If I’m not mistaken from the shown output, you’re talking about the stdout output of the client, right? This output is not controlled by the log4j.properties or logback.xml file.
However, you can use any command line tool available on your platform to redirect the stdout. For example on a Linux system, you could redirect the log output with ./flink run Flink.jar > myFile or you could use ./flink run Flink.jar | tee myFile to redirect the output and still print it. Cheers, Till On Fri, Jun 19, 2015 at 1:19 PM Chiwan Park chiwanp...@icloud.com <http://mailto:chiwanp...@icloud.com> wrote: Hi, Flink uses slf4j and log4j for logging. > > You can override default log4j configuration programmatically. [1] > Or you can use logback as a logging backend and override default logback > configuration also. [2][3] > > [1] http://stackoverflow.com/a/9003191 > [2] > http://ci.apache.org/projects/flink/flink-docs-master/internals/logging.html > [3] http://stackoverflow.com/a/3810936 > > Regards, > Chiwan Park > > > > On Jun 19, 2015, at 8:05 PM, Juan Fumero < > juan.jose.fumero.alfo...@oracle.com> wrote: > > > > Hi, > > is there any option (from API level) to redirect the log messages to a > > file instead of stdout? > > > > Thanks > > > > > > log4j:WARN No appenders could be found for logger > > (org.apache.flink.api.java.ExecutionEnvironment). > > log4j:WARN Please initialize the log4j system properly. > > log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for > > more info. > > 06/19/2015 12:14:12 Job execution switched to status RUNNING. > > .... > > 06/19/2015 12:14:12 DataSink (collect() sink)(1/1) switched to > > DEPLOYING > > 06/19/2015 12:14:12 DataSink (collect() sink)(1/1) switched to > > RUNNING > > 06/19/2015 12:14:12 DataSink (collect() sink)(1/1) switched to > > FINISHED > > 06/19/2015 12:14:12 Job execution switched to status FINISHED. > > > > > > > >
