Suppose I am using constantUsersPerSec(10). How will my test run?
In first second Gatling will inject 10 users and in second second will Gatling inject 10 users irrespective of pauses and api response. If it’s taking next 10 users irrespective of pauses would it put extra load since the previous 10 users have not exited?
Hi @Deepak,
Yes, you got it right!
Injection profile is indeed the definition of the time that each new user is injected in the test irrespective to the users already present.
Cheers!
Second question: Is there a way to control the injection rate.?
Say I want first 10 users to exit first and then next 10 users come in or I can wait for some time after each set of user that is injected?
You describe what we call Closed model for injection. But it is rarely what the application is really doing and we do warn that you should not use it if the application is not behind such a system.
Cheers!
Thanks, just wanted to know if it was possible with open model.
Two more questions.
If I am running load for 30 mins it goes beyond it.?
Will it be possible to log to datadog?
Hi @Deepak I am not sure I understand your use case, but you could investigate sequential scenarios.
So my real problem is though Gatling will push a load of 10 constant user throughout the test duration, I do not see a constant load of 10 users in my server side reports. On server the load is not evenly distributed I some times a load of 1 user and some times a load of 30 users. ?