mesh-issues: a fix's time from the move to fixing to delivered, and loud over its budget (issue 403, hq ADR 0313) #250

Merged
mesh-admin merged 3 commits from fix/403-fix-time into main 2026-10-12 19:41:32 +00:00
Contributor

Why. hq ADR 0313 (accepted 2026-10-12): the operator's five minutes start when an agent moves the issue to fixing, not at the merge. The tracker measures each issue's fix time, splits it into four phases, shows it and says a fix over its budget.

mesh-delivery (no new state):

  • Each transition event says opened (when the pull request's first head was heard, never a delivery made from the forge's word at merge), merged_at and class (fixClass: the plan's class from the controller; else none for a plan that moves nothing; else the walk's class once known). checks says the same in its delivery.
  • DeliveryPlan.Class reads the controller's new field (mesh-controller PR on this branch). Without it the fallbacks hold, so this part passes and runs on today's controller.

mesh-issues:

  • fixing-from: kept at the first move to fixing. A take-over or a move away and back does not restart it. For an issue moved before this ships, the start is read from its log (the first operation that makes it fixing; for a value carried into the log, its worker's since).
  • fix-time on the issue (and so in get), kept once every fix it names is delivered and again if a later fix moves its end. It holds from, end, took, class, budget (leaf 5m, core 10m), over, the fix delivered last, the worker, the longest phase and four phases that add up to it: implementing (to opened), review (to the first approval at the head on the forge, via gitea.gitea_pull_reviews), checks (to the merge) and merge to running (to delivered). A phase ending before the one before counts zero. A phase whose end is not known is unknown, and its time counts in the next phase, which says so.
  • Class: the slowest of the fixes (any core makes core). none (no module moved) or not said gives no budget: shown, never warned on.
  • Warning issue-tracker.fix.<class>.over-budget on the operator's channel (seat:operator-channel.notify), said once the fix time is final, naming the issue, its worker and its longest phase. The newest fix time of each class decides: over and not said is said; within clears what was said for that class. Judged after each fix time and at every sweep, so a failed ask is retried.
  • New verb fix-times (days, default 7, at most 90): each class's budget, median, worst, within and over, the issues over budget with their longest phase (newest first), and the issues with no budget apart.
  • Manifest: fix-times served; invokes gitea.gitea_pull_reviews. README section A fix's time.

Design limits I chose (say if you disagree):

  • The independent reviews in this mesh are not posted as forge reviews today (mesh-catalog #246 has 0). Until they are, review reads unknown and its time counts in checks.
  • opened is when mesh-delivery first heard the pull request's head, not the forge's created_at (the events do not carry it). For a pull request merged before mesh-delivery heard a head, it is not said, and implementing counts in review.
  • A sweep keeps a missed fix time only for an issue with fixing-from. An older issue gets one when its fix is heard of again (a transition or a move), so nothing old is warned on in a burst at rollout.

Tests.

  • fixtime_test.go: the start through diagnosing, a take-over and a move away and back. Phases add up, one ended early counts zero, one not known counts in the next. Class slowest, none and not known have no budget. No warning while undelivered; warning once delivered over; cleared by the next leaf within budget and not by a core one; retried at the sweep when the channel failed. fix-times over 7 and 30 days, with an issue before the window not counted. The start read from the log. The first approval at the head.
  • issue403_test.go (mesh-delivery): a delivered transition says the first head's opened (not the second's), merged_at and the plan's class, and checks says them too; plus fixClass's fallbacks.
  • Every new test needs the new code to compile, so each fails on main.
  • go test -race passes in both modules. MESH_GATE=<controller built from this branch> MESH_CHECK_MODULES="mesh-issues mesh-delivery" sh merge-check.sh passes.

Order. One delivery group with mesh-controller and hq on fix/403-fix-time. This includes a mesh-delivery (core) rollout: merge only while mesh-controller.plans says done. No open pull request touches mesh-delivery or mesh-issues (checked 2026-10-12; issue 407's #246 is merged).

Fixes issue 403 (with the controller and hq parts). Not merged.

