背景

flink默认会把所有日志打印到taskexecutor的日志，日志混杂在一起很难排查，并且会越来越大
可以通过的以下的方式分割离职并做Rolling

添加依赖jar包

https://mirrors.koehn.com/apache/logging/log4j/extras/1.2.17/apache-log4j-extras-1.2.17-bin.zip
下载并copy apache-log4j-extras-1.2.17.jar 包到flink/lib目录下

image.png

增加log4j.properties配置

log4j.appender.zetagAppender=org.apache.log4j.rolling.RollingFileAppender
log4j.appender.zetagAppender.RollingPolicy.FileNamePattern=/usr/local/flink/log/zetag.%d{yyyyMMdd-HH}.log
log4j.appender.zetagAppender.RollingPolicy=org.apache.log4j.rolling.TimeBasedRollingPolicy
log4j.appender.zetagAppender.append=false
log4j.appender.zetagAppender.layout=org.apache.log4j.PatternLayout
log4j.appender.zetagAppender.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss,SSS} %-5p %-60c %x - %m%n

之后重启整个集群就可以看到效果了