**Next sprint:** https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1464\ **Previous sprint:** https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1352\ **W38 results:** [end-of-week note](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1352#note_3873970957)\ **Defect tracking:** [&23325 Artifact Registry Closed Beta — Defect Tracking](https://gitlab.com/groups/gitlab-org/-/work_items/23325)\ **Roadmap:** [docs/roadmap/closed-beta.md](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/roadmap/closed-beta.md)\ **Pending-work inventory:** [`ar-virtual-registries-pending-work`](https://docs.google.com/spreadsheets/d/19ubu_Aju4xPrHpF_QZyyjkzuSZK_FIcvlAbdeyr0hZM/edit?gid=1998918332#gid=1998918332) --- > ### Corrections and decisions, 2026-09-23 > > **1. This issue's monolith/S05 figures were wrong when it was written, and are corrected here.** > It recorded steps 5, 6 and 7 as not started with step 8 in review. In fact all four had merged or > were merging: step 6 on 09-17, step 5 on 09-18, step 7 on 09-21 and step 8 on 09-22. **monolith/S05 > repository detail, virtual, is complete at 8/8.** The error came from reading the *issues'* > assignee state rather than merged MRs — the issues were unassigned while the MRs were landing. Take > stream progress from merged MR titles, not from issue assignment. > > **2. The six unowned items stay unowned, by decision.** `### Owners to assign, Monday` below asked > for owners on [#894](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/894), [#1356](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1356), > [#991](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/991), [#992](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/992) and the monolith/S05 step issues. The > monolith/S05 three are moot — that work merged. The remaining four are **deliberately left without > an owner for now**; they are not oversights and do not need chasing this week. > > **3. The daily cadence is dropped.** It had not run in W37 or W38 — neither issue received a note > during its window — so `## Cadence` below is now a plain day-by-day plan with no daily-post > obligation. The Friday review and the per-DRI status thread carry the same information. > > **4. Closed-beta dates moved.** Hosted and remote capabilities for all three formats were > **delivered 2026-09-16**, and the AR DRI set **2026-10-02** as the closed-beta target on > 2026-09-23. Virtual and everything else outstanding is fast-follow. > [!2893](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2893) carries it. > > **5. Observability is in scope — correcting point 4 as first written.** An earlier revision of > this banner said observability not complete at launch was out of closed-beta scope. **That was > wrong.** S30 19a/19b, S31 18 and S32 17 are **in** scope and on this week's list: they are the > telemetry `closed-beta-scope.md`'s virtual cache-hit ≥70% exit criterion reports, so the criterion > is measurable at launch only if they ship with it. **The in-scope count is 38, not 34** — only the > three e2e-harness steps are excluded. ## Scope of this plan **This is a virtual-repository sprint, and it has one aim: finish virtual.** Every other stream carries its tail, and nothing new opens outside virtual. The week starts from a real position rather than a standing start. W38 put virtual explicitly out of scope and it moved further than anything in scope did: **22 of the 88 rows** on the 2026-09-15 pending-work inventory landed, the P0 that inventory named as its single highest-leverage item (`virtual.Collect`, S13 step 4) merged on 09-17, S30 went from nothing started to **11 steps merged**, and the monolith S07 plan merged, lifting the guardrail-3 freeze on all 14 of its steps. **51 rows stand between here and a virtual registry a partner can use**, counted at `origin/main` `ed16a3f4c`: 47 with no MR, 4 in review. The three e2e-harness steps (S30 20, S31 19, S32 18) and the 13 conformance catalog rows are **out of scope this week** — see `## Explicitly not in scope`. ## Where virtual actually stands **Updated 2026-09-23 against `origin/main` `bb10ea3a9`.** Re-derived from **merged MR titles**, not the plan Status tables, several of which lag their streams badly (`## Plan-table hygiene`). The **Mon 09-21** column is what this issue opened with; **Now** is today. | Stream | Total | Merged Mon | Merged now | To do (in scope) | Owner | |---|---|---|---|---|---| | S13 collection mode | 6 | 4 | **6 ✅** | 0 | `@10io` | | S30 Maven virtual | 25 | 11 | **12** | **12** | `@mkhalifa3` `@jdrpereira` | | S31 npm virtual | 19 | 12 | **12** | **6** | `@10io` `@dmeshcharakou` | | S32 container virtual | 20 | 9 | **11** | **8** | `@radbatnag` `@sylviashen` | | S17 Phase 6 (management API) | 42 | 40 | **41** | **1** | `@hswimelar` | | monolith/S05 virtual detail | 8 | 7 | **8 ✅** | 0 | `@fmccawley` | | monolith/S07 Phase 3 CRUD | 14 | 0 | **3** | **11** | `@rchanila` | | **Total in scope** | | | | **38** | | **Three days in, two streams are finished and the count fell from 51 to 38.** S13 collection mode closed 6/6 on 09-21 and monolith/S05 closed 8/8 on 09-22. S32 landed steps 4, 9 and **10a** — 10a being the prerequisite S30 step 10b declared, so the only cross-stream merge order on the board is satisfied. S17 Phase 6 row 41 merged 09-22. **"In scope" excludes exactly three rows** — the e2e-harness steps S30 20, S31 19 and S32 18, which were already out on Monday. **The four observability rows are in scope**: S30 19a/19b, S31 18 and S32 17 stay on the list. **S31 npm virtual has still not moved**, and it is now the only stream at its Monday figure. Step 13 is unblocked and [#894](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/894) is unowned by decision — see `### Owners: left unowned by decision`. **S32's denominator is settled at 20.** The `/19` markers reconciled to `/20` on 09-21, so the "quote *n* of 20 and say so" instruction this issue carried is resolved: 20 is the figure. ## Assignments Set for this week. Where a row below says UNASSIGNED, the issue itself has no assignee — the pairing here is the plan, and assigning the issue is a Monday action. | Format / area | Who | What | |---|---|---| | **npm** | `@10io` `@dmeshcharakou` | S13 steps 5–6 first, then S31 steps 13–18. `@10io` takes S13 because both steps are already assigned to them; S31 step 13 is the handoff point | | **Maven** | `@mkhalifa3` `@jdrpereira` | S30 steps 10a → 19b. 13 rows, the largest single block this week | | **Container** | `@radbatnag` `@sylviashen` | S32 steps 8b, 9, 10a, 10b, 11 remainder, 12, 14–17. **Step 10a first** — see `## The critical path` | | **UI and GraphQL** | `@fmccawley` `@rchanila` `@narendran-kannan` | monolith/S05 steps 5–8 and monolith/S07 steps 1–14 | | **Management API** | `@hswimelar` | Phase 6 rows 41 and 42 are in review; landing them closes Phase 6 at 42/42 merged | | **Testing, Self-Managed** | `@suleimiahmed` | e2e tails; the harness steps are out of scope this week, so this is the existing open set | `@jdrpereira` on Maven follows from the format-specific shape of the remaining work: the resolution engine traversal is shared and landed with the S13 foundations, and what is left per format is its own tables, models, request maps, targets and metadata-merging logic. ## The critical path Four constraints decide the order. Everything else in a stream is that stream's own sequence. 1. **S32 step 10a gates S30 step 10b.** This is a declared cross-stream merge order, it is the only one on the board, and S32 step 10a has no branch. It is small. Landing it early is what keeps Maven's tail from stalling behind Container's. 2. **S31 step 13 is the one row between npm at 12/19 and a demonstrable npm virtual registry.** Its prerequisite `virtual.Collect` merged 09-17, so the block is gone. [#894](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/894) is **unassigned**; assign it before anything else in S31 starts. 3. **[#1356](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1356) gates monolith S07 step 4**, and it is open with no MR and **no assignee**. It is the ordered `upstream_repository_ids` contract on virtual create and update — the ordering semantics the whole upstream feature rests on, so it is not a step that can be worked around. 4. **S13 steps 5 and 6 edit `internal/virtual/resolve.go`**, which an S30 declaration also writes. [!2622](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2622) merged 09-15, so one side of that three-way collision is resolved; the remaining risk grows with every day S30 lands into the same file ahead of them. **What W38 unblocked, so nobody re-derives it:** monolith/S05 step 4 merged, so steps 5, 6 and 7 are free; [!2669](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2669) merged, so guardrail 3 no longer holds any S07 step shut; Phase 6 row 40 merged, so S07 step 14's cross-repo dependency is down to three; the S30 schema steps 1–3 merged, so the inventory's "cross-repo blocker with zero work started" no longer describes them. ## Decisions this week needs ### Routed, not settled here **Is virtual closed-beta scope or a GA fast-follow?