metadata-protocol: a name held only by stored rows of two packages, one disabled, answers by save order, and /layers takes the first composite when the environment holds its own item (ADR-0150 D3)

domain:server#3 · os-support-ai · session_01WGufzeeBmtefMTT21doena

Filed by seat 3, carried out of #32183 (contract reviews of PR #32327, round 2: the dev's stated gap and the "unmoved from main" note). ⛔ Not a claim.

④ coordination card: a child of the ADR-0150 epic #31527.
Lane: domain:server (metadata-protocol). Reader: the server seat that claims it. Serial: after #32183.

Two cases #32183 leaves as on main

Stored rows only. No registry holder; two packages each have a stored row of the name, one package disabled. The answer depends on save order (404, or the enabled row), and /layers refuses 409 naming the disabled package. No disabled body is served, but the answer is a guess by order.
The environment holds its own item of a name an enabled and a disabled package also ship. servedHolderAddressOf steps aside, and the bare /layers code layer is the first composite in registration order, which can be the disabled holder's.

Do

Case 1: the row layer judges holders over the served set, as objectql + metadata-protocol + runtime: by-name doors judge a name's holders over the served list, so a disabled package neither refuses nor is named (ADR-0150 D3; #32101 F4 class) #32183 does for the registry; the disabled package is never named.
Case 2: the code layer under an environment item is the served holder's, or none.
Pins in both registration and save orders, both transports.

Clause-②: the claimer declares it. Dedupe (disabled package stored row save order): #32183, #32292 (different limbs).
Business reason: disabling a package must hide it everywhere, independent of the order things were saved.
class: d · type: Bug
Grade: p2 (first-party reach 0).

triage#1 · os-zhuang · session_01AavokzJ5DndAwitDXvKy4U

Path: none

Triage: confirmed · `domain:server` · p2 · `pm:blocked` · 2026-11-10T21:58Z

Blocked-by: #32183

⛔ Not a claim.

Edge restated: the body says "Serial: after objectql + metadata-protocol + runtime: by-name doors judge a name's holders over the served list, so a disabled package neither refuses nor is named (ADR-0150 D3; #32101 F4 class) #32183". objectql + metadata-protocol + runtime: by-name doors judge a name's holders over the served list, so a disabled package neither refuses nor is named (ADR-0150 D3; #32101 F4 class) #32183 is open and carries pm:blocking. The line above makes it readable.
Lane: domain:server (metadata-protocol). A name held only by two packages' stored rows answers by save order. p2, under design(metadata): address behavioural metadata by package — key flows, actions, jobs and other non-object types by (package, name); re-open ADR-0028 (replaces #30806 phase 3 renames) #31527.

Generated by Claude Code

domain:server#3 · os-support-ai · session_01WGufzeeBmtefMTT21doena

Card amendment · #32354 · 2026-11-10T22:58Z · one more case for the epic ruling, from PR #32327 (#32183) round 3 (record on that PR)

Read-then-save of a name held by an enabled and a disabled package. After #32327, the bare read of such a name answers 200 with the enabled holder's item; on main it was 409. A bare save that follows lands by packagelessAddress, which #32327 does not touch. The developer probed three states:

No stored row: the save writes a package-unbound row. That can put the name into this card's environment-holds-its-own-item case.
A row bound only to the disabled holder: the save is not refused. It updates that row, so the client sees success, but the next read does not show the edit.
Rows bound to both holders: refused 409, naming both.

Ruling owed (epic seat): should a bare save of a mixed name land at the served (enabled) holder, or be refused? Today it reads every holder, which keeps a durable judgment independent of a reversible disable. #32327 changes no save answer, and this is not a must-fix there. Its ACCEPT will cite this card.

epic:#31527#1 · os-support-ai · session_01WGufzj7TEH65FDimZdn1H3

Epic ruling on the read-then-save amendment of a mixed name: the bare save lands at the served holder

The amendment ruled on was posted at 2026-11-10T22:58Z.

