mesh-issues: the issues are kept in the log changes, and the bucket is their fold (hq ADR 0300, issues 501, 502) #228

Merged
mesh-admin merged 5 commits from adr-0300-issues-on-a-log into main 2026-10-12 08:46:10 +00:00
Contributor

Implements novox/hq ADR 0300 (the issue tracker keeps its issues in a log of operations, with its bucket as the fold) on top of ADR 0297 (the module log) and ADR 0307 (mesh/state.history). This fixes issue 501 (the issues bucket fills toward its 64 MiB cap) and issue 502 (status changes are kept for only 7 days). The review answers are in comment 9800.

What changes (modules/mesh-issues)

  • Manifest: "logs": ["changes"]. The issues state still keeps history: 8 in this PR, so the bus cannot drop the older values before they are carried. History goes to 1 (ADR 0300 §3, one value per key) in a follow-up, once the carry has run on the mesh.
  • §1, every write is one operation (changes.go). This covers open, move, note, link, set, the triage, a delivery's word on a fix, a warning said or cleared, the ready claims, and the import (create, fill, fill-open). Each write appends one operation under the issue's number with AppendAt(expect = the issue's folded sequence), and only then writes the fold to the bucket at the revision read. A refused append (ErrChanged) writes nothing. The tracker folds what the log holds and applies the change again, a bounded number of times. A change that changes nothing appends nothing.
    • An operation names op, by and at. One that begins an issue (opened, imported, carried) holds the whole issue. Every other operation holds set (fields given a new value), add (lists added to at their end, so a note is kept as itself) and unset. The fold applies these as written and needs no knowledge of the kind.
    • One issue is kept in at most 252 KiB, so the operation holding it whole fits one log entry (256 KiB).
  • §2, an open keeps its order. The open is validated first, then the number is taken, then AppendAt(..., 0). If the append fails after the number was taken, the number is spent and the open answers the failure. If the answer was lost, the append is recognised as the open's own by the opening id in the issue's first log entry. When that cannot be confirmed, the answer says "may have been written, get N", never "nothing was opened".
  • §3, the bucket is the fold. Each issue carries folded, the sequence of its last operation. If the bucket write fails after the append, the change is answered as made, because the log holds it. The tracker logs that the bucket is behind, and the next change of that issue or the next start catches it up. On start the tracker reads the whole log (pages of 1000) and folds every issue whose last operation is newer than its folded. An emptied bucket is rebuilt whole. The counter is raised to the highest number in the log.
    • An entry that cannot be folded faults its issue: nothing folded past it is written, and its bucket value is left as it is.
    • A bucket value with no folded (written beside the log) is never folded over; it is left for its carry.
  • §4: events are emitted exactly as before.
  • §5, in memory (view.go):
    • The tracker holds every folded issue, kept current by watching its own bucket (stdio.State("issues").Watch) and by its own writes. allIssues, and therefore list, readiness and the sweeps, answer from memory once start has loaded it. Before that they read the bucket as today.
    • Title, report and notes have a full-text index, ranked by BM25 with title words weighted 3, and rebuilt on start. Choice: the index is written here, not taken from bleve. A few thousand issues need about 100 lines, and bleve would add dozens of modules to a bundle that has two.
    • list gains search (ranked, best first) and before (a cursor; the answer's next reads on). search together with before is refused, because a ranked list has no number order to page by.
  • §6, the switch is a carry on start, not a verb. It must come before any write reaches the log, and a verb could be run late or never. It also needs no change to the seat's verbs.
    • For every bucket issue without folded, it reads each value the bucket still keeps through State("issues").History(key) (ADR 0307). It appends them oldest first as carried operations, each holding the whole issue and the bucket's revision of that value, then writes the fold.
    • It resumes, matched by revision. A kept value is already in the log when its revision is at or below the highest one a carried operation records, or when the tracker wrote it itself (it has folded). Revision is used rather than content because the same content can be kept twice. So an interrupted carry, or two instances carrying at once, carry each value once.
    • A value written beside the log after the carry, by an instance from before the switch, is carried as one more operation.
    • Deletions among the kept values are passed over, not carried as a delete operation. The fold has no deletion, the tracker never deletes an issue, and every value after a deletion holds the whole issue again. They are counted (deletes) and named in the start line.
    • One issue that cannot be carried does not stop the start. It is named with its reason in faults and in the start line; the others are carried, and the start refolds, raises the counter and loads memory. Writes to that issue are refused with the same reason until it can be carried. Until then it is read and listed as before.
    • A write to an issue not yet carried carries that issue first, so no issue's log begins with anything else. The carry's answer says what it actually appended.

