Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

test: fix test-worker-memory.js for large cpu #s #27090

Closed
wants to merge 2 commits into from

Conversation

Projects
None yet
6 participants
@mhdawson
Copy link
Member

commented Apr 4, 2019

This test consistently failed on a system with a
large number of cores (~120). Cap the number of
concurrent workers so we'll stay consistently within
the "slack" allowed with respect to rss.

Checklist
  • make -j4 test (UNIX), or vcbuild test (Windows) passes
  • tests and/or benchmarks are included
  • commit message follows commit guidelines
test: fix test-worker-memory.js for large cpu #s
This test consistently failed on a system with a
large number of cores (~120). Cap the number of
concurrent workers so we'll stay consistently within
the "slack" allowed with respect to rss.
@nodejs-github-bot

This comment has been minimized.

@nodejs-github-bot

This comment has been minimized.

Show resolved Hide resolved test/parallel/test-worker-memory.js Outdated
@gireeshpunathil

This comment has been minimized.

Copy link
Member

commented Apr 5, 2019

As a side note: I had come across the same issue and attempted to address it in a similar manner, but did not proceed when I realized that there is another problem, that may be harder to solve: when too many workers run in truly parallel manner, the rss growth pattern is observed to be non-uniform and non-deterministic (not a function of the worker count, but the degree of parallelism - order in which threads were accessing the memory, scheduling, the time gap between worker exeution and the rss data collection etc.

This fix looks reasonable; but if the issue recurs even with this, we should suspect the non-deterministic behavior of rss and may be we should fix the number of workers to a lowest possible number of CPUs in our CI.

@nodejs-github-bot

This comment has been minimized.

Squash: address comment
Co-Authored-By: mhdawson <michael_dawson@ca.ibm.com>
@mhdawson

This comment has been minimized.

Copy link
Member Author

commented Apr 8, 2019

CI was good landing.

@mhdawson

This comment has been minimized.

Copy link
Member Author

commented Apr 8, 2019

See that comment was changed sinc last CI run. Lite CI to be safe: https://ci.nodejs.org/job/node-test-pull-request-lite-pipeline/3198/

@mhdawson

This comment has been minimized.

Copy link
Member Author

commented Apr 8, 2019

Landed as f96a660

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.