[SPARK-21551][Python] Increase timeout for PythonRDD.serveIterator

apache · Jul 27, 2017 · 9d3c664 · 9d3c664
1 parent 9f5647d
commit 9d3c664
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 4 deletions.
diff --git a/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala b/core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala
@@ -683,7 +683,7 @@ private[spark] object PythonRDD extends Logging {
    * Create a socket server and a background thread to serve the data in `items`,
    *
    * The socket server can only accept one connection, or close if no connection
-   * in 3 seconds.
+   * in 15 seconds.
    *
    * Once a connection comes in, it tries to serialize all the data in `items`
    * and send them into this connection.
@@ -692,8 +692,8 @@ private[spark] object PythonRDD extends Logging {
    */
   def serveIterator[T](items: Iterator[T], threadName: String): Int = {
     val serverSocket = new ServerSocket(0, 1, InetAddress.getByName("localhost"))
-    // Close the socket if no connection in 3 seconds
-    serverSocket.setSoTimeout(3000)
+    // Close the socket if no connection in 15 seconds
+    serverSocket.setSoTimeout(15000)
 
     new Thread(threadName) {
       setDaemon(true)

diff --git a/python/pyspark/rdd.py b/python/pyspark/rdd.py
@@ -127,7 +127,7 @@ def _load_from_socket(port, serializer):
         af, socktype, proto, canonname, sa = res
         sock = socket.socket(af, socktype, proto)
         try:
-            sock.settimeout(3)
+            sock.settimeout(15)
             sock.connect(sa)
         except socket.error:
             sock.close()