如何使用JMeter计算两个微服务中Post与Get API的同步耗时？

Here's a practical, step-by-step breakdown to set up your test plan and calculate the sync time between your Post and Get APIs:

1. Build the Base Test Structure #

• Start by adding a Thread Group to your test plan (right-click Test Plan > Add > Threads (Users) > Thread Group). Tweak the number of threads/iterations to match your testing needs.

2. Add the Post API & Capture Its Initiation Time #

• Under the Thread Group, add your HTTP Request for the Post API (right-click Thread Group > Add > Sampler > HTTP Request). Fill in all required details: method, URL, headers, request body, etc.
• To track exactly when the Post request starts, add a JSR223 PreProcessor to the Post request (right-click Post HTTP Request > Add > Pre Processors > JSR223 PreProcessor). Paste this code to store the start timestamp in a reusable variable:

  vars.put("postInitiationTime", String.valueOf(System.currentTimeMillis()));
  

  If you want to measure time from when the Post request completes (instead of starts), swap the pre-processor for a JSR223 PostProcessor and name the variable postCompletionTime.

3. Set Up a While Controller to Repeat the Get Request #

• After the Post request, add a While Controller (right-click Thread Group > Add > Logic Controller > While Controller). This will loop the Get request until we get a successful response.
• For the controller's condition, use an expression that checks if the Get request hasn't succeeded yet. Examples:
  • If success means a 200 response code: ${__jexl3("${getResponseCode}" != "200",)}
  • If success depends on specific content in the response body: ${__jexl3(!${__groovy(vars.get("getResponseBody").contains("success_flag")),},)}
    To avoid infinite loops, add a maximum iteration limit, e.g.: ${__jexl3(${__counter(FALSE, getLoopCount)} < 100 && "${getResponseCode}" != "200",)} (caps retries at 100)

4. Configure the Get API Inside the Loop #

• Inside the While Controller, add your HTTP Request for the Get API and fill in its details.
• Add a Response Assertion to the Get request (right-click Get HTTP Request > Add > Assertions > Response Assertion) to define what counts as a "successful" response—for example, checking for a 200 code or specific text in the response.
• Add a Fixed Timer to the Get request (right-click Get HTTP Request > Add > Timers > Fixed Timer) to add a delay between retries (e.g., 1000ms = 1 second) to avoid overwhelming your server.
• When the Get request succeeds, capture the timestamp with a JSR223 PostProcessor on the Get request:

  vars.put("getSuccessTime", String.valueOf(System.currentTimeMillis()));
  

5. Calculate & Log the Total Sync Time #

• After the While Controller, add a JSR223 Sampler (right-click Thread Group > Add > Sampler > JSR223 Sampler) to compute the time difference. Use code like this:

  // Adjust variable name if you used postCompletionTime instead
  long postStart = Long.parseLong(vars.get("postInitiationTime"));
  long getSuccess = Long.parseLong(vars.get("getSuccessTime"));
  long totalSyncMs = getSuccess - postStart;
  
  vars.put("totalSyncTime", String.valueOf(totalSyncMs));
  log.info("Total sync time (Post start to Get success): " + totalSyncMs + " milliseconds");
  

6. View Your Results #

• Add listeners like View Results Tree or Summary Report to monitor test execution and see the totalSyncTime value. You can also add a Debug Sampler to inspect all variables during testing.

内容的提问来源于stack exchange，提问作者Reem Sami