** The two authorities disagree and the disagreement decides roughly 70 rows. - [`docs/roadmap/closed-beta.md`](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/roadmap/closed-beta.md) records the posture **settled 2026-09-07 by `@jaime`**: the three virtual formats are fast-follows, and S17 Phase 6 rows 36–42 cannot be launch scope either, being downstream of one. - `unified-artifact-management`'s `closed-beta-scope.md` ([!77](https://gitlab.com/gitlab-org/ci-cd/package-stage/unified-artifact-management/-/merge_requests/77)) still marks virtual registries **IN** with the cut locked, and sets a closed-beta exit criterion of virtual cache hit ≥70% in at least one partner environment. - The pending-work inventory scored **72 of its 88 rows** as `Closed-Beta` on the locked reading and says so explicitly, so anyone sequencing off that sheet is sequencing off the unresolved reading. **This issue does not settle it.** `closed-beta.md` states that reopening the posture is a scope decision belonging in that file **and** in `closed-beta-scope.md`, not in a sprint issue. What this week needs is the MR pair, or an explicit statement that the fast-follow reading stands and the inventory's `Blocking` column is rescored to match. Until one of those lands, treat the week as fast-follow delivery: **it is not the launch gate, and the launch gate is [#1326](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1326) alone.** One consequence worth naming ahead of the decision: the observability steps (S30 19a and 19b, S31 18, S32 17) are the only way to report a cache-hit percentage. If the locked reading wins they are in scope for an exit criterion; if the fast-follow reading wins they are ordinary stream work. They sit at the end of each stream either way, so the decision changes their urgency, not their position. ### Owners: left unowned by decision Six items had no assignee at Monday. **Three are moot** — the monolith/S05 step issues (gitlab#628666, gitlab#628667, gitlab#628668) were unassigned while their MRs merged, and that stream is complete at 8/8. **The remaining four stay unowned for now, by decision on 2026-09-23.** They are recorded here so nobody re-raises them as oversights: | Item | What it blocks | Status | |---|---|---| | [#894](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/894) S31 step 13 | Everything else in S31 | Unowned by decision | | [#1356](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1356) ordered `upstream_repository_ids` | monolith S07 steps 4 and 14 | Unowned by decision | | [#991](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/991) AppSec review: Maven virtual | Partner-facing exposure of Maven virtual | Unowned by decision | | [#992](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/992) AppSec review: npm virtual | Partner-facing exposure of npm virtual | Unowned by decision | **The consequence to hold, not to fix this week:** S31 npm virtual has not moved in two sprints and #894 is the row that would unblock it, so leaving it unowned is a decision that S31 stays at 12/19 unless someone picks it up. [#993](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/993), the container virtual AppSec review, closed in W38, so two reviews remain rather than three. ## Plan-table hygiene Virtual is where the Status-table drift now is, and the single-writer convention means no step MR may fix it — the rows are recorded in batch from a dedicated `docs(plans)` branch. At `ed16a3f4c`: | Plan | Recorded | Merged | Gap | |---|---|---|---| | [S30 Maven virtual](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/plans/2026-09-10-s30-maven-virtual.md) | 0/25 | 11 | **11 — the whole table is blank** | | [S32 container virtual](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/plans/2026-08-27-s32-container-virtual.md) | 6 of 20 rows | 9 | **3** (steps 2, 3, 4) | | [monolith/S05 virtual detail](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/plans/monolith/2026-09-10-repository-detail-virtual.md) | 0/8 | 4 | **4** — steps 1 and 2 merged 09-14, 3 and 4 in the same window | | [S13 collection mode](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/plans/2026-09-04-virtual-collection-mode.md) | 4/6 | 4 | 0 | | [S31 npm virtual](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/plans/2026-08-21-s31-npm-virtual.md) | 12/19 | 12 | 0 | **S17 Phase 6 has the inverse problem, and any figure quoted off it needs a correction stated beside it.** Its table reads **42/42**, but rows 41 and 42 cite [!2780](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2780) and [!2778](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2778), both **open**. Recording at MR-open time is what the convention asks for, so the table is correct and the headline is not: Phase 6 is **40/42 merged**. [!2664](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2664) — 50 rows across 12 monolith plans — has been open and green since W38. While it is open, every monolith progress figure quoted this week comes from an unmerged document. ## Tails from W38 Carried, not the focus. Each is one action. | Item | State at 2026-09-21 | The action | |---|---|---| | [#1326](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1326) org role tuples — **the whole launch gate** | `~priority::1`, `@bmarjanovic`. [gitlab!255330](https://gitlab.com/gitlab-org/gitlab/-/merge_requests/255330) merged 09-15; [gitlab!255373](https://gitlab.com/gitlab-org/gitlab/-/merge_requests/255373) open, out of Draft, untouched since 09-16 | Land !255373. It is the last thing on the gate | | [#1326](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1326) epic parent | Still **not a child** of [&23325](https://gitlab.com/groups/gitlab-org/-/work_items/23325): the epic's open children carry `~AR-Blocks::Closed-Beta` 0 times while the label query returns 1 | Attach it. Third sprint this has been listed | | [readiness!85](https://gitlab.com/gitlab-org/architecture/readiness/-/merge_requests/85) | Open, non-draft, reports `mergeable`, **187 days** | Merge it, or name the reviewer it waits on | | Severity retirement | **Not decided.** 90 open issues carry `~severity::*` (`::2` on 3, `::3` on 32, `::4` on 55), down from 98 **by closure** | Sweep, or amend the policy. A third sprint of drift is a decision taken by default | | 17 unlabelled children of [&23325](https://gitlab.com/groups/gitlab-org/-/work_items/23325) | 7 of the 17 also carry no `~priority::*` | Two labels each | | Monolith review tail | [gitlab!253063](https://gitlab.com/gitlab-org/gitlab/-/merge_requests/253063) (20 days) and [gitlab!253638](https://gitlab.com/gitlab-org/gitlab/-/merge_requests/253638) (17 days), both green, **neither with an approval** | Two reviews. Third sprint | | [#1140](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1140), [#783](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/783) | #1140 gained `~priority::2` but no assignee; #783 has neither a label nor an assignee and is untouched since 2026-08-21 | Assign both | | Roadmap sync | [`docs/roadmap/closed-beta.md`](https://gitlab.com/gitlab-org/ops/artifact-registry/-/blob/main/docs/roadmap/closed-beta.md) still carries **W37** results | Run the `### Weekly sync workflow` for W38 | | Open MR queue | **44** non-bot non-draft, up from 36, while 201 merged | The week opened more than it reviewed. Watch it, do not chase it | ## Explicitly not in scope - **Virtual conformance — all 13 catalog rows, and the `rc#83` spec and `rc#84` plan that gate them.** Deferred until virtual implementation is done, by decision for this sprint. They are quality gates against behaviour that does not exist yet: `rc#83` is unwritten, `rc#84` is unwritten and unassigned, and 10 of the 13 rows name an S30, S31 or S32 step as their subject. One row also carries a live contradiction to record before anyone picks it up — `oci/tag-list-merge` tests a `tags/list` merge on virtual that ADR-009 makes unavailable on virtual in the MVP, so either the catalog row or the ADR is wrong. - **The three e2e harness steps** — S30 20, S31 19, S32 18. Quality gates rather than functional gates, and the inventory scores all three GA. - **`~AR-Blocks::GA`** — 39 open. Deferred, however cheap a fix looks. This includes [#1334](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1334), the only virtual item carrying an `~AR-Blocks::*` label at all, and [#1177](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1177) allow/deny rule CRUD. - **New workstream starts outside virtual.** No spec, plan or step MR opens this week for anything that is neither virtual nor a named tail above. ## Cadence The week runs **Mon 09-21 → Fri 09-25**, with Friday as the review. **There is no daily post** — it had not run in W37 or W38, so it is dropped rather than carried a third time. The Friday review and the per-DRI status thread below carry the same information. - [ ] **Mon 09-21** — assign the six unowned items in `### Owners to assign, Monday`; start S32 step 10a, since S30 step 10b waits behind it; get reviewers on [!2780](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2780) and [!2778](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2778) - [ ] **Tue 09-22** — S31 step 13 opened; S13 steps 5 and 6 opened; the scope decision routed as an MR pair or an explicit statement that fast-follow stands - [ ] **Wed 09-23** — mid-week read on the 47: what has an MR, what has not, and which stream needs a body moved. API and frontend fit checked against each other while both are still open - [ ] **Thu 09-24** — the `docs(plans)` batch that records S30's 11 rows, S32's 3 and monolith/S05's 4; [!2664](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2664) merged - [ ] **Fri 09-25** — review against the 51: landed, in review, still nothing. Name what carries to W40 and why ### Weekly Status Updates DRIs post as a thread on this issue. ``` ## W39 Update — {Format / area} - My stream at Monday: {n} rows to do. Now: {n} - Merged since yesterday: - Open and waiting on a reviewer — name the reviewer: - Blocked, and on what: - Will your stream be complete by Friday 2026-09-25? yes / no / at risk ``` --- ## Metrics > Fill in Friday 2026-09-25. The "Last week" column is the verified W38 set. > > **Take the merged count from the API's `merged_after`/`merged_before`, not from > `git log --first-parent`.