Hadoop学习之路（七）Hadoop集群shell常用命令

Hadoop常用命令

启动HDFS集群

[hadoop@hadoop1 ~]$ start-dfs.sh
Starting namenodes on [hadoop1]
hadoop1: starting namenode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-namenode-hadoop1.out
hadoop2: starting datanode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-datanode-hadoop2.out
hadoop3: starting datanode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-datanode-hadoop3.out
hadoop4: starting datanode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-datanode-hadoop4.out
hadoop1: starting datanode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-datanode-hadoop1.out
Starting secondary namenodes [hadoop3]
hadoop3: starting secondarynamenode, logging to /home/hadoop/apps/hadoop-2.7./logs/hadoop-hadoop-secondarynamenode-hadoop3.out
[hadoop@hadoop1 ~]$ 

启动YARN集群

[hadoop@hadoop4 ~]$ start-yarn.sh
starting yarn daemons
starting resourcemanager, logging to /home/hadoop/apps/hadoop-2.7./logs/yarn-hadoop-resourcemanager-hadoop4.out
hadoop2: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7./logs/yarn-hadoop-nodemanager-hadoop2.out
hadoop3: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7./logs/yarn-hadoop-nodemanager-hadoop3.out
hadoop4: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7./logs/yarn-hadoop-nodemanager-hadoop4.out
hadoop1: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7./logs/yarn-hadoop-nodemanager-hadoop1.out
[hadoop@hadoop4 ~]$ 

查看HDFS系统根目录

[hadoop@hadoop1 ~]$ hadoop fs -ls /
Found  items
drwxr-xr-x   - hadoop supergroup           -- : /test
drwx------   - hadoop supergroup           -- : /tmp
[hadoop@hadoop1 ~]$ 

创建文件夹

[hadoop@hadoop1 ~]$ hadoop fs -mkdir /a
[hadoop@hadoop1 ~]$ hadoop fs -ls /
Found  items
drwxr-xr-x   - hadoop supergroup           -- : /a
drwxr-xr-x   - hadoop supergroup           -- : /test
drwx------   - hadoop supergroup           -- : /tmp
[hadoop@hadoop1 ~]$ 

级联创建文件夹

[hadoop@hadoop1 ~]$ hadoop fs -mkdir -p /aa/bb/cc
[hadoop@hadoop1 ~]$ 

查看hsdf系统根目录下的所有文件包括子文件夹里面的文件

[hadoop@hadoop1 ~]$ hadoop fs -ls -R /aa
drwxr-xr-x - hadoop supergroup 0 2018-03-08 11:12 /aa/bb
drwxr-xr-x - hadoop supergroup 0 2018-03-08 11:12 /aa/bb/cc
[hadoop@hadoop1 ~]$

上传文件

[hadoop@hadoop1 ~]$ ls
apps data words.txt
[hadoop@hadoop1 ~]$ hadoop fs -put words.txt /aa
[hadoop@hadoop1 ~]$ hadoop fs -copyFromLocal words.txt /aa/bb
[hadoop@hadoop1 ~]$

下载文件

[hadoop@hadoop1 ~]$ hadoop fs -get /aa/words.txt ~/newwords.txt
[hadoop@hadoop1 ~]$ ls
apps  data  newwords.txt  words.txt
[hadoop@hadoop1 ~]$ hadoop fs -copyToLocal /aa/words.txt ~/newwords1.txt
[hadoop@hadoop1 ~]$ ls
apps  data  newwords1.txt  newwords.txt  words.txt
[hadoop@hadoop1 ~]$ 

合并下载

[hadoop@hadoop1 ~]$ hadoop fs -getmerge /aa/words.txt /aa/bb/words.txt ~/2words.txt
[hadoop@hadoop1 ~]$ ll
总用量
-rw-r--r--.  hadoop hadoop    3月    : 2words.txt
drwxrwxr-x.  hadoop hadoop  3月    : apps
drwxrwxr-x.  hadoop hadoop  3月    : data
-rw-r--r--.  hadoop hadoop    3月    : newwords1.txt
-rw-r--r--.  hadoop hadoop    3月    : newwords.txt
-rw-rw-r--.  hadoop hadoop    3月    : words.txt
[hadoop@hadoop1 ~]$ 