mesh-sdk is pinned at the pseudo-version of merged main 9747f3e (v0.1.14-0.20261012082113-9747f3e1b1ac), which holds both the log (ADR 0297) and KeptState.History (ADR 0307, mesh-sdk #22). It stays on that pseudo-version until a go/v0.1.14 release. That release needs the TypeScript package's version bumped first, because Go tags follow the TS package version (mesh-sdk go/README.md). That is not part of this PR.

Merge order

  1. mesh-sdk #22 is merged (9747f3e, pinned here).
  2. mesh-tools #29 (mesh/state.history) delivered to the control-node. mesh-controller #231 (logs, merged at 3bd984c) is delivered too.
  3. Then this PR.

If this were delivered before the runtime serves mesh/state.history, every issue's carry would be a fault with the runtime's reason. That is loud: the issues stay in the bucket, read and listed, and every write to them is refused, so nothing is lost. Before the runtime serves mesh/log.*, every write is refused the same way.

Reviewers' checklist

  • Every write of an issue goes through change or begin. grep -n 'Store.Create(IssuesState\|Store.Update(IssuesState' cmd/mesh-issues/*.go lists only the fold writes in changes.go.
  • A refused append writes nothing: TestAWriteWhoseAppendIsRefusedChangesNothing. An operation that came in between is not lost: TestAnOperationInBetweenIsNotLost.
  • §2: TestAnOpenRefusedAtValidationTakesNoNumber, TestAnAppendFailingAfterTheNumberSpendsIt, TestAnAppendWhoseAnswerWasLostIsTheOpensOwn, TestALostOpenIsKnownByItsFirstEntryNotItsLast, TestAnOpenWhoseWriteMayHaveLandedNeverSaysNothingWasOpened (log half), and TestANumberTakenAndNotWrittenIsSaidAndNeverHandedOutAgain (rewritten for the log).
  • §3: TestTheFoldOnStartRestoresAnEmptiedBucketExactly (byte for byte), TestABucketBehindItsLogCatchesUp, TestStartRaisesTheCounterToTheLog, and TestAnEntryThatCannotBeFoldedLeavesItsIssueAsItIs. Every harness on the fake log also asserts at cleanup that each folded issue in the bucket matches the fold of its whole log.
  • §4: TestEachWriteSaysItsEventAlongsideItsOperation checks one event and one operation per write, and that a note is an add of itself.
  • §5: TestBeforePagesThroughMoreThanFiveHundredIssues (530 issues, 6 pages, no gap or repeat, also over the tool) and TestASearchIsRankedAndTheIndexRebuiltOnStart.
  • §6: these tests all use the fake store's History, which answers like the SDK:
    • TestTheCarryIsIdempotentAndCarriesEveryValueKept
    • TestAnInterruptedCarryResumesAfterTheLastValueCarried
    • TestOneIssueThatCannotBeCarriedDoesNotStopTheStart
    • TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce
    • TestAValueWrittenBesideTheLogIsCarriedNotOverwritten
    • TestTheCarryPassesOverADeletionAmongTheKeptValues
    • TestAWriteBeforeTheCarryCarriesItsIssueFirst
    • TestAnOpenOverAnIssueKeptBeforeTheSwitchOverwritesNothing
    • Over a real bus: TestAWriteOverTheBusCarriesEveryValueTheBucketKeptFirst.
  • Over a real bus: the test runtime in bus_test.go serves mesh/log.* over a JetStream stream made as the controller makes it, and mesh/state.history. The fifty concurrent opens from two instances still get fifty distinct numbers, and TestAnIssueIsOpenedNotedLinkedMovedAndListedOverTheBus asserts one log entry per change.
  • Changed existing tests, and why:
    • tracker_test.go and import_test.go used to make the bucket fail to stand for a lost write. The record is now the log, so they make the log fail instead.
    • The size-refusal test now names 252 KiB.
  • module.json keeps history: 8 for issues; the manifest test holds it above 1 until the follow-up.

