Shell 问题记录_command substitution:line 12报错-优快云博客

本文链接：https://blog.youkuaiyun.com/cloudeagle_bupt/article/details/15028287

本文探讨了使用Shell脚本执行Hadoop任务时遇到的不同错误及其解决方案。通过四种不同的写法对比，展示了如何正确地在Shell脚本中调用Hadoop命令，并避免常见的语法错误。

实在不知道是啥原因，求达人指教.

写法一:

for i in seq `1 3`;
do hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &;
hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 & ;
hadoop fs -rmr /user/liuqiang/Pi*;
done

报错: ./mixed_jobs.sh: line 12: syntax error near unexpected token `;'
./mixed_jobs.sh: line 12: `do hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &;'

写法二:
for i in seq `1 3`;
do
"hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &;

hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 & ;

hadoop fs -rmr /user/liuqiang/Pi*";
done

报错:

./mixed_jobs.sh: line 13: 1: command not found
./mixed_jobs.sh: line 12: hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbenchseq.txt 2>&1 &;hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/piseq.txt 2>&1 & ; hadoop fs -rmr /user/liuqiang/Pi*: No such file or director

写法三:

for i in seq `1 3`;
do
`hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &;

hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 & ;

hadoop fs -rmr /user/liuqiang/Pi* `;
done

报错:

./mixed_jobs.sh: line 13: 1: command not found
./mixed_jobs.sh: command substitution: line 12: syntax error near unexpected token `;'
./mixed_jobs.sh: command substitution: line 12: `hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &;hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 & ; hadoop fs -rmr /user/liuqiang/Pi* '

写法四:

for i in seq `1 3`;
do
`hadoop jar hadoop-test-1.0.3.jar mrbench -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &

hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 &

hadoop fs -rmr /user/liuqiang/Pi* `;
done

报错:

./mixed_jobs.sh: line 13: 1: command not found

已解决: 版主给力啊,http://bbs.youkuaiyun.com/topics/390638470

for i in `seq 3`; do

hadoop jar hadoop-test-1.0.3.jar mrbench  -inputLines 80000 -maps 300 -reduces 30 >/opt/jobs/mrbench$i.txt 2>&1 &

hadoop jar hadoop-examples-1.0.3.jar pi 500 200000000 > /opt/jobs/pi$i.txt 2>&1 &

hadoop fs -rmr /user/liuqiang/Pi*

done