:          "127.0.0.2",
-		AdvertiseAddrWan: "127.0.0.2",
+		AdvertiseAddr:     "127.0.0.2",
+		AdvertiseAddrWan:  "127.0.0.2",
 		Ports: PortConfig{
 			DNS:     1,
 			HTTP:    2,

website/source/docs/agent/options.html.markdown

@@ -151,17 +151,22 @@ will exit with an error at startup.
   the use of filesystem locking, meaning some types of mounted folders (e.g. VirtualBox
   shared folders) may not be suitable.
 
+* <a name="_datacenter"></a><a href="#_datacenter">`-datacenter`</a> - This flag controls the datacenter in
+  which the agent is running. If not provided,
+  it defaults to "dc1". Consul has first-class support for multiple datacenters, but
+  it relies on proper configuration. Nodes in the same datacenter should be on a single
+  LAN.
+
 * <a name="_dev"></a><a href="#_dev">`-dev`</a> - Enable development server
   mode. This is useful for quickly starting a Consul agent with all persistence
   options turned off, enabling an in-memory server which can be used for rapid
   prototyping or developing against the API. This mode is **not** intended for
   production use as it does not write any data to disk.
 
-* <a name="_datacenter"></a><a href="#_datacenter">`-datacenter`</a> - This flag controls the datacenter in
+* <a name="_disable_host_node_id"></a><a href="#_disable_host_node_id">`-disable-host-node-id`</a> - Setting
-  which the agent is running. If not provided,
+  this to true will prevent Consul from using information from the host to generate a deterministic node ID,
-  it defaults to "dc1". Consul has first-class support for multiple datacenters, but
+  and will instead generate a random node ID which will be persisted in the data directory. This is useful
-  it relies on proper configuration. Nodes in the same datacenter should be on a single
+  when running multiple Consul agents on the same host for testing. This defaults to false.
-  LAN.
 
 * <a name="_dns_port"></a><a href="#_dns_port">`-dns-port`</a> - the DNS port to listen on.
   This overrides the default port 8600. This is available in Consul 0.7 and later.
@@ -295,11 +300,9 @@ will exit with an error at startup.
   changes. This must be in the form of a hex string, 36 characters long, such as
   `adf4238a-882b-9ddc-4a9d-5b6758e4159e`. If this isn't supplied, which is the most common case, then
   the agent will generate an identifier at startup and persist it in the <a href="#_data_dir">data directory</a>
-  so that it will remain the same across agent restarts. This is currently only exposed via
+  so that it will remain the same across agent restarts. Information from the host will be used to
-  <a href="/api/agent.html#agent_self">/v1/agent/self</a>,
+  generate a deterministic node ID if possible, unless [`-disable-host-node-id`](#_disable_host_node_id) is
-  <a href="/api/catalog.html">/v1/catalog</a>, and
+  set to true.
-  <a href="/api/health.html">/v1/health</a> endpoints, but future versions of
-  Consul will use this to better manage cluster changes, especially for Consul servers.
 
 * <a name="_node_meta"></a><a href="#_node_meta">`-node-meta`</a> - Available in Consul 0.7.3 and later,
   this specifies an arbitrary metadata key/value pair to associate with the node, of the form `key:value`.
@@ -636,6 +639,9 @@ Consul will not enable TLS for the HTTP API unless the `https` port has been ass
   `disable_anonymous_signature`</a> Disables providing an anonymous signature for de-duplication
   with the update check. See [`disable_update_check`](#disable_update_check).
 
+* <a name="disable_host_node_id"></a><a href="#disable_host_node_id">`disable_host_node_id`</a>
+  Equivalent to the [`-disable-host-node-id` command-line flag](#_disable_host_node_id).
+
 * <a name="disable_remote_exec"></a><a href="#disable_remote_exec">`disable_remote_exec`</a>
   Disables support for remote execution. When set to true, the agent will ignore any incoming
   remote exec requests. In versions of Consul prior to 0.8, this defaulted to false. In Consul