**Why.** hq ADR 0313 (accepted 2026-10-12): the operator's five minutes start when an agent moves the issue to `fixing`, not at the merge. The tracker measures each issue's fix time, splits it into four phases, shows it and says a fix over its budget. **mesh-delivery** (no new state): - Each `transition` event says `opened` (when the pull request's first head was heard, never a delivery made from the forge's word at merge), `merged_at` and `class` (`fixClass`: the plan's class from the controller; else `none` for a plan that moves nothing; else the walk's class once known). `checks` says the same in its `delivery`. - `DeliveryPlan.Class` reads the controller's new field (mesh-controller PR on this branch). Without it the fallbacks hold, so this part passes and runs on today's controller. **mesh-issues**: - `fixing-from`: kept at the first move to `fixing`. A take-over or a move away and back does not restart it. For an issue moved before this ships, the start is read from its log (the first operation that makes it `fixing`; for a value carried into the log, its worker's since). - `fix-time` on the issue (and so in `get`), kept once every fix it names is delivered and again if a later fix moves its end. It holds from, end, took, class, budget (leaf 5m, core 10m), over, the fix delivered last, the worker, the longest phase and four phases that add up to it: implementing (to `opened`), review (to the first approval at the head on the forge, via `gitea.gitea_pull_reviews`), checks (to the merge) and merge to running (to delivered). A phase ending before the one before counts zero. A phase whose end is not known is `unknown`, and its time counts in the next phase, which says so. - Class: the slowest of the fixes (any core makes core). `none` (no module moved) or not said gives no budget: shown, never warned on. - Warning `issue-tracker.fix.<class>.over-budget` on the operator's channel (`seat:operator-channel.notify`), said once the fix time is final, naming the issue, its worker and its longest phase. The newest fix time of each class decides: over and not said is said; within clears what was said for that class. Judged after each fix time and at every sweep, so a failed ask is retried. - New verb `fix-times` (`days`, default 7, at most 90): each class's budget, median, worst, within and over, the issues over budget with their longest phase (newest first), and the issues with no budget apart. - Manifest: `fix-times` served; invokes `gitea.gitea_pull_reviews`. README section *A fix's time*. **Design limits I chose (say if you disagree):** - The independent reviews in this mesh are not posted as forge reviews today (mesh-catalog #246 has 0). Until they are, review reads `unknown` and its time counts in checks. - `opened` is when mesh-delivery first heard the pull request's head, not the forge's `created_at` (the events do not carry it). For a pull request merged before mesh-delivery heard a head, it is not said, and implementing counts in review. - A sweep keeps a missed fix time only for an issue with `fixing-from`. An older issue gets one when its fix is heard of again (a transition or a move), so nothing old is warned on in a burst at rollout. **Tests.** - `fixtime_test.go`: the start through diagnosing, a take-over and a move away and back. Phases add up, one ended early counts zero, one not known counts in the next. Class slowest, none and not known have no budget. No warning while undelivered; warning once delivered over; cleared by the next leaf within budget and not by a core one; retried at the sweep when the channel failed. `fix-times` over 7 and 30 days, with an issue before the window not counted. The start read from the log. The first approval at the head. - `issue403_test.go` (mesh-delivery): a delivered transition says the first head's `opened` (not the second's), `merged_at` and the plan's class, and `checks` says them too; plus `fixClass`'s fallbacks. - Every new test needs the new code to compile, so each fails on main. - `go test -race` passes in both modules. `MESH_GATE=<controller built from this branch> MESH_CHECK_MODULES="mesh-issues mesh-delivery" sh merge-check.sh` passes. **Order.** One delivery group with mesh-controller and hq on `fix/403-fix-time`. This includes a mesh-delivery (core) rollout: merge only while `mesh-controller.plans` says done. No open pull request touches mesh-delivery or mesh-issues (checked 2026-10-12; issue 407's #246 is merged). Fixes issue 403 (with the controller and hq parts). Not merged.
mesh-admin added 1 commit 2026-10-12 13:50:41 +00:00
Measure each fix from the move to fixing to delivered, and say one over its budget (issue 403)
mesh/merge-gate pass: builds mesh-delivery, mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)
mesh/repo-check pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)
mesh/delivery-group group fix/403-fix-time ready: every member ready, and composed together they pass
mesh/delivery superseded: a newer head of the same pull request
43df7ebc5a
hq ADR 0313: the operator's five minutes start when an agent moves the issue
to fixing, not at the merge. mesh-delivery says on each transition, and in
checks, when the pull request opened, when it merged and the class of what it
moves. mesh-issues keeps the first move to fixing, the fix time with its four
phases once the last fix is delivered, serves fix-times, and says
issue-tracker.fix.<class>.over-budget on the operator's channel, cleared by the
next fix of that class within budget.
Author
Contributor

Delivery novox/mesh-catalog@005f802504fc — failed since 2026-10-12T19:50:00Z

Covered: covered after its failure: a later walk that contains it was delivered: plan-1791834103351744736 (novox/mesh-catalog at ac2d4dc4 into main, done, contains its merge ac2d4dc4)

Delivery plan — builds mesh-delivery, mesh-issues → novox; no bus step

  • tier 0: mesh-delivery, mesh-issues
  • novox: receives mesh-delivery, mesh-issues

Group fix/403-fix-time, in order: novox/hq@506d00cf74 → novox/mesh-controller@daf52b3aeb → novox/mesh-catalog@005f802504

Machines

  • mesh-controller on novox: passed — healthy 3 times over 3m38s
  • mesh-controller on the rest running it: sent
  • build-agent on ace: passed — carried by release-1791834641980601406
  • build-agent on g14: passed — carried by release-1791834641980601406
  • build-agent on shanks: passed — carried by release-1791834641980601406
  • build-agent on novox: passed — carried by release-1791834641980601406
  • build-agent on the rest running it: sent
  • mesh-delivery on novox: passed — healthy 3 times over 2m11s
  • mesh-issues on novox: passed — judged with mesh-delivery: healthy 3 times over 2m11s
  • mesh-delivery on the rest running it: sent
  • mesh-issues on the rest running it: sent