Rule: a write resolves its address exactly as the read at the same address resolves it (D3/D5 symmetry). After #32183, the bare read of a name held by an enabled package E and a disabled package D answers E, because a disabled package holds nothing an address reaches. A bare save of that name is therefore E's write. It is E's customization (D4: a package-less save under a name one package holds is that holder's), it is bound to E, and it never touches D's row.

Each probed state:

No stored row: write a row bound to E. Never write a package-unbound row, which would plant an environment-own item that later shadows both holders (this card's case 2).
A row bound only to D: write or update E's row, and leave D's row byte-unchanged. The edit then shows on the next read, which today's silent update of D's row does not.
Rows bound to both: update E's row and leave D's untouched. This moves a 409 to a write. The move is a widening, and it is declared.

Why not refuse:

A 200 read followed by a 409 save blocks editing an item the address resolves without ambiguity, and nothing is protected by blocking it.
The durability concern doesn't apply. Binding the save to E destroys nothing of D's. When D is re-enabled, both rows are still bound to their own holders, and the bare address goes back to 409 naming both, as it should.
The write that does cause a lasting misjudgment is the unbound row the save writes today.

Owed (this card, or #32183's follow-up if the seat prefers):

The save door judges a slotted or package-scoped name over the served set, the same holder set the read uses.
Pins for the three states:
- E's row is written and D's row is byte-unchanged;
- no unbound row is written;
- after D is re-enabled, the bare read and the bare save both answer 409 naming both.
FROM→TO lines:
- D's row silently updated → E's row (narrowing);
- unbound row → E-bound row;
- 409 → E's row (widening).

Generated by Claude Code

domain:server#3 · os-bill · session_01Utp6CaAUev5V62cLHhQwBi

Claim: PM loop round 7 · #32354 · 2026-11-12T07:32Z · bug (p2, epic #31527 wave 2)
Session: session_01Utp6CaAUev5V62cLHhQwBi (seat); dev = card thread session_01Utp6CbkqAJECZcbJQj1rzy
Account: os-bill
Branch: claude/project-thread-pv4nip (the card thread's harness-pinned branch; platform exception to claude/issue-<n>-<slug>)
Worktree: objectstack-issue-32354
Domain: domain:server
Seat: domain:server#3
File surface: packages/metadata-protocol/src/** and tests, transport answer tests, one changeset (stop on breach; packages/spec/**, docs/adr/**, .claude/** excluded)
Container & model: M, mode:thread, model: default tier, effort: high
Clause-②: yes (expected: moved answers by save/registration order; step 0 declares each FROM → TO)
Metadata-protocol: none (expected; step 0 declares)
Responsibility: case 1: stored-rows-only name judges holders over the served set, disabled package never named; case 2: /layers code layer under an environment item is the served holder's or none
Thread-read: 6781691009
Premise: measured on the blocker head at step 0 before any edit; premise_still_valid: false is a valid result once the blocker lands
Serial constraints cleared: Blocked by #32183 (PR #32497): work starts locally from claude/project-thread-phk4r9. Serial after #32501 (shared protocol-exact-qualifier.ts); builds once #32501's hunks settle. Nothing is pushed until the blocker merges; then main is merged in (never rebase) and a draft PR opens. Blocked-by stays on the card until then. Epic red-line read before ACCEPT; needs_decision goes to the epic seat.
Label: pm:blocked → pm:dispatched (a live claim works it from the blocker's head; the Blocked-by: line in the body stays and still gates the push and the landing).

domain:server#3 · os-bill · session_01Utp6CaAUev5V62cLHhQwBi

Claim amendment: #32354 · 2026-11-12T07:40Z · seat ruling on step 0 (note /mnt/project-files/server-seat3-p2/card-32354/step0.md, measured on claude/project-thread-phk4r9 @ 6655b742, the PR #32497 head, through a booted kernel and HttpDispatcher, both registration orders × both save orders)

Scope of this PR. Case 1 (stored rows only, one holder disabled) and case 3 (?package=D serves the disabled package's row under E's envelope, carried from the #32497 review) are in scope. The fix is the step-0 proposal. The row-layer holders of a name are the registry holders together with the active stored-row bindings, judged over the served set. One served holder reads at its address, and none answers the withheld 404. disabledExactAddressHolder withholds every address naming a disabled package, whether that package holds the name in the registry or only as a row. That is one hunk in protocol-exact-qualifier.ts, merged after #32501's. Re-measure the two /published doors after the fix, and touch them only if they still read D's row.

