diff --git a/examples/hadoop_word_count/README.txt b/examples/hadoop_word_count/README.txt
index 1ec13f7881..cf8a34469e 100644
--- a/examples/hadoop_word_count/README.txt
+++ b/examples/hadoop_word_count/README.txt
@@ -1,9 +1,16 @@
+Introduction
+============
+
 WordCount hadoop example: Inserts a bunch of words across multiple rows,
 and counts them, with RandomPartitioner. The word_count_counters example sums
 the value of counter columns for a key.
 
 The scripts in bin/ assume you are running with cwd of contrib/word_count.
 
+
+Running
+=======
+
 First build and start a Cassandra server with the default configuration*, 
 then run
 
@@ -32,3 +39,9 @@ is written to a text file in /tmp/word_count_counters.
 
 *If you want to point wordcount at a real cluster, modify the seed
 and listenaddress settings accordingly.
+
+
+Troubleshooting
+===============
+
+word_count uses conf/log4j.properties to log to wc.out.
diff --git a/examples/hadoop_word_count/bin/word_count b/examples/hadoop_word_count/bin/word_count
index 58f325e28c..a0c5aa02f1 100755
--- a/examples/hadoop_word_count/bin/word_count
+++ b/examples/hadoop_word_count/bin/word_count
@@ -30,6 +30,7 @@ if [ ! -e $cwd/../build/word_count.jar ]; then
     exit 1
 fi
 
+CLASSPATH=$CLASSPATH:$cwd/../conf
 CLASSPATH=$CLASSPATH:$cwd/../build/word_count.jar
 CLASSPATH=$CLASSPATH:$cwd/../../../build/classes/main
 CLASSPATH=$CLASSPATH:$cwd/../../../build/classes/thrift
diff --git a/examples/hadoop_word_count/conf/log4j.properties b/examples/hadoop_word_count/conf/log4j.properties
new file mode 100644
index 0000000000..070d21e33c
--- /dev/null
+++ b/examples/hadoop_word_count/conf/log4j.properties
@@ -0,0 +1,15 @@
+log4j.rootLogger=DEBUG,stdout,F
+
+#stdout
+log4j.appender.stdout=org.apache.log4j.ConsoleAppender
+log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
+log4j.appender.stdout.layout.ConversionPattern=%5p %d{HH:mm:ss,SSS} %m%n
+
+# log file
+log4j.appender.F=org.apache.log4j.FileAppender
+log4j.appender.F.Append=false
+log4j.appender.F.layout=org.apache.log4j.PatternLayout
+log4j.appender.F.layout.ConversionPattern=%5p [%t] %d{ISO8601} %F (line %L) %m%n
+# Edit the next line to point to your logs directory
+log4j.appender.F.File=wc.out
+
diff --git a/examples/hadoop_word_count/src/WordCount.java b/examples/hadoop_word_count/src/WordCount.java
index d3cee0e72a..96bcb1be27 100644
--- a/examples/hadoop_word_count/src/WordCount.java
+++ b/examples/hadoop_word_count/src/WordCount.java
@@ -93,7 +93,8 @@ public class WordCount extends Configured implements Tool
                 else
                     value = ByteBufferUtil.string(column.value());
                                
-                System.err.println("read " + ByteBufferUtil.string(key) + ":" +name + ":" + value + " from " + context.getInputSplit());
+                logger.debug("read {}:{}={} from {}",
+                             new Object[] {ByteBufferUtil.string(key), name, value, context.getInputSplit()});
 
                 StringTokenizer itr = new StringTokenizer(value);
                 while (itr.hasMoreTokens())