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

Add initial delay when loading python functions #8239

Merged
merged 7 commits into from
Feb 21, 2025
Merged

Conversation

taeold
Copy link
Contributor

@taeold taeold commented Feb 21, 2025

Anecdotally, python function discovery is flakey. We propose 2 change in this PR:

  1. For python discovery, add a small initial delay for python's admin server to boot.

2. Add a request timeout to retry call to retrieve trigger information. Previously, the default timeout would've been set to OS-level TCP timeout, which in my laptop was between 2030s.~ This breaks discovery for function packages that are slow to load.

Before the change, I was able to reliably trigger the flake ~1/10 trigger parsing session. With these changes, I'm haven't been able to trigger the flake (N=~50, YMMV).

Anecdotally, python function discovery is flakey. We propose 2 change in
this PR:

1. For python discovery, add a small initial delay for python's admin
server to boot.

2. Add a request timeout to retry call to retrieve trigger information.
Previously, the default timeout would've been set to OS-level TCP
timeout, which in my laptop was between 20~30s.
@taeold taeold requested a review from inlined February 21, 2025 01:21
@taeold taeold changed the title Improve robustness of function discovery for python Add initial delay when loading python functions Feb 21, 2025
@taeold taeold enabled auto-merge (squash) February 21, 2025 22:34
@taeold taeold merged commit caaf1af into master Feb 21, 2025
51 of 56 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants