<!--IssueSummary start--> <details> <summary> Everyone can contribute. [Help move this issue forward](https://handbook.gitlab.com/handbook/marketing/developer-relations/engineering/community-contributors-workflows/#contributor-links) while earning points, leveling up and collecting rewards. </summary> - [Label this issue](https://contributors.gitlab.com/manage-issue?action=label&projectId=278964&issueIid=631599) </details> <!--IssueSummary end--> ## Summary Roll out [the feature](https://gitlab.com/gitlab-org/gitlab/-/work_items/617932#note_3933674806) currently behind the `ci_finalize_pending_detach_partitions_daily` feature flag. - DRI: @lma-git - Team Slack channel: `#g_ci-platform` MR that introduced the FF: https://gitlab.com/gitlab-org/gitlab/-/merge_requests/258843 ### What are we expecting to happen? With the flag on, a new worker runs daily at 02:00 UTC, the lowest-traffic hour, and finalizes partitions left pending by an interrupted `DETACH ... CONCURRENTLY` that have a `detached_partitions` row. Today they stay pending until the dropper finalizes them when their drop is due (retention, usually over a week). The dropper keeps that finalize as a fallback, and both call the same finalizer. **[UPDATE 2026-10-07]:** Updated `FinalizePendingDetachPartitionsWorker` to run hourly from `00:17` to `03:17` instead of just once daily. See https://gitlab.com/gitlab-org/gitlab/-/work_items/631599#note_3974805502. With the flag on, the finalizer also first locks the tables the partition's foreign keys reference in `SHARE ROW EXCLUSIVE` mode. That's the same order the app takes them (builds before instances), so `FINALIZE` waits instead of deadlocking. This runs inside `with_lock_retries` (`lock_timeout` of at most 1s per attempt, 20 attempts), which raises when the attempts are exhausted. `FINALIZE` takes this lock anyway. Moving it to the front means we also hold it through `FINALIZE`'s wait for older snapshots. With the flag off, the new worker does nothing and the dropper finalizes exactly as it does today. ## What could go wrong? - Writes to `p_ci_builds` block while we wait for the lock and while we hold it through FINALIZE's wait for older snapshots, for up to about 2s per attempt. We detect this through lock-wait or statement timeouts on builds writes around the worker's run time. - `FINALIZE` keeps timing out because a long transaction holds a snapshot or writers hold the lock. The partition stays pending, which is safe, and the worker retries the next night. We detect this through lock-retry exhaustion logs from the worker. - A deadlock through a lock order we haven't seen, such as the app locking instances before builds. `lock_timeout` turns it into a timeout. We detect it through `ActiveRecord::Deadlocked` from the finalizer. Rollback: turn the flag off. ## Events - [Exceptions with ci_finalize_pending_detach_partitions_daily:1](https://log.gprd.gitlab.net/app/discover#/?_g=(time:(from:now-1d,to:now))&_a=(index:'7092c4e2-4eb5-46f2-8305-a7da2edad090',query:(language:kuery,query:'json.exception.feature_flag_states.keyword:%22ci_finalize_pending_detach_partitions_daily:1%22'),columns:!(json.exception.class,json.exception.message,json.extra.workflow_id,json.meta.caller_id),sort:!(!(json.time,desc)))) - [Events with ci_finalize_pending_detach_partitions_daily:1](https://log.gprd.gitlab.net/app/discover#/?_g=(time:(from:now-1d,to:now))&_a=(index:'7092c4e2-4eb5-46f2-8305-a7da2edad090',query:(language:kuery,query:'json.feature_flag_states:%22ci_finalize_pending_detach_partitions_daily:1%22'),columns:!(json.meta.caller_id,json.feature_flag_states),sort:!(!(json.time,desc)))) - [Error rate and other graphs by modifying the examples in the Visualization Library](https://log.gprd.gitlab.net/app/visualize#/?_g=h@358d019&s=FF%20Observability) ## Rollout Run all production `/chatops` in [`#production`](https://gitlab.slack.com/archives/C101F3796) and cross-post the results to `#g_ci-platform`. Background: [incremental rollout process](https://docs.gitlab.com/development/feature_flags/controls/#process), [feature actors](https://docs.gitlab.com/development/feature_flags/#feature-actors). **Non-production** ``` /chatops gitlab run feature set ci_finalize_pending_detach_partitions_daily true --dev --pre --staging --staging-ref ``` **Production** — the flag uses the `:instance` actor, so it is on or off for the whole instance. Enable it before the worker's 02:00 UTC run and watch that run: ``` /chatops gitlab run feature set ci_finalize_pending_detach_partitions_daily true ``` ## 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 ## Cleanup Remove the flag once [deemed stable](https://handbook.gitlab.com/handbook/product-development/how-we-work/product-development-flow/feature-flag-lifecycle/#feature-flag-lifecycle) — see [cleaning up](https://docs.gitlab.com/development/feature_flags/controls/#cleaning-up). Track it here, or open a follow-up [Feature Flag Cleanup issue](https://gitlab.com/gitlab-org/gitlab/-/work_items/new?description_template=Feature%20Flag%20Cleanup). Remove the flag and its YAML definition from the codebase, then: ``` /chatops gitlab run release check <merge-request-url> 19.5 /chatops gitlab run feature delete ci_finalize_pending_detach_partitions_daily --dev --pre --staging --staging-ref --production ``` ## Rollback ``` /chatops gitlab run feature set ci_finalize_pending_detach_partitions_daily false # production /chatops gitlab run feature set ci_finalize_pending_detach_partitions_daily false --dev --pre --staging --staging-ref # non-production /chatops gitlab run feature delete ci_finalize_pending_detach_partitions_daily --dev --pre --staging --staging-ref --production # remove entirely ```