Support Questions

Find answers, ask questions, and share your expertise
Announcements
Celebrating as our community reaches 100,000 members! Thank you!

Unable to view job info in the RM UI for the jobs that don't match their id

avatar
Rising Star

What are the configuration changes needed in YARN Capacity Scheduler to satisfy both below

1. prevent a user from killing another users job

2. all users should be able to view job info in the RM UI for the jobs that don't match their id

Customer has set yarn.acl.enable to true and yarn.admin.acl to yarn user, after that 1 is working as expected but 2 is not working

1 ACCEPTED SOLUTION

avatar
Master Guru

@Harini Yadav

For Capacity scheduler - you can configure acl_administer_jobs to comma separated list of users so that only those users will be able to kill jobs in particular queue.

Please see more details here - https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.3.4/bk_yarn_resource_mgt/content/controlling_ac...

For point number 2 - No config changes are needed. Every user should be able to see other user's job in YARN RM UI.

View solution in original post

4 REPLIES 4

avatar

@Harini Yadav

Have you got any solution ? If yes, can you please share before and after configs ?

avatar
Master Guru

@Harini Yadav

For Capacity scheduler - you can configure acl_administer_jobs to comma separated list of users so that only those users will be able to kill jobs in particular queue.

Please see more details here - https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.3.4/bk_yarn_resource_mgt/content/controlling_ac...

For point number 2 - No config changes are needed. Every user should be able to see other user's job in YARN RM UI.

avatar

Kuldeep Kulkarni

Good, that works. Just I'm curious to know what setting has stopped her from 2nd point(RM UI not allowing other users jobs).

avatar

Hi @Harini Yadav/ @Sri Bandaru

Only users mentioned in "mapreduce.job.acl-view-job" [mapred-site.xml] can view the private details about the job.

More information can be found by grepping for mapreduce.job.acl-view-job in https://hadoop.apache.org/docs/r2.7.1/hadoop-mapreduce-client/hadoop-mapreduce-client-core/mapred-de...