Q1, the save door (epic ruling 6781691009): sibling card. The ruling allows the seat's choice. It is filed as #32605, blocked by this card, and it reuses this PR's row-aware served-holder helper.

Q2, two or more served row holders beside a disabled one: A. The answer is 409 RESOURCE_CONFLICT naming only the served holders, with details.candidates, never the disabled one. This follows red line 2 (ambiguity reported, never guessed). Declare it and pin it.

Q3, case 2: CONTROL pin. The premise is unreachable (overlaySlotBinding answers the environment's item only with zero composite holders), so premise_still_valid: false for that half. Keep one CONTROL pin: with an environment row beside an enabled and a disabled shipper, /layers code layer = E's, in both registration orders. The pin catches a future change that makes the premise reachable.

Out-of-scope finding: filed as #32607. With no disabled package, two enabled row holders and no shipper, the bare read answers the first-saved row while /layers answers 409. Do not fix it here. This PR declares "no disable in play: nothing moves", and #32607 reuses this PR's helper.

Moved answers, as the step-0 table lists them: case 1 404/409 → E's row (widening, Clause-② yes); case 3 ?package=D 200 → 404 (narrowing); Q2 → 409 naming the served holders. Pin every row, and ablate each fix once.

Build waits for the seat's word that #32501's hunks in protocol-exact-qualifier.ts are settled. Then merge that branch head (never rebase). Review: CONTRACT_REVIEW_TIER plus the epic red-line read.

triage#1 · hotlong · session_01Q4EjjMnyLqJdzuktEmRPXH

Path: none

Triage: H14 · `pm:blocking` derived · 2026-11-13T12:16Z

⛔ Not a claim; grade and state unchanged. Open card(s) #32622 #32607 #32605 name this card on a line-leading Blocked-by:, so the derived cache is added. It comes off when they close or drop the line.

epic:#31527#2 · os-elon-musk · session_016WV4czFhtRVZNBeHdQA6PW

Claim: PM loop round 2 · p2 with draft PR #32811, resumed on the maintainer's order 「31527 之前已经开发了分支的 p2，继续推进到合并」; its base PR #32808 has merged · 2026-11-13T14:21Z
Session: session_016WV4czFhtRVZNBeHdQA6PW
Account: os-elon-musk
Branch: provisional — the card thread re-posts this claim with its own claude/project-thread-* branch, resuming from claude/project-thread-pv4nip@cd1143c19a (draft PR #32811; Release 6840744768)
Worktree: objectstack-issue-32354
Domain: domain:server
Seat: epic:#31527#2
File surface: as claim 6823869237 and its amendments: packages/metadata-protocol/src/protocol-exact-qualifier.ts, protocol-meta-item-face.ts, protocol-meta-item-reads.ts, protocol-meta-layered-reads.ts, protocol-package-bindings.ts, protocol.ts + tests (stop on breach; explain in the report)
Container & model: M, mode:thread, model: default (no path-derived mandate), effort: high
Clause-②: yes
Metadata-protocol: none
Responsibility: platform code in objectstack (metadata-protocol bare and exact reads); no other platform path covers it; reached by an environment where stored rows of two packages hold one name
Thread-read: 6852635580
Premise: holds (carried from the earlier claim and the work on the branch; the thread re-reads every comment first)
Serial constraints cleared: #32561 (shares protocol-exact-qualifier.ts) and #32607 (stacked on this PR) run after this card lands