Checks

MESH_GATE=<mesh-controller at main 3bd984c> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh passes locally at head d413da2: words, agent-home, paths, every manifest through module check (69 long-running resources without health, as kept), and go test -race of the module.

Do not merge without review.

Implements novox/hq **ADR 0300** (the issue tracker keeps its issues in a log of operations, with its bucket as the fold) on top of **ADR 0297** (the module log) and **ADR 0307** (`mesh/state.history`). This fixes **issue 501** (the issues bucket fills toward its 64 MiB cap) and **issue 502** (status changes are kept for only 7 days). The review answers are in comment 9800. ## What changes (modules/mesh-issues) - **Manifest**: `"logs": ["changes"]`. The `issues` state **still keeps `history: 8`** in this PR, so the bus cannot drop the older values before they are carried. History goes to 1 (ADR 0300 §3, one value per key) **in a follow-up**, once the carry has run on the mesh. - **§1, every write is one operation** (`changes.go`). This covers open, move, note, link, set, the triage, a delivery's word on a fix, a warning said or cleared, the ready claims, and the import (create, fill, fill-open). Each write appends one operation under the issue's number with `AppendAt(expect = the issue's folded sequence)`, and only then writes the fold to the bucket at the revision read. A refused append (`ErrChanged`) writes nothing. The tracker folds what the log holds and applies the change again, a bounded number of times. A change that changes nothing appends nothing. - An operation names `op`, `by` and `at`. One that begins an issue (`opened`, `imported`, `carried`) holds the whole issue. Every other operation holds `set` (fields given a new value), `add` (lists added to at their end, so a note is kept as itself) and `unset`. The fold applies these as written and needs no knowledge of the kind. - One issue is kept in at most **252 KiB**, so the operation holding it whole fits one log entry (256 KiB). - **§2, an open keeps its order.** The open is validated first, then the number is taken, then `AppendAt(..., 0)`. If the append fails after the number was taken, the number is spent and the open answers the failure. If the answer was lost, the append is recognised as the open's own by the opening id in the issue's **first** log entry. When that cannot be confirmed, the answer says "may have been written, get N", never "nothing was opened". - **§3, the bucket is the fold.** Each issue carries `folded`, the sequence of its last operation. If the bucket write fails after the append, the change is answered as made, because the log holds it. The tracker logs that the bucket is behind, and the next change of that issue or the next start catches it up. On start the tracker reads the whole log (pages of 1000) and folds every issue whose last operation is newer than its `folded`. An emptied bucket is rebuilt whole. The counter is raised to the highest number in the log. - An entry that cannot be folded faults its issue: nothing folded past it is written, and its bucket value is left as it is. - A bucket value with no `folded` (written beside the log) is never folded over; it is left for its carry. - **§4**: events are emitted exactly as before. - **§5, in memory** (`view.go`): - The tracker holds every folded issue, kept current by watching its own bucket (`stdio.State("issues").Watch`) and by its own writes. `allIssues`, and therefore `list`, readiness and the sweeps, answer from memory once start has loaded it. Before that they read the bucket as today. - Title, report and notes have a full-text index, ranked by BM25 with title words weighted 3, and rebuilt on start. **Choice: the index is written here, not taken from bleve.** A few thousand issues need about 100 lines, and bleve would add dozens of modules to a bundle that has two. - `list` gains `search` (ranked, best first) and `before` (a cursor; the answer's `next` reads on). `search` together with `before` is refused, because a ranked list has no number order to page by. - **§6, the switch is a carry on start, not a verb.** It must come before any write reaches the log, and a verb could be run late or never. It also needs no change to the seat's verbs. - For every bucket issue without `folded`, it reads each value the bucket still keeps through `State("issues").History(key)` (ADR 0307). It appends them oldest first as `carried` operations, each holding the whole issue and the **bucket's revision** of that value, then writes the fold. - **It resumes, matched by revision.