Error message "environment is not allowed to be empty" when trying to report results from locust load test

[afaderoTextNow]

Hi there, first of all thank you for providing this awesome tool. I'm trying to set it up for continuous uploading using the jtl_listener_service.py and then adding the listener to my load test by adding the following snippet to my test
from jtl_listener_service import JtlListener @events.init.add_listener def on_locust_init(environment, **_kwargs): JtlListener(env=environment, project_name="project name", scenario_name="scenario name", hostname="hostname", backend_url="http://IP_ADDRESS").
Issue I'm having is that I getting this error when I try to run my locust load test, which I think is being triggered by this line payload = {"environment": self.env.host} in jtl_listener_service.py
[2025-05-13 17:41:31,456] TN00939/INFO/root: Logging with token [2025-05-13 17:41:31,589] TN00939/INFO/root: Login with token returned {'jwtToken': 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI0ZTkwMWExMS04MzdkLTRmMDgtYmUxYi1lZmE2NGI3OGE5MzYiLCJpYXQiOjE3NDcxNzI0OTEsImV4cCI6MTc0NzM0NTI5MX0.HE0-ORV7HXqYc1WXKVF2BemZQxWH-CDi1BBjafExCTk'} [2025-05-13 17:41:31,589] TN00939/INFO/root: Starting async item in Reporter [2025-05-13 17:41:31,607] TN00939/INFO/root: Reporter responded with {'message': '"environment" is not allowed to be empty'} [2025-05-13 17:41:31,607] TN00939/ERROR/root: Error while starting the test [2025-05-13 17:41:31,608] TN00939/INFO/locust.runners: sys.exit(1) called (use log level DEBUG for callstack).
I started up my test environment using the docker-compose file and when I check the logs for jtl_reporter_be I have the following logs as well
{"level":"info","message":"Express server listening on port 5000","timestamp":"2025-05-13 21:40:40"} {"level":"info","message":"Bree scheduler was started","timestamp":"2025-05-13 21:40:40"} {"level":"info","message":"By using this app you agree with the use of analytics in this app to help improve user experience and the overall functionality of the app.","timestamp":"2025-05-13 21:40:40"} {"level":"info","message":"POST /api/auth/login-with-token 200 52ms","meta":{}} {"level":"debug","message":"User 4e901a11-837d-4f08-be1b-efa64b78a936 with role admin accessing a resource with allowed roles: operator,admin","timestamp":"2025-05-13 21:41:31"} {"data":null,"isBoom":true,"isServer":false,"level":"error","output":{"headers":{},"payload":{"error":"Bad Request","message":"\"environment\" is not allowed to be empty","statusCode":400},"statusCode":400},"timestamp":"2025-05-13 21:41:31"} {"level":"info","message":"POST /api/projects/local_load_tests/scenarios/send_text_message/items/start-async 400 5ms","meta":{}} Worker for job "analytics-report" online undefined Worker for job "analytics-report" exited with code 0 undefined
I've checked the documentation and I'm not sure what I'm missing and any help you can provide will be greatly appreciated. Tx

[ludeknovy]

Hi @afaderoTextNow !
Please make sure that the environment resolves to any value. It cannot be empty. So you can either set the env var HOST manually, or adjust the code so it propagates some value into there.

[afaderoTextNow]

Hi @ludeknovy, tx for getting back to me and apologies for the late ack from my side as I was away when you sent your message. I'll try what you mention and see how it goes.

[afaderoTextNow]

@ludeknovy QQ - in the docker-compose file there are 2 JWT environment variables, JWT_TOKEN, and JWT_TOKEN_LOGIN. I generated an API key which I'm using for the JWT_TOKEN, is this the same token to use for the JWT_TOKEN_LOGIN? If it's not how do I get the JWT_TOKEN_LOGIN? Also according to the docs, I have to set this environment variable JTL_API_TOKEN to the same api key as well. I just wanted to clarify if they are all the same value

[afaderoTextNow]

@ludeknovy Tx. I'll look into why no data was sent from the listener

[afaderoTextNow]

@ludeknovy I was able to find a work-around to the array being empty by adding another condition to this code block. It now works for me like this

if self._finished and len(self.results) != 0:
    ...

[afaderoTextNow]

It appears the array was getting cleared as it was looping, so the additional check seems to help

[afaderoTextNow]

@ludeknovy wrt to the scatter chart does my test need to run for a significant amount of time for it to show data, I was just testing things and running my tests for about 30s - 1m?

[ludeknovy]

wrt to the scatter chart does my test need to run for a significant amount of time for it to show data, I was just testing things and running my tests for about 30s - 1m?

Yes, I think the scatter chart needs more data to be rendered if I remember correctly.

[afaderoTextNow]

@ludeknovy QQ - My locust load tests are organized in different files in terms of function areas and I added listeners at the end of each of the files which have different scenarios. I have the ability to run individual tests and what I'm seeing is that my tests are being reported to the incorrect scenario or should I say the last scenario listed (see logs below). In this case I'm running the user_profiles tests but it's reporting to the user_sessions

[2025-05-29 12:19:45,608] TN00939/INFO/locust.main: Starting Locust 2.37.6
[2025-05-29 12:19:45,609] TN00939/INFO/locust.main: Starting web interface at http://0.0.0.0:8089, press enter to open your default browser.
scenario_name user_sessions
scenario_name messaging
scenario_name user_profiles
[2025-05-29 12:19:45,617] TN00939/INFO/locust.main: Locust is running with the UserClass Picker Enabled