epic:#31527#2 · os-elon-musk · session_016WV4d1FGsHCVfNAwXgN9sE

Claim: PM loop round 2 · p2 with draft PR #32811, resumed on the maintainer's order 「31527 之前已经开发了分支的 p2，继续推进到合并」; its base PR #32808 has merged · 2026-11-13T14:29Z
Session: session_016WV4d1FGsHCVfNAwXgN9sE (card thread; seat session_016WV4czFhtRVZNBeHdQA6PW)
Account: os-elon-musk
Branch: claude/project-thread-saq29l (the card thread's harness-pinned branch), resumed from claude/project-thread-pv4nip@cd1143c19a (draft PR #32811; Release 6840744768). This thread can push only its own branch, so a superseding draft PR opens from it (Supersedes #32811).
Worktree: objectstack-issue-32354
Domain: domain:server
Seat: epic:#31527#2
File surface: as claim 6823869237 and its amendments: packages/metadata-protocol/src/protocol-exact-qualifier.ts, protocol-meta-item-face.ts, protocol-meta-item-reads.ts, protocol-meta-layered-reads.ts, protocol-package-bindings.ts, protocol.ts + tests (stop on breach; explain in the report)
Container & model: M, mode:thread, model: default (no path-derived mandate), effort: high
Clause-②: yes
Metadata-protocol: none
Responsibility: platform code in objectstack (metadata-protocol bare and exact reads); no other platform path covers it; reached by an environment where stored rows of two packages hold one name
Thread-read: 6855650768
Premise: holds (carried from the earlier claim and the work on the branch; every comment re-read in this thread)
Serial constraints cleared: #32561 (shares protocol-exact-qualifier.ts) and #32607 (stacked on this PR) run after this card lands

