question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

arrivalCount is not working as expected

See original GitHub issue

hi@all,

i’m using the arrivalCount to simulate a fixed amount of users in a defined duration, like the documentation says (https://www.artillery.io/docs/guides/guides/test-script-reference#fixed-number-of-arrivals-per-second)

The load test i want to simulate requires the usage of arrivalCount, because arrivalRate creates at least one user per second.

For exmaple this should generate 10 users over 5 minutes:

  phases:
    - name: '10 user in 5 minutes (1 user every 30 seconds)'
      duration: 300
      arrivalCount: 10

But at the end, only three users were created:

vusers.created_by_name.S5: .................................. 1
vusers.created.total: ....................................... 3
vusers.completed: ........................................... 3
vusers.created_by_name.S6: .................................. 2

Another thing i recognized is the slow start when using arrivalCount. It needs up to two minutes before the first request is sent.

This leads to the questions:

  • Do i expect the wrong behaviour?
  • Can this depend on the scenarios i implemented?
  • Is there something wrong in my implementations?
  • Could this just be a bug in artillery?

Issue Analytics

  • State:closed
  • Created 2 years ago
  • Comments:6 (2 by maintainers)

github_iconTop GitHub Comments

1reaction
hassycommented, Dec 17, 2021

The results of the second run you posted look OK.

There’s no output between (1) and (2) because Artillery didn’t generate any VUs in that time period. As soon as one VU was created (some time between 06:27:40 and 06:27:50) and some metrics were generated, we got a report.

The VU is reported as failed because there was a 503 code returned for a request, i.e. it did not complete its scenario successfully.

Hope that makes sense!

1reaction
hassycommented, Dec 17, 2021

hi @BernhardMaier 👋 thank you for a very detailed bug report (I would love it if every bug report we get for Artillery was that detailed!)

I’m going to reopen the issue as it does look like there’s an issue in the first report you shared - we’d expect vusers.created.total to equal 10.

Read more comments on GitHub >

github_iconTop Results From Across the Web

Test Scripts - Artillery.io
Note that this will not work as expected when running distributed tests, as each node will have its own copy of the CSV...
Read more >
shoreditch-ops/artillery - Gitter
Hi Artilleriers, just started to learn socketio engine, and am having a problem with setting the query . Because my socket server uses...
Read more >
draft-ietf-ippm-duplicate-04 - IETF Datatracker
Network Working Group H. Uijterwaal Internet-Draft RIPE NCC Intended status: ... It is thus not included in the Type-P-one-way-packet-arrival- count. 2.5.
Read more >
PFA_SMF_ARRIVAL_RATE - IBM
The SMF arrival rate check issues an exception using the following four ... When PFA determines the SMF rate is lower than expected,...
Read more >
Time-Domain Analysis of Web Cache Filter Effects - CiteSeerX
sults show that (as expected) a Web cache reduces both the ... perceivable network performance problems. ... to alleviate these performance problems.
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found