Issues 316-319: holds that never lift, holds walked past, and one module's wait stalling every walk #202

2026-10-08T11:24:01Z

mesh-admin commented

2026-10-08 11:24:01 +00:00

Registers four issues (playbook 03), each with a report and a diagnosis, all at status located. No fix and no decision yet: each ends with open questions for graduation.

316, a group member held before its composed check answered stayed held after it passed. groupTurns (mesh-delivery) holds a published member when the composed check is not a pass, which includes "asked, not answered". Checked only stores a later verdict, held members are skipped, and the table has no observed row out of held. Case: feat/module-groups, mesh-catalog #125 held at 10:09:48 UTC, composed check passed at 10:10:21 UTC. Related: the trunks require only mesh/merge-gate and mesh/repo-check, so a group can merge before its composed check answers.
317, a held delivery's module was walked by a later merge's plan. The controller's supersededBy (ADR 0218) folds an older open plan's unbuilt modules into a newer plan, including a walk waiting for a held delivery's word. Three later catalogue merges each planned openrazer again, and it reached the laptop while #125 still said held. mesh-delivery has no row out of held for a superseded walk.
318, one module's wait for a new login held every other module's walk. A gated send judges every move on a machine as one verdict (judgeMoves), and ungatedIn refuses every other send where a move waits. The laptop's gate waited on openrazer (unit failed, account relogin needed, ADR 0252), so the tool runner's, the catalogue's and the controller's walks waited. At the ten-minute bound it failed, openrazer was put back, and the walk of the waiting builds failed, which only a person reopens. The issue asks whether relogin needed should count against a gate at all.
319, a held delivery that a later commit superseded has no end but release. 16 deliveries are held. retire-history refuses them all (their heads were heard), close has no transition from held, release would walk an old commit, and stop records the wrong thing. Five are past the one-day bound and are open stalled conditions for ever. The issue names what is needed: an end for a superseded or empty held delivery, automatic when it builds nothing, a person's act with why otherwise. Located in mesh-delivery's table and to-be 47.

Checks run locally: records.py, index.py, cycle.py and words.py (with the catalogue's copy) all pass.

Do not merge without the operator's approval.

Registers four issues (playbook 03), each with a report and a diagnosis, all at status `located`. No fix and no decision yet: each ends with open questions for graduation. - **316, a group member held before its composed check answered stayed held after it passed.** `groupTurns` (mesh-delivery) holds a published member when the composed check is not a pass, which includes "asked, not answered". `Checked` only stores a later verdict, held members are skipped, and the table has no observed row out of `held`. Case: `feat/module-groups`, mesh-catalog #125 held at 10:09:48 UTC, composed check passed at 10:10:21 UTC. Related: the trunks require only `mesh/merge-gate` and `mesh/repo-check`, so a group can merge before its composed check answers. - **317, a held delivery's module was walked by a later merge's plan.** The controller's `supersededBy` (ADR 0218) folds an older open plan's unbuilt modules into a newer plan, including a walk waiting for a held delivery's word. Three later catalogue merges each planned `openrazer` again, and it reached the laptop while #125 still said held. mesh-delivery has no row out of `held` for a superseded walk. - **318, one module's wait for a new login held every other module's walk.** A gated send judges every move on a machine as one verdict (`judgeMoves`), and `ungatedIn` refuses every other send where a move waits. The laptop's gate waited on `openrazer` (unit failed, account *relogin needed*, ADR 0252), so the tool runner's, the catalogue's and the controller's walks waited. At the ten-minute bound it failed, `openrazer` was put back, and the walk of the waiting builds failed, which only a person reopens. The issue asks whether *relogin needed* should count against a gate at all. - **319, a held delivery that a later commit superseded has no end but release.** 16 deliveries are held. `retire-history` refuses them all (their heads were heard), `close` has no transition from `held`, `release` would walk an old commit, and `stop` records the wrong thing. Five are past the one-day bound and are open stalled conditions for ever. The issue names what is needed: an end for a superseded or empty held delivery, automatic when it builds nothing, a person's act with why otherwise. Located in mesh-delivery's table and to-be 47. Checks run locally: `records.py`, `index.py`, `cycle.py` and `words.py` (with the catalogue's copy) all pass. Do not merge without the operator's approval.

mesh-admin added 1 commit 2026-10-08 11:24:01 +00:00

Issues 316-319: holds that never lift, a hold a later merge walks past, and one module's wait stalling every walk

mesh/merge-gate pass: the change touches no module of the mesh's graph

Details

mesh/repo-check pass: its merge-check.sh passed

Details

mesh/delivery delivered

Details

aa221abfee

Four delivery defects seen on 2026-10-08, each located, so their decisions can be made before code moves.

mesh-admin commented

2026-10-08 11:24:16 +00:00

Delivery novox/hq@aa221abfee2f — delivered since 2026-10-08T11:28:14Z

Delivery plan — builds nothing: the change touches no module of the mesh's graph

Transitions

2026-10-08 11:24 (new) → proposed (announced): novox/hq#202's head announced
2026-10-08 11:24 proposed → checked (checked): the verdict names this commit
2026-10-08 11:24 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail
2026-10-08 11:28 ready → published (merged): on the trunk its modules follow: merged there, and its walk opened — or nothing for a walk to move
2026-10-08 11:28 published → delivered (done): nothing for a walk to move

The commit's note under refs/notes/mesh-plan keeps every transition: git log --notes=mesh-plan.

**Delivery** `novox/hq@aa221abfee2f` — **delivered** since 2026-10-08T11:28:14Z **Delivery plan** — builds nothing: the change touches no module of the mesh's graph **Transitions** - 2026-10-08 11:24 (new) → proposed (announced): novox/hq#202's head announced - 2026-10-08 11:24 proposed → checked (checked): the verdict names this commit - 2026-10-08 11:24 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail - 2026-10-08 11:28 ready → published (merged): on the trunk its modules follow: merged there, and its walk opened — or nothing for a walk to move - 2026-10-08 11:28 published → delivered (done): nothing for a walk to move The commit's note under `refs/notes/mesh-plan` keeps every transition: `git log --notes=mesh-plan`.

mesh-admin merged commit 73fa97efde into main

2026-10-08 11:27:49 +00:00

mesh-admin deleted branch issues/316-319-holds-and-gates

2026-10-08 11:27:49 +00:00

mesh-admin referenced this pull request

2026-10-10 13:51:33 +00:00

ADR 0277, progressive insight: a proposal's values are whole on a channel that proves who answers (issue 383) #279

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: novox/hq#202