Question

我正在尝试在单节点设置上运行单词计数教程 http://hadoop.apache.org/docs/stable/mapred_tutorial.html

这是我的终端输出：

> hadoop jar wordcount.jar org.myorg.WordCount input output
13/08/13 16:26:59 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
13/08/13 16:26:59 WARN mapred.JobClient: Use GenericOptionsParser for parsing the arguments. Applications should implement Tool for the same.
13/08/13 16:26:59 WARN snappy.LoadSnappy: Snappy native library not loaded
13/08/13 16:26:59 INFO mapred.FileInputFormat: Total input paths to process : 2
13/08/13 16:26:59 INFO mapred.JobClient: Running job: job_local955318185_0001
13/08/13 16:26:59 INFO mapred.LocalJobRunner: Waiting for map tasks
13/08/13 16:26:59 INFO mapred.LocalJobRunner: Starting task: attempt_local955318185_0001_m_000000_0
13/08/13 16:26:59 INFO mapred.Task:  Using ResourceCalculatorPlugin : null
13/08/13 16:26:59 INFO mapred.MapTask: Processing split: file:/Users/jfk/work/hadoop/2_word/input/file02:0+24
13/08/13 16:26:59 INFO mapred.MapTask: numReduceTasks: 1
13/08/13 16:26:59 INFO mapred.MapTask: io.sort.mb = 100
13/08/13 16:27:00 INFO mapred.MapTask: data buffer = 79691776/99614720
13/08/13 16:27:00 INFO mapred.MapTask: record buffer = 262144/327680
13/08/13 16:27:00 INFO mapred.MapTask: Starting flush of map output
13/08/13 16:27:00 INFO mapred.JobClient:  map 0% reduce 0%
13/08/13 16:27:05 INFO mapred.LocalJobRunner: file:/Users/jfk/work/hadoop/2_word/input/file02:0+24
13/08/13 16:27:06 INFO mapred.JobClient:  map 50% reduce 0%

它只是卡在那里。我所能做的就是Ctrl-C。我该怎么调试呢？

以下是logs/userlogs

的内容

-> ls -l
total 0
drwx--x---  16 jfk  admin  544 Aug  7 21:51 job_201308072147_0001
drwx--x---  16 jfk  admin  544 Aug  9 10:18 job_201308091015_0001
drwx--x---   9 jfk  admin  306 Aug 13 14:59 job_201308131457_0001
drwx--x---   7 jfk  admin  238 Aug 13 14:59 job_201308131457_0002
drwx--x---   9 jfk  admin  306 Aug 13 15:02 job_201308131457_0003
drwx--x---   9 jfk  admin  306 Aug 13 15:04 job_201308131457_0005
drwx--x---   9 jfk  admin  306 Aug 13 15:13 job_201308131457_0007
drwx--x---   9 jfk  admin  306 Aug 13 15:14 job_201308131457_0009
drwx--x---   9 jfk  admin  306 Aug 13 15:15 job_201308131457_0011
drwx--x---   7 jfk  admin  238 Aug 13 15:16 job_201308131457_0012
drwx--x---  15 jfk  admin  510 Aug 13 15:28 job_201308131457_0014
drwx--x---   7 jfk  admin  238 Aug 13 15:28 job_201308131457_0015
drwx--x---  15 jfk  admin  510 Aug 13 16:20 job_201308131549_0001
drwx--x---  11 jfk  admin  374 Aug 13 16:20 job_201308131549_0002
drwx--x---   4 jfk  admin  136 Aug 13 16:13 job_201308131549_0004

job_201308131549_0004/attempt_201308131549_0004_r_000002_0/stderr的内容：

2013-08-13 16:13:10.401 java[7378:1203] Unable to load realm info from SCDynamicStore

==更新==

在Google上搜索错误消息Unable to load realm info from SCDynamicStore时，似乎有几个人在OSX上使用Hadoop时遇到了同样的问题。以下解决方案似乎适用于某些人，但不幸的是不适合我。 Hadoop on OSX "Unable to load realm info from SCDynamicStore"

Answer 1

转到http://localhost:50030/jobtracker.jsp 和选择运行中卡住的作业，找到哪个地图卡住了。转到卡住的地图作业，然后单击右侧的所有链接，这将指向您确切的错误。不确定这是否有助于你

Hadoop地图卡在Word Count教程上 - 无法从SCDynamicStore加载领域信息

1 个答案: