[jira] [Created] (MAPREDUCE-7489) Exposing Default Handling of Misconfigurations to Users in MasterPrincipal

["LoggingResearch (Jira)" <ji...@apache.org>]

LoggingResearch created MAPREDUCE-7489:
------------------------------------------

             Summary: Exposing Default Handling of Misconfigurations to Users in MasterPrincipal
                 Key: MAPREDUCE-7489
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-7489
             Project: Hadoop Map/Reduce
          Issue Type: Improvement
          Components: mapreduce-client
    Affects Versions: 3.3.6
         Environment: Version: {{`3.3.6`}}
Location: {{{}`hadoop/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/Master.java`{}}}， in {{`getMasterPrincipal`}} method, from Line 51 - 59.
            Reporter: LoggingResearch
         Attachments: MasterTest.java, original-vs-log-enhanced.md

When users attempt to configure the {{MasterPrincipal}} for local mode using the system user {{{}root{}}}, they might set {{mapreduce.framework.name}} to {{local}} and {{mapreduce.jobtracker.kerberos.principal}} to {{{}root{}}}, following the configuration documentation. Although the job executes, it may not behave as expected. Users may spend considerable time troubleshooting before discovering that setting {{mapreduce.framework.name}} to {{classic}} yields the desired outcome. This confusion arises from the silent default handling in the code, which is counterintuitive, as "classic" is not typically associated with local mode. Even worse, there is no system output to indicate this default handling, leaving users puzzled.
 
*Expected Behavior:* 
The system should provide output indicating to users any unseen default handling regarding {{{}MasterPrincipal{}}}.
 
*How-to-Fix:*
We propose improving system transparency by adding logging.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-dev-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-dev-help@hadoop.apache.org