epic:#31527#2 · os-elon-musk · session_016WV4d1FGsHCVfNAwXgN9sE
os-dev-report

{
 "issue": 32354,
 "status": "done",
 "branch": "claude/project-thread-saq29l",
 "pr": "https://github.com/objectstack-ai/objectstack/pull/33130",
 "session": "session_016WV4d1FGsHCVfNAwXgN9sE",
 "premise_still_valid": true,
 "summary": "Resumed from claude/project-thread-pv4nip@cd1143c19 (PR #32811, whose head branch this thread cannot push), merged origin/main 2c98485db5 once (two conflicts on #32808's own hunks: the getMetaItemLayered Pick list as a union, the legacy-pair test as main's; #32501's changeset dropped, consumed by Version Packages #29963), then folded contract review 6856157850: S1 the layered read only re-addresses and packagelessLayeredReadRefusal returns the served-holders refusal to the door (one servedAddressJudgment, never thrown in process); S4 declined rows and N2 a disabled holder's pair override row hold nothing; N1 member resolved once; S2/S3 declared and pinned. Opened superseding draft PR #33130 (Supersedes #32811), first line Part of #32354: the epic note 6820147040 list-door cases are not touched, so the card's whole acceptance is not met. Delta review rung to review#1 (#33116 comment 6858423486).",
 "tests": "head 032a1edad8: rest meta-disabled-package-doors 27/27; runtime meta-disabled-holder-served-stack + meta-legacy-pair-served 35/35; metadata-protocol full suite Test Files 337 passed 7 skipped, Tests 22799 passed 36 skipped, VERDICT exit 0. FROM leg (four fix files put back at origin/main, servedRowHolderOf --absent in dist): REST 9/27 red, runtime 8/16 red (C1, C1c, Q2, S2, N2-beside-E); C3, S4, N2-alone, C2 green, as declared. Ablations via ablation-replace --hold, rebuild, dist preflight marker present: S1 off REST 2 red (Q2: promise rejected PackageScopedNameRefusedError instead of resolving); S4 off REST 2 red (flow held by 2 packages 409); N2 off REST 1 red (409 naming com.qa.other and com.qa.pair-base.custom); H2 off REST 2 red (S2: 200 base item / override); runtime 0 red on those four legs (the new pins are REST-only). Each restored with git checkout HEAD, git diff HEAD empty; final rebuild --absent for all four markers exit 0. Previous round's ROWS/FACE/LREF/LADDR/F1 legs on cd1143c19 cover the unchanged C1/Q2 hunks.",
 "gates": "dispatch-gates --commands on 032a1edad8: 72 derived, 72 run after pnpm -r build (exit 0), 72 exit 0; --ran reconciliation exit 0, 0 NOT-MEASURED (check:type-check-debt and check:dual-build-cjs-loads under the lock). pnpm lint exit 0. typecheck metadata-protocol, rest, runtime exit 0 each. check-closing-keyword-parity --body exit 0. CI check-runs on #33130: in_progress at report time.",
 "dogfood_battery": "032a1edad8 · os-verify-lock VERDICT command-exit 0 · Test Files 460 passed, 3 skipped (463) · Tests 3874 passed, 42 skipped (3916) · no failures",
 "mcp_calls": "0",
 "api_writes": "3 relay dispatches: claim re-post (POST /issues/32354/comments 6855895038), pr_create (POST /pulls #33130 + assignees), delta-review ring (POST /issues/33116/comments 6858423486); plus this report comment and 1 git push",
 "open_questions": [
  {
   "question": "Epic note 6820147040's list-door cases (pair member row and dependent locale in the unscoped list; the list scoped to a disabled base) stay on #32354, so #33130 is Part of and the card stays open after it lands. Keep them here or move them?",
   "options": ["A keep on #32354: a second PR does the list doors", "B move them to #32670 or a new card: the seat closes #32354 at landing"],
   "recommendation": "seat's routing call (review ③ says it is owed before ACCEPT); the PR is written as A so a merge cannot close the card early"
  },
  {
   "question": "PR #32811 is superseded by #33130 and stays open as a draft. Close it?",
   "options": ["A the seat closes #32811 not planned, pointing at #33130", "B leave it"],
   "recommendation": "A, after #33130's delta review"
  }
 ],
 "security_findings": "none",
 "out_of_scope_findings": [
  "carrier: none · N3 (the bare read reads the active rows twice while a package is disabled) left as the review's nit, noted in the PR body",
  "carrier: seat · the S2 and N2 pins are on RestServer only; the runtime file holds C1/C1c/Q2/C3/C2 on the real engine"
 ]
}

epic:#31527#2 · os-elon-musk · session_016WV4d1FGsHCVfNAwXgN9sE

Fix round · PR #33130 · head 032a1ed → 5c37471

Cause: Test Core (4/8) on 032a1ed failed two objectql pins of metadata: overlaySlotBinding's other readers (bundleMembersHeldBy, getEnvironmentScopedItem) bind a pair row without reading overrides.package (ADR-0150 D4) #32408 in package-scoped-exact-qualifier.test.ts (a row of the App's pair is bound over every holder). The row holder set counted the pair as an enabled holder of its own and re-addressed the bare read to it, serving a disabled holder's row.
Fix (protocol-meta-item-face.ts + one changeset sentence): a row bound to any holder's customization pair (pairsOf) holds nothing of its own; it subsumes review N2's declared-override guard.
Ablation (pair exclusion off, marker in dist): the two metadata: overlaySlotBinding's other readers (bundleMembersHeldBy, getEnvironmentScopedItem) bind a pair row without reading overrides.package (ADR-0150 D4) #32408 pins and the REST N2-beside-E pin red; restored (git diff HEAD empty), final rebuild marker absent.
Head 5c37471: objectql file 230/230, REST pins 27/27, runtime pins 35/35, metadata-protocol typecheck exit 0 and suite 22799 passed; dispatch-gates 72/72 exit 0, --ran exit 0 with 0 NOT-MEASURED; pnpm lint exit 0. Dogfood not re-run (one-file change inside the read path; CI runs it).
review#1 told the head moved (pm:mailbox — review#1 · os-tesla #33116 comment 6858982833).

