Track GitHub CI performance degradation and anomalies

[coderabbitai]

Summary

Track performance degradation and anomalies in GitHub CI.

Rationale

GitHub CI duration and reliability can change over time because of code changes, dependency changes, runner variability, network conditions, or infrastructure behaviour. The project needs a consistent method to detect, investigate, and report these changes.

Affected areas

• GitHub Actions workflows in this repository
• CI job and workflow duration data
• CI failures, retries, timeouts, and unusual execution patterns
• Performance-sensitive end-to-end and integration test workloads

Required changes

1. Define the CI workflows, jobs, and measurements to track.
2. Collect or query historical CI execution data.
3. Establish a baseline and thresholds for expected duration and reliability.
4. Detect and report performance regressions and anomalous CI runs.
5. Document an investigation procedure that identifies whether an anomaly is caused by the repository, test workload, GitHub-hosted runners, or external services.

Acceptance criteria

• The project tracks duration and outcome data for the selected CI workflows or jobs.
• The project defines baseline values and anomaly thresholds.
• The project provides a repeatable method to identify performance degradation.
• The project provides a record or report for detected anomalies.
• The project documents the escalation or investigation process.

References

• Pull request fix(huntsman): Enable TCP_NODELAY on the storage and scheduler gRPC servers' accepted sockets. #459: fix(huntsman): Enable TCP_NODELAY on the storage and scheduler gRPC servers' accepted sockets. #459
• Request comment by @sitaowang1998 on pull request fix(huntsman): Enable TCP_NODELAY on the storage and scheduler gRPC servers' accepted sockets. #459: fix(huntsman): Enable TCP_NODELAY on the storage and scheduler gRPC servers' accepted sockets. #459