Transitions

  • 2026-10-12 19:29 (new) → proposed (announced): novox/mesh-catalog#250's head announced
  • 2026-10-12 19:36 proposed → checked (checked): a verdict names this commit and the trunk's head as its trunk commit
  • 2026-10-12 19:36 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail
  • 2026-10-12 19:43 ready → published (merged): on the trunk its modules follow: merged there, and its walk opened — or nothing for a walk to move
  • 2026-10-12 19:43 published → delivering (go): its walk started: let go by this owner in its group's order, by a person, or on the controller's own path — not a failed walk that carried a later commit for it, after which it waits for a walk of its own commit
  • 2026-10-12 19:50 delivering → failed (failed): its walk failed, carrying its own commit: a gate on a first machine (what it carried put back), a build, a machine

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

<!-- mesh-delivery:view --> **Delivery** `novox/mesh-catalog@005f802504fc` — **failed** since 2026-10-12T19:50:00Z **Covered**: covered after its failure: a later walk that contains it was delivered: plan-1791834103351744736 (novox/mesh-catalog at ac2d4dc4 into main, done, contains its merge ac2d4dc4) **Delivery plan** — builds mesh-delivery, mesh-issues → novox; no bus step - tier 0: mesh-delivery, mesh-issues - novox: receives mesh-delivery, mesh-issues **Group** `fix/403-fix-time`, in order: novox/hq@506d00cf7428 → novox/mesh-controller@daf52b3aeb79 → novox/mesh-catalog@005f802504fc - novox/mesh-controller@daf52b3aeb79 before novox/mesh-catalog@005f802504fc: built by - composed together: pass — every machine composes with the change as it did without (4 of 4 compose) **Machines** - mesh-controller on novox: passed — healthy 3 times over 3m38s - mesh-controller on the rest running it: sent - build-agent on ace: passed — carried by release-1791834641980601406 - build-agent on g14: passed — carried by release-1791834641980601406 - build-agent on shanks: passed — carried by release-1791834641980601406 - build-agent on novox: passed — carried by release-1791834641980601406 - build-agent on the rest running it: sent - mesh-delivery on novox: passed — healthy 3 times over 2m11s - mesh-issues on novox: passed — judged with mesh-delivery: healthy 3 times over 2m11s - mesh-delivery on the rest running it: sent - mesh-issues on the rest running it: sent **Transitions** - 2026-10-12 19:29 (new) → proposed (announced): novox/mesh-catalog#250's head announced - 2026-10-12 19:36 proposed → checked (checked): a verdict names this commit and the trunk's head as its trunk commit - 2026-10-12 19:36 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail - 2026-10-12 19:43 ready → published (merged): on the trunk its modules follow: merged there, and its walk opened — or nothing for a walk to move - 2026-10-12 19:43 published → delivering (go): its walk started: let go by this owner in its group's order, by a person, or on the controller's own path — not a failed walk that carried a later commit for it, after which it waits for a walk of its own commit - 2026-10-12 19:50 delivering → failed (failed): its walk failed, carrying its own commit: a gate on a first machine (what it carried put back), a build, a machine The commit's note under `refs/notes/mesh-plan` keeps every transition: `git log --notes=mesh-plan`.
Author
Contributor

Merge check at 43df7ebc

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues)
  • mesh/repo-check: PASS — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)

Delivery plan — builds mesh-delivery, mesh-issues → novox; no bus step

  • tier 0: mesh-delivery, mesh-issues
  • novox: receives mesh-delivery, mesh-issues
  • read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go

Run by the build seat on g14 as build-1791813064755283752 (builds --log build-1791813064755283752).

--- the gate: mesh-delivery, mesh-issues
mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB)
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here
judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T13:33:41Z says they run
composing every machine as the mesh is
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
composing every machine with the change
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose)
judged by controller development build against the facts of 2026-10-12T13:33:41Z

modules the change moves:
  - mesh-issues (changed)

machines:
  - a machine (f35): composes
  - holds mesh-dns-resolver (idt): composes
  - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes
  - a machine (yvbpsm): composes

a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues

notes:
  - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280)
--- its merge-check.sh (go toolchain)
ok  	git.novox.be/novox/mesh-catalog/checks/words	0.002s
words: 12 retired words and 70 daemons' lines, none in the 44015 strings of 696 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.010s
agent-home: no archive and no file written into under the agent's home
ok  	git.novox.be/novox/mesh-catalog/checks/paths	0.006s
paths: no module names another's state on its machine (3 known, listed to go)
long-running resources without health: 69, as kept
ok  	git.novox.be/novox/mesh-catalog/checks/deps	0.211s
testing modules/mesh-delivery
ok  	mesh-delivery/cmd/mesh-delivery	2.571s
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	11.944s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `43df7ebc` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues) - `mesh/repo-check`: **PASS** — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript) **Delivery plan** — builds mesh-delivery, mesh-issues → novox; no bus step - tier 0: mesh-delivery, mesh-issues - novox: receives mesh-delivery, mesh-issues - read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go Run by the build seat on g14 as `build-1791813064755283752` (`builds --log build-1791813064755283752`). ``` --- the gate: mesh-delivery, mesh-issues mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB) long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T13:33:41Z says they run composing every machine as the mesh is the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued composing every machine with the change the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose) judged by controller development build against the facts of 2026-10-12T13:33:41Z modules the change moves: - mesh-issues (changed) machines: - a machine (f35): composes - holds mesh-dns-resolver (idt): composes - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes - a machine (yvbpsm): composes a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues notes: - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280) --- its merge-check.sh (go toolchain) ok git.novox.be/novox/mesh-catalog/checks/words 0.002s words: 12 retired words and 70 daemons' lines, none in the 44015 strings of 696 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.010s agent-home: no archive and no file written into under the agent's home ok git.novox.be/novox/mesh-catalog/checks/paths 0.006s paths: no module names another's state on its machine (3 known, listed to go) long-running resources without health: 69, as kept ok git.novox.be/novox/mesh-catalog/checks/deps 0.211s testing modules/mesh-delivery ok mesh-delivery/cmd/mesh-delivery 2.571s testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 11.944s --- its merge-check-typescript.sh (typescript toolchain) ```
mesh-admin left a comment
Author
Contributor

