Remove post_launch bootstrap phase

With this commit we remove the unused bootstrap phase `post_launch`
which was intended to provide a car/plugin bootstrap phase after
Elasticsearch has been launched.

Relates elastic#481

esrally/mechanic/launcher.py

@@ -35,20 +35,17 @@ class ClusterLauncher:
     The cluster launcher performs cluster-wide tasks that need to be done in the startup / shutdown phase.
 
     """
-    def __init__(self, cfg, metrics_store, on_post_launch=None, client_factory_class=client.EsClientFactory):
+    def __init__(self, cfg, metrics_store, client_factory_class=client.EsClientFactory):
         """
 
         Creates a new ClusterLauncher.
 
         :param cfg: The config object.
         :param metrics_store: A metrics store that is configured to receive system metrics.
-        :param on_post_launch: An optional function that takes the Elasticsearch client as a parameter. It is invoked after the
-                               REST API is available.
         :param client_factory_class: A factory class that can create an Elasticsearch client.
         """
         self.cfg = cfg
         self.metrics_store = metrics_store
-        self.on_post_launch = on_post_launch
         self.client_factory = client_factory_class
 
     def start(self):
@@ -98,8 +95,6 @@ def start(self):
             logger.error("REST API layer is not yet available. Forcefully terminating cluster.")
             self.stop(c)
             raise exceptions.LaunchError("Elasticsearch REST API layer is not available. Forcefully terminated cluster.")
-        if self.on_post_launch:
-            self.on_post_launch(es_default)
         return c
 
     def stop(self, c):

esrally/mechanic/mechanic.py

@@ -360,9 +360,7 @@ def receiveMsg_NodesStopped(self, msg, sender):
         self.transition_when_all_children_responded(sender, msg, "cluster_stopping", "cluster_stopped", self.on_all_nodes_stopped)
 
     def on_all_nodes_started(self):
-        # we might not have a car if we benchmark external clusters
-        post_launch_handler = PostLaunchHandler([self.car]) if self.car else None
-        self.cluster_launcher = launcher.ClusterLauncher(self.cfg, self.metrics_store, on_post_launch=post_launch_handler)
+        self.cluster_launcher = launcher.ClusterLauncher(self.cfg, self.metrics_store)
         # Workaround because we could raise a LaunchError here and thespian will attempt to retry a failed message.
         # In that case, we will get a followup RallyAssertionError because on the second attempt, Rally will check
         # the status which is now "nodes_started" but we expected the status to be "nodes_starting" previously.
@@ -410,21 +408,6 @@ def on_all_nodes_stopped(self):
         # do not self-terminate, let the parent actor handle this
 
 
-class PostLaunchHandler:
-    def __init__(self, components, hook_handler_class=team.BootstrapHookHandler):
-        self.handlers = []
-        if components:
-            for component in components:
-                handler = hook_handler_class(component)
-                if handler.can_load():
-                    handler.load()
-                    self.handlers.append(handler)
-
-    def __call__(self, client):
-        for handler in self.handlers:
-            handler.invoke(team.BootstrapPhase.post_launch.name, client=client)
-
-
 @thespian.actors.requireCapability('coordinator')
 class Dispatcher(thespian.actors.ActorTypeDispatcher):
     def __init__(self):

esrally/mechanic/team.py

@@ -416,7 +416,6 @@ def __eq__(self, other):
 
 class BootstrapPhase(Enum):
     post_install = 10
-    post_launch = 20
 
     @classmethod
     def valid(cls, name):

tests/mechanic/launcher_test.py

@@ -110,25 +110,7 @@ class ClusterLauncherTests(TestCase):
     test_host = opts.TargetHosts("10.0.0.10:9200,10.0.0.11:9200")
     client_options = opts.ClientOptions('timeout:60')
 
-    def test_launches_cluster_with_post_launch_handler(self):
-        on_post_launch = mock.Mock()
-        cfg = config.Config()
-        cfg.add(config.Scope.application, "client", "hosts", self.test_host)
-        cfg.add(config.Scope.application, "client", "options", self.client_options)
-        cfg.add(config.Scope.application, "mechanic", "telemetry.devices", [])
-        cfg.add(config.Scope.application, "mechanic", "telemetry.params", {})
-
-        cluster_launcher = launcher.ClusterLauncher(cfg, MockMetricsStore(),
-                                                    on_post_launch=on_post_launch, client_factory_class=MockClientFactory)
-        cluster = cluster_launcher.start()
-
-        self.assertEqual([{"host": "10.0.0.10", "port":9200}, {"host": "10.0.0.11", "port":9200}], cluster.hosts)
-        self.assertIsNotNone(cluster.telemetry)
-        # this requires at least Python 3.6
-        # on_post_launch.assert_called_once()
-        self.assertEqual(1, on_post_launch.call_count)
-
-    def test_launches_cluster_without_post_launch_handler(self):
+    def test_launches_cluster(self):
         cfg = config.Config()
         cfg.add(config.Scope.application, "client", "hosts", self.test_host)
         cfg.add(config.Scope.application, "client", "options", self.client_options)
@@ -161,18 +143,14 @@ def test_launches_cluster_with_telemetry_client_timeout_enabled(self):
 
     @mock.patch("time.sleep")
     def test_error_on_cluster_launch(self, sleep):
-        on_post_launch = mock.Mock()
         cfg = config.Config()
         cfg.add(config.Scope.application, "client", "hosts", self.test_host)
         # Simulate that the client will raise an error upon startup
         cfg.add(config.Scope.application, "client", "options", opts.ClientOptions("raise-error-on-info:true"))
-        #cfg.add(config.Scope.application, "client", "options", {"raise-error-on-info": True})
         cfg.add(config.Scope.application, "mechanic", "telemetry.devices", [])
         cfg.add(config.Scope.application, "mechanic", "telemetry.params", {})
 
-        cluster_launcher = launcher.ClusterLauncher(cfg, MockMetricsStore(),
-                                                    on_post_launch=on_post_launch, client_factory_class=MockClientFactory)
+        cluster_launcher = launcher.ClusterLauncher(cfg, MockMetricsStore(), client_factory_class=MockClientFactory)
         with self.assertRaisesRegex(exceptions.LaunchError,
                                     "Elasticsearch REST API layer is not available. Forcefully terminated cluster."):
             cluster_launcher.start()
-        self.assertEqual(0, on_post_launch.call_count)

tests/mechanic/team_test.py

@@ -223,5 +223,5 @@ def test_cannot_register_for_unknown_phase(self):
         handler.loader.registration_function = hook
         with self.assertRaises(exceptions.SystemSetupError) as ctx:
             handler.load()
-        self.assertEqual("Unknown bootstrap phase [this_is_an_unknown_install_phase]. Valid phases are: ['post_install', 'post_launch'].",
+        self.assertEqual("Unknown bootstrap phase [this_is_an_unknown_install_phase]. Valid phases are: ['post_install'].",
                          ctx.exception.args[0])