🏥 CI FailureTransient HTTP 502 errors during GitHub CDN downloads

[github-actions]

This issue tracks transient HTTP 502 Bad Gateway errors that occur during downloads from GitHub's CDN (releases, Copilot CLI, etc.). These are infrastructure failures, not code issues.

Latest Occurrence

Run: 22031343904
Date: 2026-02-15 06:52 UTC
Job: detection (Issue Monster workflow)
Error:

curl: (22) The requested URL returned error: 502
##[error]Process completed with exit code 22.
```

The job failed while downloading GitHub Copilot CLI v0.0.405 from:
```
https://github.com/github/copilot-cli/releases/download/v0.0.405/copilot-linux-x64.tar.gz

Pattern Analysis

This is the 3rd documented occurrence in the past week:

Run ID	Date	Workflow	Download Target	Status
21833584727	2026-02-09	CodeQL	git clone (checkout)	HTTP 500/502
22024528697	2026-02-14	Test Action	awf binary (v0.7.0)	HTTP 502
22031343904	2026-02-15	Issue Monster	Copilot CLI (v0.0.405)	HTTP 502

Common characteristics:

• All errors occur during downloads from GitHub infrastructure (releases, git, etc.)
• Self-resolving within minutes (verified with post-failure curl tests)
• No code or configuration issues identified
• Typically affects 1 job while other jobs succeed
• Install scripts already have retry logic (3 attempts)

Impact

• Severity: Low (self-resolving, no data loss)
• Frequency: ~3 occurrences per week (based on documented failures)
• Affected Jobs: Typically 1 out of 4-7 jobs per workflow
• Resolution Time: <5 minutes (automatic recovery)

Root Cause

GitHub infrastructure/CDN transient failures. Not a problem with:

• ✅ Repository code or configuration
• ✅ Workflow definitions
• ✅ Release assets (verified to exist and be valid)
• ✅ Network configuration
• ✅ Retry logic (already implemented in install scripts)

Recommended Actions

Short Term

• No immediate action required - these failures self-resolve
• Re-run failed workflows when they occur
• Continue monitoring frequency via cache-memory entries

Long Term (if frequency increases)

1. Check GitHub Status: Monitor (www.githubstatus.com/redacted) during failures
2. Add exponential backoff: Enhance retry logic with delays between attempts
3. Implement fallback mirrors: Consider caching critical binaries
4. Escalate to GitHub Support: If frequency exceeds 5 occurrences per week

Monitoring

All occurrences are logged to /tmp/gh-aw/cache-memory/ci-failure-*.md for pattern tracking.

Latest analysis: ci-failure-22031343904.md

---

📋 Classification: Infrastructure/Transient
⏱️ Auto-resolves: Yes (typically <5 minutes)
🔄 Action Required: None (monitor only)

---

🏥 Automatically tracked by CI Doctor

AI generated by CI Doctor