Need Help for Generating and uploading JTL file with run locust more than 1 workers

[RifkiNurgraha]

i use jtl reporter for locust framework. but if i use workers more than one. csv log only create one not mix from all workers. any insight about this ? thank you

[ludeknovy]

Hi @RifkiNurgraha what version of locust do you use and what listener do you use?

[ludeknovy]

Run it using your repository (test_user2.py) and it still looks correct to me:

The CSV has +1 line because of the header.

I ran it with 3 workers.

Btw. in the repository you dont use -u flag with number of users. It defaults to 1. It means only 1 worker will generate the load, the others will be idle.

[RifkiNurgraha]

sorry, I think we misunderstood 😂 . let me explain about the issue again.

• first i run my script (test_user2.py) with 3 workers.
  
  as you can see at the picture, i swarm 100 user with have 3 workers each produce 33 users.
• second i stop the test. and automatically upload to jtl dasboard , using jtl_listener.py and upload_jtl.py at 'locust-pt/libs/'
  
  but as you can see in the dashboard the result of total user only 34, which is the correct is 100 ( total from 3 workers), right ?

any solution regarding this ?

[ludeknovy]

Ahh, ok. I see the issue now.

You need to provide "Hostname" column.

Add self._worker_id = random.randint(10000,9999999) to the JtlListener constructor.

Then add Hostname to the csv_headers. And in add_result method add Hostname = str(self._worker_id) and in the same method put Hostname into row.

After doing this It showed me 100 VU in the app.

I will prepare fix for it later. Thanks for raising this issue.

[RifkiNurgraha]

Thanks a bunch for your fast response. you nailed it. 🥳

[ludeknovy]

@RifkiNurgraha could you please mark it as answered, please?

Thanks again for reporting it!