** The merge train went live 2026-09-16: 12 of W38's 201 merges arrived > as second parents of an `Automatic merge` chain, so first-parent read 189. Use first-parent only > to separate merges into stacked parent branches, which were 0 in W38 and 63 in W37. | Metric | This week | Last week | Trend | |--------|-----------|-----------|-------| | MRs merged (feat/fix/docs) | | 146 (125 excl. bots) | | | MRs merged (other/bot) | | 55 (28 excl. bots) | | | Stacked-branch MRs excluded | | 0 | | | MRs open / in review | | 44 non-bot non-draft | | | Review queue depth (oldest open MR, days) | | 37 days ([!1598](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/1598)) | | | Specs approved (of total needed) | | 34/56 Approved; **41/56** approved-or-better | | | **Virtual rows to do** | | **51** at 2026-09-21 → **38** at 2026-09-23 | | | … S13 / S30 / S31 / S32 | | 0 / 12 / 6 / 8 at 2026-09-23 | | | … Phase 6 / monolith S05 / monolith S07 | | 1 / 0 / 11 at 2026-09-23 | | | Virtual plan rows merged but unrecorded | | 19 at 2026-09-23 | | | Open `~AR-Blocks::Closed-Beta` (gate — must be 0) | | 1 | | | … of which `~priority::1` / `~priority::2` | | 1 / 0 | | | Unassigned closed-beta blockers | | 0 | | | Open `~AR-Blocks::GA` (out of scope, tracked) | | 39 | | --- ## Closing Checklist * [x] S13 collection mode complete — steps 5 and 6 merged 2026-09-21 * [ ] S31 npm virtual — steps 13 to 18 merged, observability included (only 19, the e2e harness, is out of scope) * [ ] S30 Maven virtual — steps 10b to 19b merged, observability included (only 20, the e2e harness, is out of scope) * [x] S32's `/19` marker reconciled with its 20-row table — done 2026-09-21 * [ ] S32 container virtual — steps 8b, 10b, 11, 12, 14, 15, 16 and 17 merged, observability included (only 18, the e2e harness, is out of scope) * [ ] S17 Phase 6 **merged** 42/42 — [!2780](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2780) and [!2778](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2778) landed * [x] monolith/S05 virtual detail 8/8 — completed 2026-09-22 * [x] monolith/S07 Phase 3 CRUD — state how many of 14 landed and what carries — 8 of 14; 5, 7, 10, 11, 12, 14 carry * [x] The unowned items resolved — three moot, four deferred by decision 2026-09-23 * [x] The virtual scope question routed — the fast-follow posture stands; [!2893](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2893) records it and names what `spine.md` and `closed-beta-scope.md` still need * [x] The `spine.md` MR opened for the 2026-10-02 target, with approval and stakeholder notification — superseded: Product's [UAM!151](https://gitlab.com/gitlab-org/ci-cd/package-stage/unified-artifact-management/-/merge_requests/151) set Public Beta 2026-10-06 and GA January 2027, merged 09-23 * [ ] Virtual Status tables recorded in batch — S30's 12 rows, S32's 5, monolith/S05's 4 — and [!2664](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2664) merged * [x] [#1326](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1326) closed, or the launch gate stated with it still open — gate is 1: both MRs merged, the issue awaits verification * [ ] [#1326](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1326) attached to [&23325](https://gitlab.com/groups/gitlab-org/-/work_items/23325) * [x] [readiness!85](https://gitlab.com/gitlab-org/architecture/readiness/-/merge_requests/85) merged, or its reviewer named — merged 2026-09-21 * [ ] Severity decided: swept or the policy amended * [x] Metrics table filled in, taken from the API rather than first-parent — [W39 results note](https://gitlab.com/gitlab-org/ops/artifact-registry/-/work_items/1416#note_3913910784) * [x] Roadmap updated at `docs/roadmap/closed-beta.md` for W38 — [!2893](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2893) * [x] Roadmap updated for W39 at the Friday close — [!2997](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2997), stacked on [!2921](https://gitlab.com/gitlab-org/ops/artifact-registry/-/merge_requests/2921) * [ ] Sprint summary posted to #f_artifact-registry Slack channel **Sprint summary for Slack** (copy-paste when closing): > Sprint W39 closed. Virtual: {n} of 51 rows landed. npm S31 {n}/19 · Maven S30 {n}/25 · Container > S32 {n}/19 · Phase 6 {n}/42 merged · monolith S05 {n}/8 · monolith S07 {n}/14. Closed-beta gate: > {N} open. Virtual scope question: {routed / still open}. Sprint W40: {link}