[FLINK-7722] MiniCluster does not appear to honor Log4j settings

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Not A Problem
• Affects Version/s: 1.3.2
• Component/s: Runtime / Task
• Assignee: Unassigned
• Reporter: Elias Levy
• Created: 27/Sep/17 20:22
• Updated: 04/Oct/17 10:47
• Resolved: 27/Sep/17 22:04

Description

When executing a job from the command line for testing, it will output logs like:

Connected to JobManager at Actor[akka://flink/user/jobmanager_1#1337544104] with leader session id 59dd0d9c-938e-4e79-a0eb-709c5cf73014.
09/27/2017 13:15:13	Job execution switched to status RUNNING.
09/27/2017 13:15:13	Source: Custom File Source(1/1) switched to SCHEDULED 
09/27/2017 13:15:13	Source: Collect

It will do so even if the log4j.properties file contains:

log4j.rootLogger=ERROR, stdout
log4j.logger.org.apache.flink=ERROR
log4j.logger.akka=ERROR
log4j.logger.org.apache.kafka=ERROR
log4j.logger.org.apache.hadoop=ERROR
log4j.logger.org.apache.zookeeper=ERROR

It seems that the MiniCluster does not honor Log4j settings, or at least that is my guess.