Verdict: request changes at 43df7ebc5a

Independent review of issue 403 (ADR 0313) across the group fix/403-fix-time: mesh-controller #265, this pull request, and hq #331. I read the diff through the forge.

Blocking: the review phase will never be measured on this mesh. firstApproval counts only reviews whose state is APPROVED. The reviews on this forge are posted by the same account that authors the pull requests (mesh-admin). Gitea refuses an approval of your own pull request, so the independent reviews are COMMENT reviews whose first line is Verdict: approve at <head sha>. hq #329 shows this: its approving review has state COMMENT, and gitea_pull_reviews says "0 approve the head". With the code as written, every fix's review phase is unknown and its time is counted in checks. That does not meet ADR 0313 §4, which splits the time into implementing, review, checks and merge to running, and it makes §7 (weighing a lever by phase) blind to review time. A fix: also count a COMMENT review at the head whose body's first line is Verdict: approve at <full head sha>, where that sha equals the review's commit and the head, and add a test with that shape. If the forge's own approval is meant to replace the convention, say so in the README and the issue instead.

Checklist

  1. ADR 0313 clause by clause: the start is the first move to fixing (§1), and the leaf 5m and core 10m budgets are counted from it (§2). The class is the slowest fix's, and none or not known gives no budget (§2). There are four phases, a phase that ended early counts zero, an unknown phase is carried into the next, and the phases sum to the fix time (§4). get and fix-times show it (§5). The warning is said only once every fix is delivered and names the issue, worker and longest phase; it clears on the next fix of the same class delivered within budget (§6). All of this matches, except the review-phase finding above.
  2. Burst, repeat and retry: judgeFixBudgets says at most the newest over-budget fix time per class, Warned stops a repeat at each sweep, and a failed notify leaves Warned false so the next sweep retries. Tests cover all three. Not blocking: right after rollout, the first issue to get a fix time may be an old one, because afterFix and OnMoved now see the new class field as news and re-time the issue from its log start. That can raise one warning per class for an issue fixed long ago. Consider warning only when fixing-from was kept by the move rather than read back from the log, or only for an end after the module's start.
  3. fixing-from is set only when it is nil and the move enters fixing from another status, so a take-over and a move away and back keep it. The log fallback takes the first status set to fixing, which is right for move operations. Minor: for a carried or imported snapshot it uses the worker's since, which ADR 0313 §1 says is not the start when diagnosing came first. It is the best moment the log has for those issues; say so in the code comment.
  4. Class none and unknown never warn: FixBudgets has no entry for them, and a test covers it. The slowest class wins across fixes (core beats leaf).
  5. The new event fields (class, opened, merged_at) and the checks fields are additive and omitempty, so today's consumers are unaffected. The fallback is right for a plan that moves nothing (none). I did not verify that d.Time is already set when the delivered transition is emitted, which the Time.Class fallback depends on until #265 runs. If it is set after, the delivered event carries no class and the fix has no budget until it is heard again. Worth a test.
  6. gitea.gitea_pull_reviews is declared in module.json invokes and asserted in manifest_test.go. The arguments (owner, repo, number) match the tool's schema.
  7. Tests: I did not run go vet or go test -race locally. This session was refused a local clone, so I rely on mesh/repo-check (merge-check.sh in go and typescript, pass). I have not shown that the new tests fail on main.
  8. Statuses at 43df7ebc5a: mesh/merge-gate pass (mesh-delivery and mesh-issues to novox, 4 of 4 compose), mesh/repo-check pass, mesh/delivery ready, mesh/delivery-group "every member ready, and composed together they pass".

Minor, not blocking: openedOf reads the earliest head delivery still held. If earlier heads' deliveries are pruned, implementing is shortened and review lengthened.

