获取hadoop-yarn 的列表信息

阅读更多

一.背景

     一般hadoop集群的东西，需要去监控，或者杀东西之类的，所以需要调用下他们的API，下面是基本操作，获取正在跑的flink 程序，当然，也能获取CPU 内存等信息：

 

 

二.代码

    public static void main(String[] args) throws IOException, YarnException {

        YarnConfiguration yarnConf = YarnConfLoader.getYarnConf("/Users/qqr/Downloads/yarn-conf");
        YarnClient yarnClient = YarnClient.createYarnClient();
        yarnClient.init(yarnConf);
        yarnClient.start();
        Set set = new HashSet<>();
        // ApplicationType
        set.add("Apache Flink");
        EnumSet enumSet = EnumSet.noneOf(YarnApplicationState.class);
        enumSet.add(YarnApplicationState.RUNNING);
        List reportList = yarnClient.getApplications(set, enumSet);
        for(ApplicationReport report : reportList){
            System.out.println("id-name:"+report.getApplicationId()+":"+report.getName());
            System.out.println("mem:"+report.getApplicationResourceUsageReport().getNeededResources().getMemory());
            System.out.println();
        }
    }

 

 

输出：

id-name:application_1546069948045_128203:huoguo

mem:2048

 

id-name:application_1542000413153_5869:binlog_hbase

mem:2048