Multi‑Chain Performance Testing Demo Using FunTester

In recent days the author completed basic browser request replay functionality and now shares a multi‑chain performance testing demo that also resolves the ordering issue of multiple interface requests during replay.

Overall approach :

Copy requests from different chains into separate files.

Use the FileUtil utility to retrieve all files in a directory.

Convert each file into a List<HttpRequestBase> using CurlUtil .

Feed the resulting list into the multithreaded FunTester performance‑testing framework.

Because each chain has varying response times, the demo adopts a fixed‑duration (fixed‑time) testing model with a fixed number of threads rather than a fixed request count or fixed QPS.

The implementation requires writing a custom multithreaded class that repeatedly fetches HttpRequestBase objects from the list and executes them.

Demo code (Java) :

package com.okayqa;

import com.alibaba.fastjson.JSONObject;
import com.fun.base.constaint.ThreadLimitTimeCount;
import com.fun.base.exception.FailException;
import com.fun.base.interfaces.MarkThread;
import com.fun.frame.httpclient.FanLibrary;
import com.fun.utils.FileUtil;
import com.okayqa.common.Common;
import com.okayqa.composer.base.OkayBase;
import org.apache.http.client.methods.HttpRequestBase;

import static com.fun.utils.CurlUtil.getRequests;

class TTT extends OkayBase {

    public static void main(String[] args) {
        // path is the absolute directory + filename
        String path = "";
        // time is the test duration in seconds
        int time = 1000;
        // description, keep within 20 characters
        String desc = "FunTester multi‑chain load test Demo";
        // retrieve all log files
        def file = FileUtil.getAllFile(path);
        def tasks = [];
        file.each {
            tasks << new FunTester(it, time, null);
        }
        new Concurrent(tasks, desc).start();
        FanLibrary.testOver();
    }

    /**
     * FunTester test class
     */
    static class FunTester extends ThreadLimitTimeCount
{
        List
reqs;

        FunTester(String s, int time, MarkThread markThread) {
            super(s, time, markThread);
            reqs = getRequests(s);
        }

        @Override
        void before() {
            super.before();
            // initialize thread mark to avoid NPE
            threadmark = threadmark == null ? EMPTY : threadName;
        }

        @Override
        protected void doing() throws Exception {
            reqs.each {
                it.removeHeaders(Common.REQUEST_ID.getName());
                def header = Common.getRequestIdHeader();
                it.addHeader(header);
                threadmark << header + CONNECTOR;
                JSONObject simple = FanLibrary.executeSimlple(it);
                if (!isRight(simple))
                    FailException.fail("Response error!" + it.getURI().toString());
            }
        }
    }
}

The author adds initialization for threadmark and handles marking inside the doing() method, along with a generic response validation step. Users can extend the validation logic for specific interfaces as needed.

Console output example (truncated):

INFO-> 当前用户：fv，IP：10.60.193.37，工作目录：/Users/fv/Documents/workspace/okay_test/,系统编码格式:UTF-8,系统Mac OS X版本:10.16
INFO-> FunTester多链路压测Demo进度:▍▍▍▍▍▍▍▍▍▍▍  16.41%
... (progress updates) ...
INFO-> 总计50个线程，共用时：31.121 s,执行总数:1917,错误数:0,失败数:0
INFO-> 数据保存成功！文件名：/Users/fv/Documents/workspace/okay_test/long/data/FunTester多链路压测Demo031618,50
... JSON result ...
{"rt":796,"total":1917,"qps":62.814,"failRate":0.0,"threads":50,"startTime":"2021-02-03 16:18:14","endTime":"2021-02-03 16:18:45","errorRate":0.0,"executeTotal":1917,"mark":"FunTester多链路压测Demo031618","table":"..."}

Links to the source code repositories are provided (Gitee and GitHub), along with references to related articles on FunTester, fixed‑QPS testing, IoT testing, and automated testing lifecycle.