Verdict: request changes at 43df7ebc5aa21c0bceacd1580f01f67d83d3495d Independent review of issue 403 (ADR 0313) across the group fix/403-fix-time: mesh-controller #265, this pull request, and hq #331. I read the diff through the forge. **Blocking: the review phase will never be measured on this mesh.** `firstApproval` counts only reviews whose state is `APPROVED`. The reviews on this forge are posted by the same account that authors the pull requests (mesh-admin). Gitea refuses an approval of your own pull request, so the independent reviews are `COMMENT` reviews whose first line is `Verdict: approve at <head sha>`. hq #329 shows this: its approving review has state `COMMENT`, and `gitea_pull_reviews` says "0 approve the head". With the code as written, every fix's review phase is `unknown` and its time is counted in checks. That does not meet ADR 0313 §4, which splits the time into implementing, review, checks and merge to running, and it makes §7 (weighing a lever by phase) blind to review time. A fix: also count a `COMMENT` review at the head whose body's first line is `Verdict: approve at <full head sha>`, where that sha equals the review's commit and the head, and add a test with that shape. If the forge's own approval is meant to replace the convention, say so in the README and the issue instead. **Checklist** 1. ADR 0313 clause by clause: the start is the first move to fixing (§1), and the leaf 5m and core 10m budgets are counted from it (§2). The class is the slowest fix's, and none or not known gives no budget (§2). There are four phases, a phase that ended early counts zero, an unknown phase is carried into the next, and the phases sum to the fix time (§4). `get` and `fix-times` show it (§5). The warning is said only once every fix is delivered and names the issue, worker and longest phase; it clears on the next fix of the same class delivered within budget (§6). All of this matches, except the review-phase finding above. 2. Burst, repeat and retry: `judgeFixBudgets` says at most the newest over-budget fix time per class, `Warned` stops a repeat at each sweep, and a failed notify leaves `Warned` false so the next sweep retries. Tests cover all three. Not blocking: right after rollout, the first issue to get a fix time may be an old one, because `afterFix` and `OnMoved` now see the new `class` field as news and re-time the issue from its log start. That can raise one warning per class for an issue fixed long ago. Consider warning only when `fixing-from` was kept by the move rather than read back from the log, or only for an end after the module's start. 3. `fixing-from` is set only when it is nil and the move enters fixing from another status, so a take-over and a move away and back keep it. The log fallback takes the first `status` set to fixing, which is right for move operations. Minor: for a carried or imported snapshot it uses the worker's `since`, which ADR 0313 §1 says is not the start when diagnosing came first. It is the best moment the log has for those issues; say so in the code comment. 4. Class none and unknown never warn: `FixBudgets` has no entry for them, and a test covers it. The slowest class wins across fixes (core beats leaf). 5. The new event fields (`class`, `opened`, `merged_at`) and the `checks` fields are additive and `omitempty`, so today's consumers are unaffected. The fallback is right for a plan that moves nothing (none). I did not verify that `d.Time` is already set when the delivered transition is emitted, which the Time.Class fallback depends on until #265 runs. If it is set after, the delivered event carries no class and the fix has no budget until it is heard again. Worth a test. 6. `gitea.gitea_pull_reviews` is declared in module.json `invokes` and asserted in manifest_test.go. The arguments (`owner`, `repo`, `number`) match the tool's schema. 8. Tests: I did not run go vet or go test -race locally. This session was refused a local clone, so I rely on mesh/repo-check (merge-check.sh in go and typescript, pass). I have not shown that the new tests fail on main. 9. Statuses at 43df7ebc5aa2: mesh/merge-gate pass (mesh-delivery and mesh-issues to novox, 4 of 4 compose), mesh/repo-check pass, mesh/delivery ready, mesh/delivery-group "every member ready, and composed together they pass". Minor, not blocking: `openedOf` reads the earliest head delivery still held. If earlier heads' deliveries are pruned, implementing is shortened and review lengthened.
jschoubben added 1 commit 2026-10-12 14:16:05 +00:00
Read the mesh's comment verdicts as approvals, and warn on no fix timed from the log (issue 403, review 69)
mesh/delivery-group group fix/403-fix-time checking: composed check out of date: novox/mesh-controller@7ca5a45ecc5a was judged against trunk ?, the trunk is no…
mesh/merge-gate out of date: judged against trunk 2fe5601397ae, the trunk is now 0863ec2a059d; judged again on a recheck or when it is merged
mesh/repo-check out of date: judged against trunk 2fe5601397ae, the trunk is now 0863ec2a059d; judged again on a recheck or when it is merged
mesh/delivery superseded: a newer head of the same pull request
aed7e0c968
The forge refuses an approval of one's own pull request, so the independent
reviews are comment reviews opening with "Verdict: approve at <sha>"; read
only APPROVED, the review phase was never measured. A fix whose start was
read back from the log was fixed before fix times were measured: it is kept
and shown, never warned on. A carried value at fixing after one at diagnosing
starts at its worker's active (the move), not its since (the diagnosis).
Author
Contributor

Merge check at aed7e0c9

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues)
  • mesh/repo-check: PASS — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)

Delivery plan — builds mesh-delivery, mesh-issues → novox; no bus step

  • tier 0: mesh-delivery, mesh-issues
  • novox: receives mesh-delivery, mesh-issues
  • read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go

Run by the build seat on novox as build-1791814592556277576 (builds --log build-1791814592556277576).

