[SPARK-2313] Use socket to communicate GatewayServer port back to Python driver

This patch changes PySpark so that the GatewayServer's port is communicated back to the Python process that launches it over a local socket instead of a pipe.  The old pipe-based approach was brittle and could fail if `spark-submit` printed unexpected to stdout.

To accomplish this, I wrote a custom `PythonGatewayServer.main()` function to use in place of Py4J's `GatewayServer.main()`.

Closes #3424.

Author: Josh Rosen <joshrosen@databricks.com>

Closes #4603 from JoshRosen/SPARK-2313 and squashes the following commits:

6a7740b [Josh Rosen] Remove EchoOutputThread since it's no longer needed
0db501f [Josh Rosen] Use select() so that we don't block if GatewayServer dies.
9bdb4b6 [Josh Rosen] Handle case where getListeningPort returns -1
3fb7ed1 [Josh Rosen] Remove stdout=PIPE
2458934 [Josh Rosen] Use underscore to mark env var. as private
d12c95d [Josh Rosen] Use Logging and Utils.tryOrExit()
e5f9730 [Josh Rosen] Wrap everything in a giant try-block
2f70689 [Josh Rosen] Use stdin PIPE to share fate with driver
8bf956e [Josh Rosen] Initial cut at passing Py4J gateway port back to driver via socket

2 files changed, 65 insertions, 3 deletions

diff --git a/core/src/main/scala/org/apache/spark/api/python/PythonGatewayServer.scala b/core/src/main/scala/org/apache/spark/api/python/PythonGatewayServer.scala
new file mode 100644
index 0000000000..164e950815
--- /dev/null
+++ b/core/src/main/scala/org/apache/spark/api/python/PythonGatewayServer.scala
@@ -0,0 +1,64 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.api.python
+
+import java.io.DataOutputStream
+import java.net.Socket
+
+import py4j.GatewayServer
+
+import org.apache.spark.Logging
+import org.apache.spark.util.Utils
+
+/**
+ * Process that starts a Py4J GatewayServer on an ephemeral port and communicates the bound port
+ * back to its caller via a callback port specified by the caller.
+ *
+ * This process is launched (via SparkSubmit) by the PySpark driver (see java_gateway.py).
+ */
+private[spark] object PythonGatewayServer extends Logging {
+  def main(args: Array[String]): Unit = Utils.tryOrExit {
+    // Start a GatewayServer on an ephemeral port
+    val gatewayServer: GatewayServer = new GatewayServer(null, 0)
+    gatewayServer.start()
+    val boundPort: Int = gatewayServer.getListeningPort
+    if (boundPort == -1) {
+      logError("GatewayServer failed to bind; exiting")
+      System.exit(1)
+    } else {
+      logDebug(s"Started PythonGatewayServer on port $boundPort")
+    }
+
+    // Communicate the bound port back to the caller via the caller-specified callback port
+    val callbackHost = sys.env("_PYSPARK_DRIVER_CALLBACK_HOST")
+    val callbackPort = sys.env("_PYSPARK_DRIVER_CALLBACK_PORT").toInt
+    logDebug(s"Communicating GatewayServer port to Python driver at $callbackHost:$callbackPort")
+    val callbackSocket = new Socket(callbackHost, callbackPort)
+    val dos = new DataOutputStream(callbackSocket.getOutputStream)
+    dos.writeInt(boundPort)
+    dos.close()
+    callbackSocket.close()
+
+    // Exit on EOF or broken pipe to ensure that this process dies when the Python driver dies:
+    while (System.in.read() != -1) {
+      // Do nothing
+    }
+    logDebug("Exiting due to broken pipe from Python driver")
+    System.exit(0)
+  }
+}
diff --git a/core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala b/core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala
index 54399e99c9..012a89a31b 100644
--- a/core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala
+++ b/core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala
@@ -39,7 +39,6 @@ import org.apache.ivy.plugins.resolver.{ChainResolver, IBiblioResolver}
 
 import org.apache.spark.SPARK_VERSION
 import org.apache.spark.deploy.rest._
-import org.apache.spark.executor._
 import org.apache.spark.util.{ChildFirstURLClassLoader, MutableURLClassLoader, Utils}
 
 /**
@@ -284,8 +283,7 @@ object SparkSubmit {
     // If we're running a python app, set the main class to our specific python runner
     if (args.isPython && deployMode == CLIENT) {
       if (args.primaryResource == PYSPARK_SHELL) {
-        args.mainClass = "py4j.GatewayServer"
-        args.childArgs = ArrayBuffer("--die-on-broken-pipe", "0")
+        args.mainClass = "org.apache.spark.api.python.PythonGatewayServer"
       } else {
         // If a python file is provided, add it to the child arguments and list of files to deploy.
         // Usage: PythonAppRunner <main python file> <extra python files> [app arguments]