How correlation can be done from resources that are fetched in parallel?

Gatling (Open-Source)
1

Hi,Greeting everyone!

In order to avoid multiple exec being called in my Gatling script, I’m trying to allow resources in parallel. (I have set throttle in the code for load testing. I do not want certain calls in the scenario, to be included in the throttle target. Hence trying to make use of .resources() to fetch these calls in parallel.**)
But the correlations done from these HTTP resources are not retrieved. While configuring the code with multiple exec calls for each HTTP requests, the correlations are working as expected.

I want to extract values from the resources to be passed to the next request.
Is it possible to retrieve values and do checks on the resources? Can someone please help me with this?

Code snippet having issue :

SNIPPET.PNG
SNIPPET.PNG931×592 22.7 KB

Getting error :

---- Errors --------------------------------------------------------------------
> REDIRECTION_1: Failed to build request: No attribute named 're 1 ( 5.56%)
direction_1’ is defined
> REDIRECTION_2: Failed to build request: No attribute named 're 1 ( 5.56%)
direction_2’ is defined
> LOGIN: Failed to build request: No attribute named 'Correlatio 1 ( 5.56%)
nParameter’ is defined

Code snippet which is working fine :

WORK.PNG
WORK.PNG902×488 19.9 KB

Regards,
Tenu

2

What you’re doing/expecting doesn’t make any sense:
• no, using resources won’t bypass throttling
• resources are executed in parallel (with a limit in the number of connections by host), so there’s no way to guarantee that some request will terminate first so that some data can be captured so another resource request can be computed
• redirect chains are by definition sequential, not concurrent

3

Moreover, what you’re doing with disabling redirects is useless (I guess we’ll end up removing this option in a future release).
There’s a currentLocation check that let’s you capture data from the landing url. Please read the documentation.

4

Okay, I understood. Thanks for the response Stephane!

Is there any other way to include multiple calls in a single exec, such that it runs sequentially and we can capture data for the subsequent calls.