HADOOP-261. Record an error message when map outputs are lost. Contributed by Owen.

git-svn-id: https://svn.apache.org/repos/asf/lucene/hadoop/trunk@448334 13f79535-47bb-0310-9956-ffa450edef68

CHANGES.txt

@@ -3,35 +3,38 @@ Hadoop Change Log
 
 Trunk (unreleased changes)
 
-1. HADOOP-243.  Fix rounding in the display of task and job progress
-   so that things are not shown to be 100% complete until they are in
-   fact finished.  (omalley via cutting) 
+ 1. HADOOP-243.  Fix rounding in the display of task and job progress
+    so that things are not shown to be 100% complete until they are in
+    fact finished.  (omalley via cutting) 
 
-2. HADOOP-438.  Limit the length of absolute paths in DFS, since the
-   file format used to store pathnames has some limitations.
-   (Wendy Chien via cutting)
+ 2. HADOOP-438.  Limit the length of absolute paths in DFS, since the
+    file format used to store pathnames has some limitations.
+    (Wendy Chien via cutting)
+
+ 3. HADOOP-530.  Improve error messages in SequenceFile when keys or
+    values are of the wrong type.  (Hairong Kuang via cutting)
 
-3. HADOOP-530.  Improve error messages in SequenceFile when keys or
-   values are of the wrong type.  (Hairong Kuang via cutting)
+ 4. HADOOP-288.  Add a file caching system and use it in MapReduce to
+    cache job jar files on slave nodes.  (Mahadev Konar via cutting)
 
-4. HADOOP-288.  Add a file caching system and use it in MapReduce to
-   cache job jar files on slave nodes.  (Mahadev Konar via cutting)
+ 5. HADOOP-533.  Fix unit test to not modify conf directory.
+   (Hairong Kuang via cutting)
 
-5. HADOOP-533.  Fix unit test to not modify conf directory.
-  (Hairong Kuang via cutting)
+ 6. HADOOP-527.  Permit specification of the local address that various
+    Hadoop daemons should bind to.  (Philippe Gassmann via cutting)
 
-6. HADOOP-527.  Permit specification of the local address that various
-   Hadoop daemons should bind to.  (Philippe Gassmann via cutting)
+ 7. HADOOP-542.  Updates to contrib/streaming: reformatted source code,
+    on-the-fly merge sort, a fix for HADOOP-540, etc.
+    (Michel Tourn via cutting)
 
-7. HADOOP-542.  Updates to contrib/streaming: reformatted source code,
-   on-the-fly merge sort, a fix for HADOOP-540, etc.
-   (Michel Tourn via cutting)
+ 8. HADOOP-545.  Remove an unused config file parameter.
+    (Philippe Gassmann via cutting)
 
-8. HADOOP-545.  Remove an unused config file parameter.
-   (Philippe Gassmann via cutting)
+ 9. HADOOP-548.  Add an Ant property "test.output" to build.xml that
+    causes test output to be logged to the console.  (omalley via cutting)
 
-9. HADOOP-548.  Add an Ant property "test.output" to build.xml that
-   causes test output to be logged to the console.  (omalley via cutting)
+10. HADOOP-261.  Record an error message when map output is lost.
+    (omalley via cutting)
 
 
 Release 0.6.2 (unreleased)

src/java/org/apache/hadoop/mapred/TaskTracker.java

@@ -357,7 +357,7 @@ public class TaskTracker
       this.mapOutputFile = new MapOutputFile();
       this.mapOutputFile.setConf(conf);
       int httpPort = conf.getInt("tasktracker.http.port", 50060);
-      String httpBindAddress = conf.get("tasktracker.http.bindAddress", "0.0.0.0");;
+      String httpBindAddress = conf.get("tasktracker.http.bindAddress", "0.0.0.0");
       this.server = new StatusHttpServer("task", httpBindAddress, httpPort, true);
       int workerThreads = conf.getInt("tasktracker.http.threads", 40);
       server.setThreads(1, workerThreads);
@@ -990,11 +990,14 @@ public class TaskTracker
         /**
          * The map output has been lost.
          */
-        public synchronized void mapOutputLost() throws IOException {
+        public synchronized void mapOutputLost(String failure
+                                               ) throws IOException {
             if (runstate == TaskStatus.SUCCEEDED) {
               LOG.info("Reporting output lost:"+task.getTaskId());
               runstate = TaskStatus.FAILED;       // change status to failure
               progress = 0.0f;
+              reportDiagnosticInfo("Map output lost, rescheduling: " + 
+                                   failure);
               runningTasks.put(task.getTaskId(), this);
               mapTotal++;
             } else {
@@ -1120,10 +1123,11 @@ public class TaskTracker
     /**
      * A completed map task's output has been lost.
      */
-    public synchronized void mapOutputLost(String taskid) throws IOException {
+    public synchronized void mapOutputLost(String taskid,
+                                           String errorMsg) throws IOException {
         TaskInProgress tip = (TaskInProgress) tasks.get(taskid);
         if (tip != null) {
-          tip.mapOutputLost();
+          tip.mapOutputLost(errorMsg);
         } else {
           LOG.warn("Unknown child with bad map output: "+taskid+". Ignored.");
         }

src/webapps/task/getMapOutput.jsp

@@ -40,9 +40,10 @@
     TaskTracker tracker = 
        (TaskTracker) application.getAttribute("task.tracker");
     Log log = (Log) application.getAttribute("log");
-    log.warn("Http server (getMapOutput.jsp): " +
-                StringUtils.stringifyException(ie));
-    tracker.mapOutputLost(mapId);
+    String errorMsg = "getMapOutput(" + mapId + "," + reduceId + ") failed : "+
+                       StringUtils.stringifyException(ie);
+    log.warn(errorMsg);
+    tracker.mapOutputLost(mapId, errorMsg);
     throw ie;
   } finally {
     if (inStream != null) {