## Current state - [Infrastructure is managed by the team](https://gitlab.com/gitlab-org/quality/triage-ops/-/blob/master/doc/reactive/production.md?ref_type=heads). - The `triage-ops` service isn't integrated into GitLab. It's reacting –through the public API– to webhooks sent from `gitlab-org` & `gitlab-com` groups. - It also has a "scheduled" part, which is also an ad-hoc solution, based on [the `gitlab-triage` gem](https://gitlab.com/gitlab-org/ruby/gems/gitlab-triage) and pipeline schedules. ## Proposal - :rocket: https://gitlab.com/gitlab-org/quality/triage-ops/-/issues/471+s - https://gitlab.com/gitlab-org/quality/engineering-productivity/team/-/issues/343+s - This allows groups to be in control of their data and reports without involving the EP team. - Infrastructure improvements: - Better isolate infrastructure to its own GCP project: https://gitlab.com/gitlab-org/quality/triage-ops/-/issues/1179+s - https://gitlab.com/groups/gitlab-org/-/epics/11434+s - https://gitlab.com/gitlab-org/quality/engineering-productivity-infrastructure/-/issues/103+s - [Runway project](https://gitlab.com/gitlab-com/gl-infra/platform/runway) - [Runway blueprint](https://docs.gitlab.com/ee/architecture/blueprints/runway/) - Integrate [Tanuki Stan](https://gitlab.com/gitlab-org/ml-ops/tanuki-stan) into the service. - team~9032372 [Integrate `triage-ops` as a GitLab feature](https://gitlab.com/groups/gitlab-org/-/epics/7487) that supports both the reactive and scheduled use-cases. Potential solutions: - [Temporal PoC/showcase](https://gitlab.com/gitlab-org/ci-cd/section-showcases/-/issues/54). - [GitLab Steps](https://docs.gitlab.com/ee/architecture/blueprints/gitlab_steps/). - [GitLab CI Events](https://docs.gitlab.com/ee/architecture/blueprints/gitlab_ci_events/) (based on [GitLab Events Platform](https://docs.gitlab.com/ee/architecture/blueprints/gitlab_events_platform/)). - [No code/low code](https://about.gitlab.com/direction/create/nolowcode/).