--- the gate: mesh-delivery, mesh-issues
mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB)
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here
judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T14:13:26Z says they run
composing every machine as the mesh is
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
composing every machine with the change
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose)
judged by controller development build against the facts of 2026-10-12T14:13:26Z

modules the change moves:
  - mesh-issues (changed)

machines:
  - a machine (f35): composes
  - holds mesh-dns-resolver (idt): composes
  - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes
  - a machine (yvbpsm): composes

a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues

notes:
  - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280)
--- its merge-check.sh (go toolchain)
ok  	git.novox.be/novox/mesh-catalog/checks/words	0.003s
words: 12 retired words and 70 daemons' lines, none in the 44027 strings of 696 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.015s
agent-home: no archive and no file written into under the agent's home
ok  	git.novox.be/novox/mesh-catalog/checks/paths	0.007s
paths: no module names another's state on its machine (3 known, listed to go)
long-running resources without health: 69, as kept
ok  	git.novox.be/novox/mesh-catalog/checks/deps	0.220s
testing modules/mesh-delivery
ok  	mesh-delivery/cmd/mesh-delivery	2.983s
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	13.083s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `aed7e0c9` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues) - `mesh/repo-check`: **PASS** — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript) **Delivery plan** — builds mesh-delivery, mesh-issues → novox; no bus step - tier 0: mesh-delivery, mesh-issues - novox: receives mesh-delivery, mesh-issues - read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go Run by the build seat on novox as `build-1791814592556277576` (`builds --log build-1791814592556277576`). ``` --- the gate: mesh-delivery, mesh-issues mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB) long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T14:13:26Z says they run composing every machine as the mesh is the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued composing every machine with the change the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose) judged by controller development build against the facts of 2026-10-12T14:13:26Z modules the change moves: - mesh-issues (changed) machines: - a machine (f35): composes - holds mesh-dns-resolver (idt): composes - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes - a machine (yvbpsm): composes a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues notes: - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280) --- its merge-check.sh (go toolchain) ok git.novox.be/novox/mesh-catalog/checks/words 0.003s words: 12 retired words and 70 daemons' lines, none in the 44027 strings of 696 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.015s agent-home: no archive and no file written into under the agent's home ok git.novox.be/novox/mesh-catalog/checks/paths 0.007s paths: no module names another's state on its machine (3 known, listed to go) long-running resources without health: 69, as kept ok git.novox.be/novox/mesh-catalog/checks/deps 0.220s testing modules/mesh-delivery ok mesh-delivery/cmd/mesh-delivery 2.983s testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 13.083s --- its merge-check-typescript.sh (typescript toolchain) ```
mesh-admin left a comment
Author
Contributor

Verdict: approve at aed7e0c968

Re-review after review 69. The blocking finding is fixed, and the two suggestions I made are done.

Review phase: fixed. firstApproval now also counts a COMMENT review at the head whose first line is Verdict: approve at <sha>. The sha must be the review's commit or a hex prefix of it of at least 7 characters. A stale or dismissed review does not count, and the earliest approval wins. The test covers these cases: a request-changes verdict, a verdict that is not on the first line, a 6-character prefix, another commit, and a review with no commit, which are all refused; and a 7-character prefix, which is accepted and wins as the earliest. The reviews on this forge carry commit, body and at_head in exactly that shape (checked against hq #329).

Fix times read from the log are never warned on: done. startOfFixing now also says whether the start came from the log, and judgeFixBudgets skips an over-budget newest fix time marked from-log. TestAFixWhoseStartCameFromTheLogIsNeverWarnedOn shows that such a fix time is kept and shown, that no warning is raised even after a sweep, and that the next fix measured from a kept start is warned as before.

A carried issue starts at the move: done. After a carried value at another status, the first carried value at fixing uses its worker's active, which is the move; it no longer uses since, which a diagnosis keeps. Only a first value already at fixing falls back to since. TestACarriedIssueStartsAtTheMoveNotTheDiagnosis covers both cases.

Tests I ran myself on a clone at aed7e0c:

  • go vet and go test -race -count=1 pass in mesh-issues (13.2s) and in mesh-delivery (2.7s).
  • The new tests do not build on origin/main, and the new fixtime_test.go does not build on the previous head 43df7eb. In both cases the functions they test are missing there.

Statuses at aed7e0c968: mesh/merge-gate pass (mesh-delivery and mesh-issues to novox, 4 of 4 compose). mesh/repo-check pass. mesh/delivery ready. mesh/delivery-group says "every member ready, and composed together they pass" (14:29, after the new heads).

Still open from review 69 and not blocking: openedOf depends on the deliveries of earlier heads still being held, and the walk's-class fallback depends on d.Time being set before the delivered transition is emitted. #265 makes the second one moot once it runs.

