You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kylin.apache.org by "ShengJun Zheng (Jira)" <ji...@apache.org> on 2021/01/04 11:14:00 UTC
[jira] [Created] (KYLIN-4859) Log4J reinitialized/reconfigured by
Spark Logging
ShengJun Zheng created KYLIN-4859:
-------------------------------------
Summary: Log4J reinitialized/reconfigured by Spark Logging
Key: KYLIN-4859
URL: https://issues.apache.org/jira/browse/KYLIN-4859
Project: Kylin
Issue Type: Improvement
Components: Others
Affects Versions: v4.0.0-alpha
Reporter: ShengJun Zheng
Spark Logging was intrucoded in kylin 4.0-alpha, it uses SL4J too, but it will reinitialize log4j when RootLoggers's Appender is empty.
See: https://github.com/apache/spark/blob/45e19bb99acd5066723fec2bbdc0c99c696c3daf/core/src/main/scala/org/apache/spark/internal/Logging.scala#L120
this causes some logs of a query logged in file(kylin.log), the other logged in stdout(kylin.out) using configuration file log4j-defaults.properties in spark-core_2.11-2.x.x.jar
not firendly to read and anlyze a query performance, and may cause other logging confusions
To Avoid this is, default appenders for rootLoggers should be set in kylin-server-log4j.properties
--
This message was sent by Atlassian Jira
(v8.3.4#803005)