Allow k6 scenario to be defined within k6 test #1797
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #1797 +/- ##
==========================================
+ Coverage 89.84% 89.86% +0.03%
==========================================
Files 71 71
Lines 19597 19631 +34
==========================================
+ Hits 17604 17639 +35
+ Misses 1993 1992 -1 ☔ View full report in Codecov by Sentry. |
…fix/k6-scenario-within-k6-test
…fix/k6-scenario-within-k6-test
|
Hi, I have fixed up the tests and improved code coverage. All checks are now passing, thanks. |
|
Can someone please review this? Thanks |
|
Hi, are any maintainers able to review this PR please? |
set dev version to 0.0.1-DEV to allow local install
…fix/k6-scenario-within-k6-test
…enpearson84/taurus into fix/k6-scenario-within-k6-test
|
Any maintainers able to review this PR please? |
Sync upstream changes
…fix/k6-scenario-within-k6-test
upstream changes to my branch
bzt/modules/k6.py
Outdated
|
|
||
| def startup(self): | ||
| cmdline = [self.k6.tool_name, "run", "--out", f"csv={self.kpi_file}"] | ||
| cmdline = [self.k6.tool_name, "run"] #, "--out", f"csv={self.kpi_file}"] |
There was a problem hiding this comment.
This change breaks Taurus' handling of K6 results when more than one scenario is executed in parallel.
In the previous version, Taurus managed the creation of files like kpi-1.csv, kpi-2.csv, etc. automatically.
With this change, we now have to manually configure this setup.
Error:
Child Process Error: Empty results, most likely xxx (K6Executor) failed. Actual reason for this can be found in logs under...
(but the execution was a success. However, due to this error status, the execution is considered a failure in the pipeline)
For a complex execution involving multiple scenarios running in parallel, this increases the complexity of the user's code for something that was previously managed internally.
Therefore, I’d like to understand the reasoning behind removing this default functionality, if possible. Thank you.
There was a problem hiding this comment.
Yeah Im aware of this. However, this PR has been raised for over a year with no communication from the project maintainers at all so we have been using this Fork instead. We decided we dont need the results output to file as we output to an InfluxDB instead and we dont care about the overall Taurus result so this 'quick' fix works for us.
If this is going to be seriously considered for merging then Ill spend some more time looking at it but Im not sure the project maintainers are interested.
There was a problem hiding this comment.
@aldrichksb this has now been made optional. Disabled via the modifications -> disable block inside the scenarios block.
…fix/k6-scenario-within-k6-test
|
Just keeping this updated with main |
5 participants
Allow for load scenario to be defined within the K6 script as well as in the Taurus yml file. Currently if no iterations is defined in the Taurus scenario then only 1 iteration is executed before the test ends. This change adds k6 its own get-load method to allow for infinite iterations.
Closes #1791