Flume Practical Application Analysis

Business background:

The output of the log files generated by the Java project to flume

The first step:

Output the log to flume, write the log4j in the Java program, and specify the output to which Flume server

Log4j.rootlogger=info,flumelog4j.appender.flume= Org.apache.flume.clients.log4jappender.log4jappenderlog4j.appender.flume.hostname= 192.168.13.132log4j.appender.flume.port=41414

Step Two:

Import Java.util.date;import Org.apache.log4j.logger;public class Flume {public static final Logger Logger = Logger.getlo Gger (flume.class);p ublic static void Main (string[] args) throws Exception {while (true) {Logger.info (string.valueof (new Date (). GetTime ())); Thread.Sleep (2000);}}

Step Three:

Reference

Case 15:avro Load Balancer dynamically gets the contents of the file in start 132 and 133, you can also specify the priority, reference case 17:avro load balancer plus priority configuration

For example, three different logs are generated on different servers, such as 132 generated by Access 133 is generated by the Ugcheader can be

Log4j.rootlogger=info,access, Ugcheader log4j.appender.access = Org.apache.flume.clients.log4jappender.Log4jAppenderlog4j.appender.access. Hostname=192.168.13.132log4j.appender.access. port=41414

Log4j.rootlogger=info,access, Ugcheader Log4j.appender.ugcheader = Org.apache.flume.clients.log4jappender.Log4jAppenderlog4j.appender.ugcheader. Hostname=192.168.13.133log4j.appender.ugcheader. port=41414

Flume Practical Application Analysis