Verdict: approve at aed7e0c96887344950886a14d58393bfbcb247bf Re-review after review 69. The blocking finding is fixed, and the two suggestions I made are done. **Review phase: fixed.** `firstApproval` now also counts a `COMMENT` review at the head whose first line is `Verdict: approve at <sha>`. The sha must be the review's commit or a hex prefix of it of at least 7 characters. A stale or dismissed review does not count, and the earliest approval wins. The test covers these cases: a request-changes verdict, a verdict that is not on the first line, a 6-character prefix, another commit, and a review with no commit, which are all refused; and a 7-character prefix, which is accepted and wins as the earliest. The reviews on this forge carry `commit`, `body` and `at_head` in exactly that shape (checked against hq #329). **Fix times read from the log are never warned on: done.** `startOfFixing` now also says whether the start came from the log, and `judgeFixBudgets` skips an over-budget newest fix time marked `from-log`. `TestAFixWhoseStartCameFromTheLogIsNeverWarnedOn` shows that such a fix time is kept and shown, that no warning is raised even after a sweep, and that the next fix measured from a kept start is warned as before. **A carried issue starts at the move: done.** After a carried value at another status, the first carried value at fixing uses its worker's `active`, which is the move; it no longer uses `since`, which a diagnosis keeps. Only a first value already at fixing falls back to `since`. `TestACarriedIssueStartsAtTheMoveNotTheDiagnosis` covers both cases. **Tests I ran myself** on a clone at aed7e0c: - go vet and go test -race -count=1 pass in mesh-issues (13.2s) and in mesh-delivery (2.7s). - The new tests do not build on origin/main, and the new fixtime_test.go does not build on the previous head 43df7eb. In both cases the functions they test are missing there. **Statuses at aed7e0c96887:** mesh/merge-gate pass (mesh-delivery and mesh-issues to novox, 4 of 4 compose). mesh/repo-check pass. mesh/delivery ready. mesh/delivery-group says "every member ready, and composed together they pass" (14:29, after the new heads). Still open from review 69 and not blocking: `openedOf` depends on the deliveries of earlier heads still being held, and the walk's-class fallback depends on `d.Time` being set before the delivered transition is emitted. #265 makes the second one moot once it runs.
Author
Contributor

Merge check at aed7e0c9
Judged: head aed7e0c96887 merged into trunk 2fe5601397ae

  • mesh/merge-gate: FAIL — head aed7e0c968 does not merge cleanly into trunk 2fe5601397: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go (modules: mesh-delivery, mesh-issues)
  • mesh/repo-check: FAIL — head aed7e0c968 does not merge cleanly into trunk 2fe5601397: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go

Delivery plan — builds mesh-delivery, mesh-issues → novox; no bus step

  • tier 0: mesh-delivery, mesh-issues
  • novox: receives mesh-delivery, mesh-issues
  • read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go

Run by the build seat on ace as build-1791831891053807859 (builds --log build-1791831891053807859).

head aed7e0c96887 does not merge cleanly into trunk 2fe5601397ae: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go
**Merge check** at `aed7e0c9` Judged: head `aed7e0c96887` merged into trunk `2fe5601397ae` - `mesh/merge-gate`: **FAIL** — head aed7e0c96887 does not merge cleanly into trunk 2fe5601397ae: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go (modules: mesh-delivery, mesh-issues) - `mesh/repo-check`: **FAIL** — head aed7e0c96887 does not merge cleanly into trunk 2fe5601397ae: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go **Delivery plan** — builds mesh-delivery, mesh-issues → novox; no bus step - tier 0: mesh-delivery, mesh-issues - novox: receives mesh-delivery, mesh-issues - read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go Run by the build seat on ace as `build-1791831891053807859` (`builds --log build-1791831891053807859`). ``` head aed7e0c96887 does not merge cleanly into trunk 2fe5601397ae: in conflict modules/mesh-delivery/cmd/mesh-delivery/model.go ```
jschoubben added 1 commit 2026-10-12 19:29:39 +00:00
Merge remote-tracking branch 'origin/main' into fix/403-fix-time
mesh/merge-gate head 005f802504fc merged into trunk eb5f33f2b475 — pass: builds mesh-delivery, mesh-issues → novox; no bus step; every machine composes wit…
mesh/repo-check head 005f802504fc merged into trunk eb5f33f2b475 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791834103351744736 (novox/mesh-catalog at ac2d4dc4 into main,…
mesh/delivery-group group fix/403-fix-time merged: every member is merged
005f802504
# Conflicts:
#	modules/mesh-delivery/cmd/mesh-delivery/model.go
Author
Contributor

Merge check at 005f8025
Judged: head 005f802504fc merged into trunk eb5f33f2b475; read beside it: mesh-controller 4dbd4066ca54, mesh-controller-main 4dbd4066ca54, mesh-host 303d4037e84e, mesh-tools 954e1a53eef9

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues)
  • mesh/repo-check: PASS — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)

Delivery plan — builds mesh-delivery, mesh-issues → novox; no bus step

  • tier 0: mesh-delivery, mesh-issues
  • novox: receives mesh-delivery, mesh-issues
  • read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go

Run by the build seat on g14 as build-1791833394369210612 (builds --log build-1791833394369210612).

