api: fix description for ORCA utilization precedence and document weight derivation#46002
api: fix description for ORCA utilization precedence and document weight derivation#46002jukie wants to merge 4 commits into
Conversation
Signed-off-by: jukie <10012479+jukie@users.noreply.github.com>
|
CC @envoyproxy/api-shepherds: Your approval is needed for changes made to |
Signed-off-by: jukie <10012479+jukie@users.noreply.github.com>
Signed-off-by: jukie <10012479+jukie@users.noreply.github.com>
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request updates the protobuf and RST documentation for client-side weighted round-robin (CSWRR) and load-aware locality load balancing policies to clarify how endpoint utilization is resolved from ORCA reports, including the default precedence of named metrics over application utilization. The review feedback correctly identifies an inaccuracy in the updated documentation: the code actually checks if the final utilization (resolved utilization plus any error penalty) is greater than 0, rather than just the resolved utilization itself. Actionable suggestions were provided to correct this in both the proto comments and the RST documentation.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> Signed-off-by: Isaac Wilson <isaac.wilson514@gmail.com>
Commit Message: api: fix description for ORCA utilization precedence
Additional Description:
The
metric_names_for_computing_utilizationfield comments in client_side_weighted_round_robin/load_aware_locality still describe the legacy utilization-precedence order. After #44196,envoy.reloadable_features.orca_weight_manager_use_named_metrics_firstnow defaults to true so the described order and gating was incorrect. They now match the runtime-dependent behavior of the sharedgetUtilizationFromOrcaReporthelper:application_utilization, used when > 0cpu_utilizationAlso includes clarification about rps_fractional requirement and weight derivation.
Risk Level: Low
Testing: N/A
Docs Changes: Updated load_aware_locality docs
Release Notes: None
Platform Specific Features: N/A
API Considerations: Comment only changes