Skip to content

Improve error handling and cleanup logic for benchmarks#813

Draft
zhixiangli wants to merge 3 commits intofsspec:mainfrom
zhixiangli:zhixiangli/benchmark-clean-up-on-failure
Draft

Improve error handling and cleanup logic for benchmarks#813
zhixiangli wants to merge 3 commits intofsspec:mainfrom
zhixiangli:zhixiangli/benchmark-clean-up-on-failure

Conversation

@zhixiangli
Copy link
Copy Markdown
Collaborator

@zhixiangli zhixiangli commented Apr 15, 2026

This PR improves error handling and cleanup logic for benchmarks. It depends on #810, and the diff will be clearer once #810 is merged.

- Introduced `_BENCHMARK_TYPE` substitution variable (defaults to "IO").
- Modified existing VM creation step (4a) to only run when type is "IO" (uses c4-standard-192).
- Added new VM creation step (4b) for "METADATA" benchmarks (uses c4-standard-8).
- Updated `run-benchmarks` to wait for both VM creation steps.
- Replaced `/workspace/failure` with `/workspace/FAILED` to track failing steps.
- Added `_SKIP_IF_FAILED` check to skip benchmark execution if prior steps fail.
- Added `allowFailure: true` to benchmark steps to ensure cleanup steps always run.
- Updated cleanup steps (SSH key, VM, buckets) to only run if respective resources were initialized.
- Added robust error reporting in the final step to list failed steps.
@zhixiangli zhixiangli marked this pull request as draft April 15, 2026 12:22
@codecov
Copy link
Copy Markdown

codecov bot commented Apr 15, 2026

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 79.81%. Comparing base (5596df9) to head (1bb8f20).

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #813      +/-   ##
==========================================
+ Coverage   79.55%   79.81%   +0.26%     
==========================================
  Files          16       16              
  Lines        3042     3042              
==========================================
+ Hits         2420     2428       +8     
+ Misses        622      614       -8     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Comment thread cloudbuild/benchmarks/benchmarks-cloudbuild.yaml Outdated
Comment thread cloudbuild/benchmarks/benchmarks-cloudbuild.yaml Outdated
@zhixiangli
Copy link
Copy Markdown
Collaborator Author

/gcbrun

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