hadoop streaming 同一个命令 不同的表达方式报错

1.为何 hadoop streaming 命令  。bin/hadoop jar hadoop-streaming-1.1.2.jar  -input  sjtu   -output  shuchu  -mapper cat  -reducer wc   可以成功运行

bin/hadoop jar hadoop-streaming-1.1.2.jar  -input  sjtu   -output  shuchu  -mapper cat  -reducer wc  -l （有空格）  却会报错

bin/hadoop jar hadoop-streaming-1.1.2.jar  -input  sjtu   -output  shuchu  -mapper cat  -reducer wc -l (无空格） 运行一段时间才会报错

自己编写了两个脚本 sh1.sh (cat) sh2.sh(wc )   命令  bin/hadoop jar hadoop-streaming-1.1.2.jar  -input  sjtu   -output  sh  -mapper /home/wangmeng/sh1.sh  -reducer /home/wangmeng/sh2.sh  却会在最后报错 

13/07/21 22:40:42 INFO streaming.StreamJob:  map 0%  reduce 0%
13/07/21 22:41:12 INFO streaming.StreamJob:  map 100%  reduce 100%
13/07/21 22:41:12 INFO streaming.StreamJob: To kill this job, run:
13/07/21 22:41:12 INFO streaming.StreamJob: /home/wangmeng/hadoop-1.1.2/libexec/../bin/hadoop job  -Dmapred.job.tracker=master:9001 -kill job_201307212156_0013
13/07/21 22:41:12 INFO streaming.StreamJob: Tracking URL: http://master:50030/jobdetails.jsp?jobid=job_201307212156_0013
13/07/21 22:41:12 ERROR streaming.StreamJob: Job not successful. Error: # of failed Map Tasks exceeded allowed limit. FailedCount: 1. LastFailedTask: task_201307212156_0013_m_000000
13/07/21 22:41:12 INFO streaming.StreamJob: killJob...
Streaming Command Failed!