How to add variety to the endpoints I test

[Sandeep]

Hello,

I’m a little stuck how to create a representative test. (I have completed the academy tutorials)

Our api contains numerous endpoints so what I’ve done is captured a log file from Kibana. Out of the 100 or so requests the top 5 contribute to around 50% of everything captured. The remainder is just varying paths

e.g …/game/48390 or …/game/30435

Is it sufficient enough to just run the most frequent endpoints used to test our service or should I apply a feeder to switch up the paths and request queries.

It’s probably a conversation I need to have with the devs I assume.

Thank you

[sbrevet]

Hi @Sandeep

@Sandeep

It’s probably a conversation I need to have with the devs I assume.

Indeed!

As Gatling is a tool, we can only help you with how to use the tool, not what to do with it.

@Sandeep

The remainder is just varying paths

e.g …/game/48390 or …/game/30435

From my personal point of view, this is the same request with a parameter (the game id, I guess).
In your position, I will group all such request together, because it’s really the same usage (only value change, not computation)

Hope this helps!

Cheers!

[Sandeep]

Thank you so much for the quick response!

Very helpful. Just to clarify one point If I have the below

game/monday/34343
game/tuesday/34234

grouping my test all all under …/game is sufficient?

[sbrevet]

It feels to me that is still the same request… with two params: day and id

.../api/game/{day}/{id}

But you (or the backend devs) are the best people to know what to test for your own project.

Cheers!