✅ Snyk checks have passed. No issues have been found so far.

Status	Scanner	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.

Greptile Overview

Greptile Summary

This PR fixes a critical issue with audit log streaming where retry attempts were failing due to conflicting timeout mechanisms. The change removes AbortSignal.timeout() from all audit log stream providers (Splunk, Azure, Datadog, Cribl, Custom), keeping only the axios timeout option.

Problem identified: When axios-retry attempted to retry failed requests, it reused the same AbortSignal which had already started counting down or expired, causing immediate cancellation of retry attempts with ERR_CANCELED errors. This created memory pressure from accumulated timer handles.

Changes made:

• Removed signal: AbortSignal.timeout(AUDIT_LOG_STREAM_TIMEOUT) from all provider factories
• Kept timeout: AUDIT_LOG_STREAM_TIMEOUT (5 seconds) which is sufficient for request timeouts
• Applied consistently across 5 provider implementations

Impact: This fix ensures retries work correctly with fresh timeouts per attempt, resolving memory pressure issues and allowing proper retry behavior when downstream services return errors like 503s.

Confidence Score: 5/5

• This PR is safe to merge with minimal risk - it fixes a legitimate bug in retry behavior
• The change removes redundant timeout mechanism that was causing retry failures. The axios timeout option alone provides sufficient timeout protection, and this fix resolves memory pressure and ERR_CANCELED errors. The change is consistent across all providers, maintains backward compatibility (no API changes), and aligns with axios-retry best practices. No security concerns introduced.
• No files require special attention

Important Files Changed

_{Last reviewed commit: a75dfb9}