2

My use case is to find out on how many users the api will break. For eg. I should be able to tell that this particular api breaks for 100 users and works fine for 99 users. Can we achieve the same? And if someone knows its approach then let me know.

Andremoniy
  • 34,031
  • 20
  • 135
  • 241
Rocky
  • 21
  • 6

1 Answers1

0

In my experience teams use trial and error, you should be able to pass some command-line or environment-variable to your test with the number of concurrent users.

But you can experiment with rampUsers() etc. here it is just Gatling concepts so refer to their docs if needed.

Peter Thomas
  • 54,465
  • 21
  • 84
  • 248
  • Thanks for you help. Can we have any feature to dynamically increase the load of users and timings ? for eg. while running the load test for rampUsers() by passing args via maven command line we can pass fix set of users and fixed set of time to run that load test but in between its execution if we were able to increase the users & timings then we can find the breakpoint and then will solve the issue. – Rocky Aug 18 '20 at 07:47
  • 1
    @Rocky there is no "issue" - you are welcome to find ways to do it and contribute code if needed. maybe you should ask this as a `gatling` question (not `karate`) so you get answers from that team. all the best – Peter Thomas Aug 18 '20 at 07:49
  • 1
    thanks for help will check with Gatling team – Rocky Aug 18 '20 at 07:53