From 2ba90c93d71aa2d30ee9ed431750c10c685e5599 Mon Sep 17 00:00:00 2001
From: Akira Ajisaka <aajisaka@apache.org>
Date: Fri, 21 Aug 2015 10:41:54 +0900
Subject: [PATCH] MAPREDUCE-6357. MultipleOutputs.write() API should document
 that output committing is not utilized when input path is absolute.
 Contributed by Dustin Cote.

---
 hadoop-mapreduce-project/CHANGES.txt               |  4 ++++
 .../mapreduce/lib/output/MultipleOutputs.java      | 14 +++++++++++++-
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/hadoop-mapreduce-project/CHANGES.txt b/hadoop-mapreduce-project/CHANGES.txt
index 43d0fafb77..361a19b695 100644
--- a/hadoop-mapreduce-project/CHANGES.txt
+++ b/hadoop-mapreduce-project/CHANGES.txt
@@ -555,6 +555,10 @@ Release 2.8.0 - UNRELEASED
     MAPREDUCE-5817. Mappers get rescheduled on node transition even after all
     reducers are completed. (Sangjin Lee via kasha)
 
+    MAPREDUCE-6357. MultipleOutputs.write() API should document that output
+    committing is not utilized when input path is absolute.
+    (Dustin Cote via aajisaka)
+
 Release 2.7.2 - UNRELEASED
 
   INCOMPATIBLE CHANGES
diff --git a/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/output/MultipleOutputs.java b/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/output/MultipleOutputs.java
index c31cab7ddd..a3a0e76ab8 100644
--- a/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/output/MultipleOutputs.java
+++ b/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/output/MultipleOutputs.java
@@ -120,7 +120,11 @@
  * 
  * <p>
  * Use <code>MultipleOutputs.write(KEYOUT key, VALUEOUT value, String baseOutputPath)</code> to write key and 
- * value to a path specified by <code>baseOutputPath</code>, with no need to specify a named output:
+ * value to a path specified by <code>baseOutputPath</code>, with no need to specify a named output.
+ * <b>Warning</b>: when the baseOutputPath passed to MultipleOutputs.write
+ * is a path that resolves outside of the final job output directory, the
+ * directory is created immediately and then persists through subsequent
+ * task retries, breaking the concept of output committing:
  * </p>
  * 
  * <pre>
@@ -418,6 +422,10 @@ public <K, V> void write(String namedOutput, K key, V value)
    * @param value          the value
    * @param baseOutputPath base-output path to write the record to.
    * Note: Framework will generate unique filename for the baseOutputPath
+   * <b>Warning</b>: when the baseOutputPath is a path that resolves
+   * outside of the final job output directory, the directory is created
+   * immediately and then persists through subsequent task retries, breaking
+   * the concept of output committing.
    */
   @SuppressWarnings("unchecked")
   public <K, V> void write(String namedOutput, K key, V value,
@@ -442,6 +450,10 @@ public <K, V> void write(String namedOutput, K key, V value,
    * @param value     the value
    * @param baseOutputPath base-output path to write the record to.
    * Note: Framework will generate unique filename for the baseOutputPath
+   * <b>Warning</b>: when the baseOutputPath is a path that resolves
+   * outside of the final job output directory, the directory is created
+   * immediately and then persists through subsequent task retries, breaking
+   * the concept of output committing.
    */
   @SuppressWarnings("unchecked")
   public void write(KEYOUT key, VALUEOUT value, String baseOutputPath)