[SPARK-21159][CORE] Don't try to connect to launcher in standalone cluster mode.

Monitoring for standalone cluster mode is not implemented (see SPARK-11033), but
the same scheduler implementation is used, and if it tries to connect to the
launcher it will fail. So fix the scheduler so it only tries that in client mode;
cluster mode applications will be correctly launched and will work, but monitoring
through the launcher handle will not be available.

Tested by running a cluster mode app with "SparkLauncher.startApplication".

Author: Marcelo Vanzin <[email protected]>

Closes apache#18397 from vanzin/SPARK-21159.

(cherry picked from commit bfd73a7)
Signed-off-by: Wenchen Fan <[email protected]>

Author: Marcelo Vanzin

core/src/main/scala/org/apache/spark/scheduler/cluster/StandaloneSchedulerBackend.scala

@@ -58,7 +58,13 @@ private[spark] class StandaloneSchedulerBackend(
 
   override def start() {
     super.start()
-    launcherBackend.connect()
+
+    // SPARK-21159. The scheduler backend should only try to connect to the launcher when in client
+    // mode. In cluster mode, the code that submits the application to the Master needs to connect
+    // to the launcher instead.
+    if (sc.deployMode == "client") {
+      launcherBackend.connect()
+    }
 
     // The endpoint for executors to talk to us
     val driverUrl = RpcEndpointAddress(