【备忘/shell】hadoop 常见shell 与相关进程操作命令 ing

本文介绍hdfs的相关操作shell，Hadoop格式化、启动、以及其他常见的进程操作命令，以便在操作hadoop时能够快速查看命令，起到备忘的作用。

一. HDFS shell

# 1. 上传文件
hdfs dfs -put ./wuguo.txt /sanguo

# 2. 追加文件
hdfs dfs -appendToFile liubei.txt /sanguo/shuguo.txt

#3. 下载文件
hdfs dfs -get /sanguo/shuguo.txt ./shuguo2.txt

#3. 修改文件所属权限
hdfs dfs  -chmod 666  /sanguo/shuguo.txt
hdfs dfs  -chown  atguigu:atguigu   /sanguo/shuguo.txt

#4. 文件大小
hdfs dfs -du -s -h /jinguo

说明：27表示文件大小；81表示27*3个副本；/jinguo表示查看的目录

#5. 递归删除
hdfs dfs -rm -r /sanguo
#删除
hdfs dfs -rm /sanguo/shuguo.txt 

# 展示文件大小
hadoop fs -du -h hdfs://nn1node:9000/file_path/test_performance_10m
6.7 G  20.0 G  hdfs://path/test_performance_10m/pday=20230101
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

官网：shell 官网

 
 

二. Yarn常见命令

1. application操作

1）列出所有Application：yarn application -list

yarn application -list

2021-02-06 10:21:19,238 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Total number of applications (application-types: [], states: [SUBMITTED, ACCEPTED, RUNNING] and tags: []):0
                Application-Id      Application-Name      Application-Type        User       Queue               State         Final-State         Progress                        Tracking-URL
1
2
3
4
5

2）根据Application状态获取app列表：yarn application -list -appStates

（所有状态：ALL、NEW、NEW_SAVING、SUBMITTED、ACCEPTED、RUNNING、FINISHED、FAILED、KILLED）

yarn application -list -appStates FINISHED

2021-02-06 10:22:20,029 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Total number of applications (application-types: [], states: [FINISHED] and tags: []):1
                Application-Id      Application-Name      Application-Type        User       Queue               State         Final-State         Progress                        Tracking-URL
application_1612577921195_0001            word count             MAPREDUCE     atguigu     default            FINISHED           SUCCEEDED             100% http://hadoop102:19888/jobhistory/job/job_1612577921195_0001
1
2
3
4
5
6

3）Kill掉Application

yarn application -kill application_1612577921195_0001

2021-02-06 10:23:48,530 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Application application_1612577921195_0001 has already finished
1
2
3
4

 

2. 查看日志

1）查询Application日志：

yarn logs -applicationId <ApplicationId>
1

2）查询Container日志：yarn logs -applicationId < AppId> -containerId < ContainerId >

yarn logs -applicationId application_1612577921195_0001 -containerId container_1612577921195_0001_01_000001
1

 

3. 查看container

1）列出所有Container：yarn container -list < ApplicationAttemptId >

yarn container -list appattempt_1612577921195_0001_000001

2021-02-06 10:28:41,396 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Total number of containers :0
                  Container-Id	          Start Time	         Finish Time	               State	                Host	   Node Http Address	
1
2
3
4
5

2）打印Container状态： yarn container -status < ContainerId >

yarn container -status container_1612577921195_0001_01_000001

2021-02-06 10:29:58,554 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Container with id 'container_1612577921195_0001_01_000001' doesn't exist in RM or Timeline Server.
1
2
3
4

注：只有在任务跑的途中才能看到container的状态

 

4. node状态

列出所有节点：yarn node -list -all

yarn node -list -all

2021-02-06 10:31:36,962 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8032
Total Nodes:3
         Node-Id	     Node-State	Node-Http-Address	Number-of-Running-Containers
 hadoop103:38168	        RUNNING	   hadoop103:8042	                           0
 hadoop102:42012	        RUNNING	   hadoop102:8042	                           0
 hadoop104:39702	        RUNNING	   hadoop104:8042
1
2
3
4
5
6
7
8

 

5. 更新队列

更新队列：yarn rmadmin -refreshQueues

yarn rmadmin -refreshQueues
2021-02-06 10:32:03,331 INFO client.RMProxy: Connecting to ResourceManager at hadoop103/192.168.10.103:8033
1
2

 

三. 启停命令

1. 启停方式一

启动

sbin/start-dfs.sh
sbin/start-yarn.sh
1
2

停止

sbin/yarn-dfs.sh
sbin/yarn-yarn.sh
1
2

 

2. 启停方式二

hdfs --daemon start/stop journalnode/namenode/datanode/zkfc
yarn --daemon start/stop resourcemanager/nodemanager
1
2

 

四. 一些场景命令ing

总结一些常见的hadoop场景

1. 关闭安全模式

hadoop dfsadmin -safemode leave
1

2. 查看那个节点是activeNamenode

hdfs haadmin -getServiceState nn1
#active
hdfs haadmin -getServiceState nn2
# standby
1
2
3
4