First, thank you for creating and maintaining this tool. I’ve been recommending this tool widely since discovering it.
I am using Gatling v2.2.0.
We are using Gatling to simulate 1 million users across many Gatling host nodes. The test takes approximately 2 hours to run. We then use gatling.sh -ro to merge the logs from all the nodes into one HTML report.
We found a problem in our HTML reports. The Details page for one of the requests was missing some of the OK request records near the end of the test. There may have been more missing from earlier in the test; we have not looked into that specifically.
I’ve created two simpler log files that reproduce the problem. The attached no_problem.log contains 3 requests and generates a report as expected. The other file, problem.log, generates empty charts in the Detail page for the request.
The only difference between the two log files is that problem.log’s last request occurs some time after the last request in no_problem.log.
Are we expecting Gatling to do something it cannot do?
problem.log (394 Bytes)
no_problem.log (394 Bytes)