--- the gate: mesh-delivery, mesh-issues
mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB)
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here
judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T19:24:51Z says they run
composing every machine as the mesh is
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
composing every machine with the change
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued
merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose)
judged by controller development build against the facts of 2026-10-12T19:24:51Z

modules the change moves:
  - mesh-issues (changed)

machines:
  - a machine (f35): composes
  - holds mesh-dns-resolver (idt): composes
  - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes
  - a machine (yvbpsm): composes

a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues

notes:
  - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280)
--- its merge-check.sh (go toolchain)
ok  	git.novox.be/novox/mesh-catalog/checks/words	0.004s
words: 12 retired words and 70 daemons' lines, none in the 44870 strings of 704 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.010s
agent-home: no archive and no file written into under the agent's home
ok  	git.novox.be/novox/mesh-catalog/checks/paths	0.005s
paths: no module names another's state on its machine (3 known, listed to go)
long-running resources without health: 69, as kept
ok  	git.novox.be/novox/mesh-catalog/checks/deps	0.198s
testing modules/mesh-delivery
ok  	mesh-delivery/cmd/mesh-delivery	2.876s
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	11.407s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `005f8025` Judged: head `005f802504fc` merged into trunk `eb5f33f2b475`; read beside it: mesh-controller `4dbd4066ca54`, mesh-controller-main `4dbd4066ca54`, mesh-host `303d4037e84e`, mesh-tools `954e1a53eef9` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: mesh-delivery, mesh-issues) - `mesh/repo-check`: **PASS** — its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript) **Delivery plan** — builds mesh-delivery, mesh-issues → novox; no bus step - tier 0: mesh-delivery, mesh-issues - novox: receives mesh-delivery, mesh-issues - read by no module's build: modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go Run by the build seat on g14 as `build-1791833394369210612` (`builds --log build-1791833394369210612`). ``` --- the gate: mesh-delivery, mesh-issues mesh-delivery: ok, 1 tool(s), invokes seat:mesh-controller.delivery-plan, seat:mesh-controller.delivery-order, seat:mesh-controller.delivery-check, seat:mesh-controller.deliver, seat:mesh-controller.delivery-stop, seat:mesh-controller.delivery-walks, gitea.gitea_note_append, gitea.gitea_delivery_view, gitea.gitea_commit_status, gitea.gitea_commit_statuses, gitea.gitea_contains, gitea.gitea_open_pulls, keeps state deliveries, groups mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, gitea.gitea_pull_reviews, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations, keeps log changes (1024 MiB) long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 2 manifest(s) checked. Judged against the seats this binary carries; a claim on one of the mesh's own seats is judged fully at registration, and a seat declared by a module not given here reads as unknown. Identities judged on a 6-character machine name, against the bounds of the providers given here judging 1 module(s) changed against 4 machine(s), as the snapshot of 2026-10-12T19:24:51Z says they run composing every machine as the mesh is the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued composing every machine with the change the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued the bus's user list leaves out 5 user(s) the mesh has minted no credential for: controller, node.f35, node.idt, node.ygcyr, node.yvbpsm. Each is a user that cannot connect until one is issued merge gate: PASS — every machine composes with the change as it did without (4 of 4 compose) judged by controller development build against the facts of 2026-10-12T19:24:51Z modules the change moves: - mesh-issues (changed) machines: - a machine (f35): composes - holds mesh-dns-resolver (idt): composes - holds git, holds mesh-artifact-store, holds mesh-broker, holds mesh-catalog, holds mesh-delivery, holds mesh-dns-resolver, holds mesh-store, holds mesh-vault, holds npm-package-registry, the control node, the hub (ygcyr): composes - a machine (yvbpsm): composes a merge would rebuild 2 module(s) in 1 tier(s): mesh-delivery, mesh-issues notes: - modules/mesh-delivery/cmd/mesh-delivery/issue403_test.go, modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/fixtime_test.go and modules/mesh-issues/cmd/mesh-issues/manifest_test.go read by no module's build, and rebuild nothing (issue 280) --- its merge-check.sh (go toolchain) ok git.novox.be/novox/mesh-catalog/checks/words 0.004s words: 12 retired words and 70 daemons' lines, none in the 44870 strings of 704 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.010s agent-home: no archive and no file written into under the agent's home ok git.novox.be/novox/mesh-catalog/checks/paths 0.005s paths: no module names another's state on its machine (3 known, listed to go) long-running resources without health: 69, as kept ok git.novox.be/novox/mesh-catalog/checks/deps 0.198s testing modules/mesh-delivery ok mesh-delivery/cmd/mesh-delivery 2.876s testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 11.407s --- its merge-check-typescript.sh (typescript toolchain) ```
Author
Contributor

Re-check at 005f802504 (merge of main into the approved aed7e0c9): the only conflict was DeliveryPlan's last fields in mesh-delivery's model.go (main's Trunk, this PR's Class), both kept, nothing else changed. Review 77's approval carries to this head.

Re-check at 005f802504fcbdda659d3f866a40ed277768d61e (merge of main into the approved aed7e0c9): the only conflict was DeliveryPlan's last fields in mesh-delivery's model.go (main's Trunk, this PR's Class), both kept, nothing else changed. Review 77's approval carries to this head.
mesh-admin merged commit ac2d4dc46f into main 2026-10-12 19:41:32 +00:00
mesh-admin deleted branch fix/403-fix-time 2026-10-12 19:41:34 +00:00
Sign in to join this conversation.
No Reviewers
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: novox/mesh-catalog#250