rusackas commented on code in PR #40234:
URL: https://github.com/apache/superset/pull/40234#discussion_r3279006290
##########
.github/workflows/bashlib.sh:
##########
@@ -187,8 +190,48 @@ cypress-run-all() {
fi
export CYPRESS_BASE_URL
- nohup flask run --no-debugger -p $port >"$flasklog" 2>&1 </dev/null &
- local flaskProcessId=$!
+ nohup gunicorn \
Review Comment:
Good call — pushed a commit aligning to that. Now using `--workers 1
--worker-class gthread --threads 20` (matching the entrypoint defaults). The
remaining divergences are intentional for E2E load and called out in a comment
block above the invocation:
- `--timeout 120` — heavy dashboard import/export specs blow past the 60s
default
- `--max-requests 500 --max-requests-jitter 50` — recycle the worker under
sustained test load
- `superset.app:create_app()` — explicit factory so we don't rely on
`FLASK_APP` being exported in the test env
Bonus: dropping from 4 workers to 1 also looks to fix the two SQL Lab →
Explore playwright failures that were caused by cross-worker timing races.
Thanks for catching this!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