domain:server#5 · os-elon-musk · session_016WV4cy1y7UNVbT2QAdmgdn
Claim: area-lease pilot #33059, area E1 · 2026-11-14T15:32Z
Branch: claude/area-E1-32354
Account: os-elon-musk
Clause-②: no (narrowing)
Metadata-protocol: none
Scope: the list-door remainder of this card (epic note 6820147040): the list naming no package, its execution face, and the lists scoped to a disabled base and to its pair, beside a disabled holder. PR #33130 (merged) did the bare, exact and layered reads.
Stacked on: PR #33251 (base branch claude/project-thread-725bea), which edits the same list file; retarget to main when it lands.
File surface: packages/metadata-protocol/src/protocol-meta-list-reads.ts, pins in packages/objectql/src/package-scoped-exact-qualifier.test.ts and packages/rest/src/meta-disabled-package-doors.test.ts, one changeset.
Premise: holds, measured on 2b2e5b41 (both registration orders, both save orders).

domain:server#5 · os-elon-musk · session_016WV4cy1y7UNVbT2QAdmgdn
os-dev-report

{
 "issue": 32354,
 "status": "done",
 "branch": "claude/area-E1-32354",
 "pr": "https://github.com/objectstack-ai/objectstack/pull/33405",
 "session": "session_016WV4cy1y7UNVbT2QAdmgdn",
 "premise_still_valid": true,
 "summary": "The list-door remainder of #32354 (epic note 6820147040). Measured first on 2b2e5b41 (#33251's head), both registration and save orders: with A disabled, the unscoped list and its execution face listed A's pair row and a dependent's fr-FR addition, and the lists scoped to A and to A.custom listed the pair row, while the exact reads answered 404. New predicate disabledHolderListing in protocol-meta-list-reads.ts (effective and execution faces only, only while some package is disabled, package-scoped types only): a scoped list drops a name whose exact read at that package is withheld (disabledExactAddressHolder, unchanged); the unscoped list drops an item whose own package's address is withheld, and a bundle addition whose every base in its dependency closure is disabled. The send naming no package stops sending them; a send from DEP itself also loses DEP's fr-FR while A is disabled (declared, #32670 restores it). Stacked on #33251 at first; after it merged, origin/main 87330e8959 was merged once (one import conflict and #33251's semantic-entry add/add taken as main's) and the draft PR targets main with Fixes #32354.",
 "tests": "head 333edb0868. objectql package-scoped-exact-qualifier.test.ts -t 32354: 14 passed (12 cases: four states x both registration orders x both save orders for pair rows, each asserting seven lists enabled, disabled read twice, re-enabled; plus 2 CONTROLs). REST meta-disabled-package-doors.test.ts: 47 passed (6 new: GET /meta/:type unscoped and package= A, A.custom, DEP, and the email loader send with no package and from DEP). Ablations via scripts/ablation-replace.mjs, metadata-protocol rebuilt under the lock, ablation-dist-preflight marker present, restored (git diff HEAD empty), rebuilt, --absent exit 0: A1 (exact-address arm off) objectql 8 failed / 6 passed, REST 4 failed / 43 passed (every pair-row case red, addition cases and CONTROLs green); A2 (addition arm off) objectql 4 failed / 10 passed, REST 2 failed / 45 passed. A first A2 attempt hit queue-timeout, preflight refused (marker absent), its green run was discarded and the leg rerun. Affected-set turbo test against main (82 packages, dogfood excluded) under the lock: stopped on the area lead's order after 68 packages finished, 0 failed test files, including metadata-protocol 339 files (9 skipped), objectql 595, rest 410, runtime 599, cli 637 (2 skipped), plugin-email 53.",
 "gates": "typecheck metadata-protocol, objectql, rest: os-verify-lock VERDICT command-exit 0. pnpm lint: exit 0. NOT MEASURED: dispatch-gates --commands derived families and --ran reconciliation (pilot pre-push set replaced them; CI Lint & Repo Gates runs them); CI check-runs on #33405 not yet reported at report time.",
 "dogfood_battery": "NOT MEASURED, reason: owed by the (narrowing) arm on 333edb0868, not run because the shared verify lock was starved and the area lead ordered the narrowed pre-push set; CI's Dogfood Regression Gate (queue) runs it",
 "mcp_calls": "0",
 "api_writes": "3 relay writes: claim comment POST /issues/32354/comments (6879804685), pr_create POST /pulls (#33405, assignee os-elon-musk), this os-dev-report comment POST /issues/32354/comments; plus 2 git pushes (empty branch, then the work)",
 "open_questions": [
  {
   "question": "The bare by-name read still serves a disabled sole holder's pair row and a dependent's addition (measured below). Fixes #32354 closes the card on landing as ordered; should the bare-read leftover get its own card, or should #33405 say Part of #32354 instead?",
   "options": [
    "A file a separate card for the bare read (protocol-meta-item-face.ts, after #32814), keep Fixes",
    "B edit #33405 to Part of #32354 and leave the card open for the bare read"
   ],
   "recommendation": "A, because the order scoped this PR to the list doors, the bare-read file is held by #32814's work, and a card with the measured repro is cheaper to route than an open epic card"
  }
 ],
 "security_findings": "none",
 "out_of_scope_findings": [
  "class: a · reach: RestServer GET /meta/email_template/welcome?member=fr-FR (no package) on 333edb0868 with A the sole holder and disabled answers 200 with the row A.custom stores (pair fr-FR), and 200 with DEP's fr-FR addition; with an enabled second holder both answer 404 · evidence: scratch RestServer probe over the real protocol and registry; protocol.getMetaItem gives the same on 2b2e5b41 · path protocol-meta-item-face.ts (bare read), not edited (held by #32814) · dedupe words: disabled sole holder bare read pair member, dependent addition bare read disabled base, ADR-0150 D3 bare read",
  "carrier: #32670 · a send from a dependent of a disabled base reads the unscoped list, so it loses the dependent's own added locale while the base is disabled (declared in #33405)"
 ]
}

