## Summary Roll out [the feature](https://gitlab.com/gitlab-org/gitlab/-/work_items/630626) currently behind the `duo_code_review_standard_structured_findings` feature flag. - DRI: @kinsingh - Team Slack channel: `#g_code_review` > [!note] > Process and guidance live in the docs, this issue is just the commands and a place to track the rollout. > "Rolling out" means incrementally enabling the flag on GitLab.com to validate stability, it is not the same as releasing the feature, which happens when the flag is removed. > [Feature flag controls](https://docs.gitlab.com/development/feature_flags/controls/) · [Feature flag lifecycle](https://handbook.gitlab.com/handbook/product-development/how-we-work/product-development-flow/feature-flag-lifecycle/#feature-flag-lifecycle) ## What could go wrong? - The flag routes the user's Standard Duo Code Review runs to `code_review` 4.0.0, which returns findings through the `code_review_findings` schema and posts them with a deterministic publish step, the same as the deep flow. Comment format changes for flagged users: severity and category headers, a computed summary, and a confidence gate. - Enabling the flag before 4.0.0 is deployed to Duo Workflow Service fails every review for flagged users, because the flow version cannot be resolved. This is not inert. - The flow version resolver and the comment processor both check the flag for the same user. If the flag is turned off while a review is running, the JSON output is rejected and the review reports an error instead of posting "nothing to comment on". - The `duo_code_review_deep_flow` flag takes precedence, so users with both flags keep getting the deep flow. - Quality: Qodo evals on 100 MRs showed no regression against earlier Standard runs (hit F1 0.567 vs 0.553-0.562, precision flat). Production comment volume may still shift because the confidence gate now drops findings below 4. - Any regression is contained to flagged users. Disabling the flag returns them to the previous flow version immediately, and no state persists across the change. Detection: Duo Workflow Service logs for `code_review` 4.0.0 failures, Rails logs and Sentry for `Gitlab::Duo::CodeReview::FindingsParser::Error`, Duo Code Review error notes on MRs, and feedback in `#g_ai_coding`. ## Rollout Run all production `/chatops` in [`#production`](https://gitlab.slack.com/archives/C101F3796) and cross-post the results to `#g_ai_coding`. Background: [incremental rollout process](https://docs.gitlab.com/development/feature_flags/controls/#process), [feature actors](https://docs.gitlab.com/development/feature_flags/#feature-actors). Before enabling for anyone, confirm both prerequisites are deployed: 1. ai-assist MR https://gitlab.com/gitlab-org/modelops/applied-ml/code-suggestions/ai-assist/-/merge_requests/7444 deployed to the GitLab-managed Duo Workflow Service fleet. Without it, flagged reviews fail. 1. Monolith MR https://gitlab.com/gitlab-org/gitlab/-/merge_requests/261961 deployed to production. Without it the flag does not exist. **Non-production** ``` /chatops gitlab run feature set duo_code_review_standard_structured_findings true --dev --pre --staging --staging-ref ``` **Production** This rollout is actor based. Step 1: enable for named Code Review team members. Step 2: enable for all internal GitLab team members. Step 3: percentage of actors. ``` /chatops gitlab run feature set --user=<username> duo_code_review_standard_structured_findings true ``` ``` /chatops gitlab run feature set --feature-group=gitlab_team_members duo_code_review_standard_structured_findings true ``` ``` /chatops gitlab run feature set duo_code_review_standard_structured_findings <percentage> --actors ``` ## Before global rollout Confirm the relevant gotchas before going to 100%, see [enabling a feature for GitLab.com](https://docs.gitlab.com/development/feature_flags/controls/#enabling-a-feature-for-gitlabcom): - [ ] [Docs + version history](https://docs.gitlab.com/development/documentation/feature_flags/) updated - [ ] [Breaking changes](https://docs.gitlab.com/development/documentation/release_notes/#deprecations-removals-and-breaking-changes) announced, if any - [ ] [Change management issue](https://handbook.gitlab.com/handbook/engineering/infrastructure-platforms/change-management/#feature-flags-and-the-change-management-process) opened, if required - [ ] [External API consumers](https://docs.gitlab.com/development/feature_flags/#do-not-use-feature-flags-in-external-api-consumers) handled with a fail-open mechanism, if applicable - [ ] Self-managed and Dedicated instances on an older AI Gateway that lacks `code_review` 4.0.0 handled before the flag is default-on or removed ## Cleanup Once rolled out, make `code_review` ^4.0.0 the default flow version for Standard reviews and let the comment processor accept structured output without a flag check. Then remove the flag and its YAML definition from the codebase, and: ``` /chatops gitlab run release check <merge-request-url> <milestone> /chatops gitlab run feature delete duo_code_review_standard_structured_findings --dev --pre --staging --staging-ref --production ``` ## Rollback ``` /chatops gitlab run feature set duo_code_review_standard_structured_findings false # production /chatops gitlab run feature set duo_code_review_standard_structured_findings false --dev --pre --staging --staging-ref # non-production /chatops gitlab run feature delete duo_code_review_standard_structured_findings --dev --pre --staging --staging-ref --production # remove entirely ```