yarn命令

根据状态查看Yarn全部运行应用程序

# 查看全部运行过的yarn程序
yarn application -list -appStates ALL
# 查看正在运行的yarn程序，等同于：yarn application -list
yarn application -list -appStates RUNNING

-appStates

与-list一起使用，可根据输入的逗号分隔的应用程序状态列表来过滤应用程序。有效的应用程序状态可以是以下之一：ALL，NEW，NEW_SAVING，SUBMITTED，ACCEPTED，RUNNING，FINISHED，FAILED，KILLED

---

根据提交程序代码提交类型查看运行程序

# 查看Flink提交到yarn的程序
# yarn application -list -appTypes 'Apache Flink'Application-Id	    Application-Name	    Application-Type	      User	     Queue	             State	       Final-State	       Progress	                       Tracking-URL
application_1651044751782_0093	avris-profile-platform	        Apache Flink	      hdfs	root.users.hdfs	           RUNNING	         UNDEFINED	           100%	     http://dev-bigdata-24-149:4681

-appTypes #与-list一起使用，可以根据输入的逗号分隔的应用程序类型列表来过滤应用程序。

yarn top 查看正在运行的状态

YARN top - 17:03:12, up 13d, 6:24, 0 active users, queue(s): root
NodeManager(s): 3 total, 3 active, 0 unhealthy, 0 decommissioned, 0 lost, 0 rebooted
Queue(s) Applications: 7 running, 119 submitted, 0 pending, 69 completed, 43 killed, 0 failed
Queue(s) Mem(GB): 16 available, 70 allocated, 0 pending, 0 reserved
Queue(s) VCores: 67 available, 29 allocated, 0 pending, 0 reserved
Queue(s) Containers: 29 allocated, 0 pending, 0 reservedAPPLICATIONID USER             TYPE      QUEUE PRIOR   #CONT  #RCONT  VCORES RVCORES     MEM    RMEM  VCORESECS    MEMSECS %PROGR       TIME NAME                                            application_1655951936605_0067 hadoop     apache flink    default     0       9       0       9       0     26G      0G    1636659    4728125 100.00   02:02:31 avris-clue-ruleapplication_1655951936605_0120 hadoop     apache flink    default     0       6       0       6       0     12G      0G     509589    1019188 100.00   00:23:35 avris-profile-platformapplication_1655951936605_0122 hadoop     apache flink    default     0       4       0       4       0     11G      0G       4945      13599 100.00   00:00:20 avris-strategy-ruleapplication_1655951936605_0048 hadoop     apache flink    default     0       3       0       3       0      9G      0G    1287268    3861784 100.00   04:23:11 avris-scEvent-toCDPapplication_1655951936605_0121 hadoop     apache flink    default     0       3       0       3       0      5G      0G     253043     421733 100.00   00:23:25 avris-behavior-labelRegineapplication_1655951936605_0003 hadoop     apache flink    default     0       2       0       2       0      3G      0G    2290410    3435610 100.00   13:06:06 avris-userinfo-changeapplication_1655951936605_0002 hadoop     apache flink    default     0       2       0       2       0      4G      0G    2290768    4581525 100.00   13:06:09 avris-event-hbase

说明：

参数	说明
#RCONT	使用的容器container个数
RVCORES	使用的cpu核数
RMEM	使用的内存大小
APPLICATIONID	yarn任务ID
USER	提交任务用户
TYPE	任务类型，如：apache flink
QUEUE	提交任务所在队列

yarn top整体资源使用查看

YARN top - 17:08:52, up 13d, 6:29, 0 active users, queue(s): root
NodeManager(s): 3 total, 3 active, 0 unhealthy, 0 decommissioned, 0 lost, 0 rebooted
Queue(s) Applications: 7 running, 119 submitted, 0 pending, 69 completed, 43 killed, 0 failed
Queue(s) Mem(GB): 16 available, 70 allocated, 0 pending, 0 reserved
Queue(s) VCores: 67 available, 29 allocated, 0 pending, 0 reserved
Queue(s) Containers: 29 allocated, 0 pending, 0 reserved

参数	说明
available	剩余可用的资源
allocated	已经使用的资源
reserved	已经申请正在分配的资源
pending	等待申请的资源
available + allocated	Yarn的总资源

Yarn web页面

工具脚本

根据yarn应用名称kill进程

[hadoop@10 /data]$ cat kill.sh 
applicationId=`yarn application -list | grep '提交程序在Yarn名称' |awk '{print $1}'`
yarn application -kill $applicationId
yarn application -list

根据yarn应用名称查看日志

applicationId=`yarn application -list | grep '提交程序在Yarn名称' |awk '{print $1}'`
yarn logs -applicationId $applicationId