我想用flink写一个需要两个阶段的计算。
在第一阶段，我创建一个图并获取其顶点ID：

List<String> ids = graph.getVertexIds().collect();

在第二阶段，我想使用这些ID为每个顶点运行singlesourceshortestpath。

for (String id: ids){
        System.out.println("Source Id: "+id);
        graph.run( new SingleSourceShortestPaths<String, String>(id, 10)).print();
    }

它在本地工作（在intellijide和命令行中使用 ./bin/flink run ... )，但当我使用flink的webui提交作业时，程序只执行到 collect() 方法，而不运行程序的其余部分（对于语句和 print() ).
有什么问题？
这是我的密码：

import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.graph.Edge;
import org.apache.flink.graph.Graph;
import org.apache.flink.graph.library.SingleSourceShortestPaths;

import java.util.ArrayList;
import java.util.List;

public class Main {
    public static void main(String[] args) throws Exception {

        ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

        Edge<String, Double> e1 = new Edge<String, Double>("1", "2", 0.5);
        Edge<String, Double> e2 = new Edge<String, Double>("2", "3", 0.5);
        Edge<String, Double> e3 = new Edge<String, Double>("4", "5", 0.5);
        Edge<String, Double> e4 = new Edge<String, Double>("5", "6", 0.5);
        Edge<String, Double> e5 = new Edge<String, Double>("7", "8", 0.5);

        List<Edge<String, Double>> edgeList = new ArrayList<Edge<String, Double>>();
        edgeList.add(e1);
        edgeList.add(e2);
        edgeList.add(e3);
        edgeList.add(e4);
        edgeList.add(e5);

        Graph<String, String, Double> graph = Graph.fromCollection(edgeList,
                new MapFunction<String, String>() {
                    public String map(String value) {
                        return value;
                    }
                }, env);

        List<String> ids = graph.getVertexIds().collect();

        for (String id: ids){
            System.out.println("Source Id: "+id);
            graph.run( new SingleSourceShortestPaths<String, String>(id, 10)).print();
        }
    }
}