复制

从HDFS一个路径拷贝到HDFS另一个路径

[hadoop@hadoop1 ~]$ hadoop fs -ls /a
[hadoop@hadoop1 ~]$ hadoop fs -cp /aa/words.txt /a
[hadoop@hadoop1 ~]$ hadoop fs -ls /a
Found  items
-rw-r--r--    hadoop supergroup          -- : /a/words.txt
[hadoop@hadoop1 ~]$ 

移动

在HDFS目录中移动文件

[hadoop@hadoop1 ~]$ hadoop fs -ls /aa/bb/cc
[hadoop@hadoop1 ~]$ hadoop fs -mv /a/words.txt /aa/bb/cc
[hadoop@hadoop1 ~]$ hadoop fs -ls /aa/bb/cc
Found  items
-rw-r--r--    hadoop supergroup          -- : /aa/bb/cc/words.txt
[hadoop@hadoop1 ~]$ 

删除

删除文件或文件夹

[hadoop@hadoop1 ~]$ hadoop fs -rm /aa/bb/cc/words.txt
// :: INFO fs.TrashPolicyDefault: Namenode trash configuration: Deletion interval =  minutes, Emptier interval =  minutes.
Deleted /aa/bb/cc/words.txt
[hadoop@hadoop1 ~]$ hadoop fs -ls /aa/bb/cc
[hadoop@hadoop1 ~]$ 

删除空目录

[hadoop@hadoop1 ~]$ hadoop fs -rmdir /aa/bb/cc/
[hadoop@hadoop1 ~]$ hadoop fs -ls /aa/bb/
Found  items
-rw-r--r--    hadoop supergroup          -- : /aa/bb/words.txt
[hadoop@hadoop1 ~]$ 

强制删除

[hadoop@hadoop1 ~]$ hadoop fs -rm /aa/bb/
rm: `/aa/bb': Is a directory
[hadoop@hadoop1 ~]$ hadoop fs -rm -r /aa/bb/
// :: INFO fs.TrashPolicyDefault: Namenode trash configuration: Deletion interval =  minutes, Emptier interval =  minutes.
Deleted /aa/bb
[hadoop@hadoop1 ~]$ hadoop fs -ls /aa
Found  items
-rw-r--r--    hadoop supergroup          -- : /aa/words.txt
[hadoop@hadoop1 ~]$ 

从本地剪切文件到HDFS上

[hadoop@hadoop1 ~]$ ls
apps  data  hello.txt
[hadoop@hadoop1 ~]$ hadoop fs -moveFromLocal ~/hello.txt /aa
[hadoop@hadoop1 ~]$ ls
apps  data
[hadoop@hadoop1 ~]$ 

追加文件

追加之前hello.txt到words.txt之前

[hadoop@hadoop1 ~]$ hadoop fs -appendToFile ~/hello.txt /aa/words.txt
[hadoop@hadoop1 ~]$ 

追加之前hello.txt到words.txt之后

查看文件内容

[hadoop@hadoop1 ~]$ hadoop fs -cat /aa/hello.txt
hello
hello
hello
[hadoop@hadoop1 ~]$ 

chgrp

使用方法：hadoop fs -chgrp [-R] GROUP URI [URI …] Change group association of files. With -R, make the change recursively through the directory structure. The user must be the owner of files, or else a super-user. Additional information is in the Permissions User Guide. -->

改变文件所属的组。使用-R将使改变在目录结构下递归进行。命令的使用者必须是文件的所有者或者超级用户。更多的信息请参见HDFS权限用户指南。

chmod

使用方法：hadoop fs -chmod [-R] <MODE[,MODE]... | OCTALMODE> URI [URI …]

改变文件的权限。使用-R将使改变在目录结构下递归进行。命令的使用者必须是文件的所有者或者超级用户。更多的信息请参见HDFS权限用户指南。

chown

使用方法：hadoop fs -chown [-R] [OWNER][:[GROUP]] URI [URI ]