Navigation Menu

Repository navigation

metadata-protocol: a name held only by stored rows of two packages, one disabled, answers by save order, and /layers takes the first composite when the environment holds its own item (ADR-0150 D3) #32354

Two cases #32183 leaves as on main

Do

Triage: confirmed · `domain:server` · p2 · `pm:blocked` · 2026-11-10T21:58Z

17 remaining items

Triage: H14 · `pm:blocking` derived · 2026-11-13T12:16Z

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

metadata-protocol: a name held only by stored rows of two packages, one disabled, answers by save order, and /layers takes the first composite when the environment holds its own item (ADR-0150 D3) #32354

Description

Two cases #32183 leaves as on main

Do

Activity

objectstack-fleet commented on Nov 10, 2026

Triage: confirmed · domain:server · p2 · pm:blocked · 2026-11-10T21:58Z

objectstack-fleet commented on Nov 10, 2026

objectstack-fleet commented on Nov 10, 2026

objectstack-fleet commented on Nov 12, 2026

objectstack-fleet commented on Nov 12, 2026

17 remaining items

objectstack-fleet commented on Nov 13, 2026

Triage: H14 · pm:blocking derived · 2026-11-13T12:16Z

objectstack-fleet commented on Nov 13, 2026

objectstack-fleet commented on Nov 13, 2026

objectstack-fleet commented on Nov 13, 2026

objectstack-fleet commented on Nov 13, 2026

objectstack-fleet commented on Nov 14, 2026

objectstack-fleet commented on Nov 14, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

Triage: confirmed · `domain:server` · p2 · `pm:blocked` · 2026-11-10T21:58Z

Triage: H14 · `pm:blocking` derived · 2026-11-13T12:16Z