改变文件的拥有者。使用-R将使改变在目录结构下递归进行。命令的使用者必须是超级用户。更多的信息请参见HDFS权限用户指南。

du

使用方法：hadoop fs -du URI [URI …]

显示目录中所有文件的大小，或者当只指定一个文件时，显示此文件的大小。
示例：
hadoop fs -du /user/hadoop/dir1 /user/hadoop/file1 hdfs://host:port/user/hadoop/dir1 
返回值：
成功返回0，失败返回-1。

dus

使用方法：hadoop fs -dus <args>

显示文件的大小。

expunge

使用方法：hadoop fs -expunge

清空回收站。请参考HDFS设计文档以获取更多关于回收站特性的信息。

setrep

使用方法：hadoop fs -setrep [-R] <path>

改变一个文件的副本系数。-R选项用于递归改变目录下所有文件的副本系数。

示例：

• hadoop fs -setrep -w 3 -R /user/hadoop/dir1

返回值：

成功返回0，失败返回-1。

tail

使用方法：hadoop fs -tail [-f] URI

将文件尾部1K字节的内容输出到stdout。支持-f选项，行为和Unix中一致。

示例：

• hadoop fs -tail pathname

返回值：
成功返回0，失败返回-1。

test

使用方法：hadoop fs -test -[ezd] URI

选项：
-e 检查文件是否存在。如果存在则返回0。
-z 检查文件是否是0字节。如果是则返回0。 
-d 如果路径是个目录，则返回1，否则返回0。

示例：

• hadoop fs -test -e filename

查看集群的工作状态

[hadoop@hadoop1 ~]$ hdfs dfsadmin -report
Configured Capacity:  (68.68 GB)
Present Capacity:  (49.16 GB)
DFS Remaining:  (49.16 GB)
DFS Used:  ( KB)
DFS Used%: 0.00%
Under replicated blocks:
Blocks with corrupt replicas:
Missing blocks:
Missing blocks (with replication factor ): 

-------------------------------------------------
Live datanodes ():

Name: 192.168.123.102: (hadoop1)
Hostname: hadoop1
Decommission Status : Normal
Configured Capacity:  (17.17 GB)
DFS Used:  ( KB)
Non DFS Used:  (4.00 GB)
DFS Remaining:  (12.29 GB)
DFS Used%: 0.00%
DFS Remaining%: 71.57%
Configured Cache Capacity:  ( B)
Cache Used:  ( B)
Cache Remaining:  ( B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers:
Last contact: Thu Mar  :: CST 

Name: 192.168.123.105: (hadoop4)
Hostname: hadoop4
Decommission Status : Normal
Configured Capacity:  (17.17 GB)
DFS Used:  ( KB)
Non DFS Used:  (4.00 GB)
DFS Remaining:  (12.29 GB)
DFS Used%: 0.00%
DFS Remaining%: 71.58%
Configured Cache Capacity:  ( B)
Cache Used:  ( B)
Cache Remaining:  ( B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers:
Last contact: Thu Mar  :: CST 

Name: 192.168.123.103: (hadoop2)
Hostname: hadoop2
Decommission Status : Normal
Configured Capacity:  (17.17 GB)
DFS Used:  ( KB)
Non DFS Used:  (4.00 GB)
DFS Remaining:  (12.29 GB)
DFS Used%: 0.00%
DFS Remaining%: 71.58%
Configured Cache Capacity:  ( B)
Cache Used:  ( B)
Cache Remaining:  ( B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers:
Last contact: Thu Mar  :: CST 

Name: 192.168.123.104: (hadoop3)
Hostname: hadoop3
Decommission Status : Normal
Configured Capacity:  (17.17 GB)
DFS Used:  ( KB)
Non DFS Used:  (4.00 GB)
DFS Remaining:  (12.29 GB)
DFS Used%: 0.00%
DFS Remaining%: 71.57%
Configured Cache Capacity:  ( B)
Cache Used:  ( B)
Cache Remaining:  ( B)
Cache Used%: 100.00%
Cache Remaining%: 0.00%
Xceivers:
Last contact: Thu Mar  :: CST 

[hadoop@hadoop1 ~]$