** A kept value is already in the log when its revision is at or below the highest one a carried operation records, or when the tracker wrote it itself (it has `folded`). Revision is used rather than content because the same content can be kept twice. So an interrupted carry, or two instances carrying at once, carry each value once. - A value written beside the log after the carry, by an instance from before the switch, is carried as one more operation. - **Deletions among the kept values are passed over, not carried as a delete operation.** The fold has no deletion, the tracker never deletes an issue, and every value after a deletion holds the whole issue again. They are counted (`deletes`) and named in the start line. - **One issue that cannot be carried does not stop the start.** It is named with its reason in `faults` and in the start line; the others are carried, and the start refolds, raises the counter and loads memory. Writes to that issue are refused with the same reason until it can be carried. Until then it is read and listed as before. - A write to an issue not yet carried carries that issue first, so no issue's log begins with anything else. The carry's answer says what it actually appended. mesh-sdk is pinned at the pseudo-version of merged main **9747f3e** (`v0.1.14-0.20261012082113-9747f3e1b1ac`), which holds both the log (ADR 0297) and `KeptState.History` (ADR 0307, mesh-sdk #22). It stays on that pseudo-version until a go/v0.1.14 release. That release needs the TypeScript package's version bumped first, because Go tags follow the TS package version (mesh-sdk go/README.md). That is not part of this PR. ## Merge order 1. mesh-sdk #22 is merged (9747f3e, pinned here). 2. mesh-tools **#29** (`mesh/state.history`) delivered to the control-node. mesh-controller #231 (`logs`, merged at 3bd984c) is delivered too. 3. Then this PR. If this were delivered before the runtime serves `mesh/state.history`, every issue's carry would be a fault with the runtime's reason. That is loud: the issues stay in the bucket, read and listed, and every write to them is refused, so nothing is lost. Before the runtime serves `mesh/log.*`, every write is refused the same way. ## Reviewers' checklist - [ ] Every write of an issue goes through `change` or `begin`. `grep -n 'Store.Create(IssuesState\|Store.Update(IssuesState' cmd/mesh-issues/*.go` lists only the fold writes in `changes.go`. - [ ] A refused append writes nothing: `TestAWriteWhoseAppendIsRefusedChangesNothing`. An operation that came in between is not lost: `TestAnOperationInBetweenIsNotLost`. - [ ] §2: `TestAnOpenRefusedAtValidationTakesNoNumber`, `TestAnAppendFailingAfterTheNumberSpendsIt`, `TestAnAppendWhoseAnswerWasLostIsTheOpensOwn`, `TestALostOpenIsKnownByItsFirstEntryNotItsLast`, `TestAnOpenWhoseWriteMayHaveLandedNeverSaysNothingWasOpened` (log half), and `TestANumberTakenAndNotWrittenIsSaidAndNeverHandedOutAgain` (rewritten for the log). - [ ] §3: `TestTheFoldOnStartRestoresAnEmptiedBucketExactly` (byte for byte), `TestABucketBehindItsLogCatchesUp`, `TestStartRaisesTheCounterToTheLog`, and `TestAnEntryThatCannotBeFoldedLeavesItsIssueAsItIs`. Every harness on the fake log also asserts at cleanup that each folded issue in the bucket matches the fold of its whole log. - [ ] §4: `TestEachWriteSaysItsEventAlongsideItsOperation` checks one event and one operation per write, and that a note is an `add` of itself. - [ ] §5: `TestBeforePagesThroughMoreThanFiveHundredIssues` (530 issues, 6 pages, no gap or repeat, also over the tool) and `TestASearchIsRankedAndTheIndexRebuiltOnStart`. - [ ] §6: these tests all use the fake store's `History`, which answers like the SDK: - `TestTheCarryIsIdempotentAndCarriesEveryValueKept` - `TestAnInterruptedCarryResumesAfterTheLastValueCarried` - `TestOneIssueThatCannotBeCarriedDoesNotStopTheStart` - `TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce` - `TestAValueWrittenBesideTheLogIsCarriedNotOverwritten` - `TestTheCarryPassesOverADeletionAmongTheKeptValues` - `TestAWriteBeforeTheCarryCarriesItsIssueFirst` - `TestAnOpenOverAnIssueKeptBeforeTheSwitchOverwritesNothing` - Over a real bus: `TestAWriteOverTheBusCarriesEveryValueTheBucketKeptFirst`. - [ ] Over a real bus: the test runtime in `bus_test.go` serves `mesh/log.*` over a JetStream stream made as the controller makes it, and `mesh/state.history`. The fifty concurrent opens from two instances still get fifty distinct numbers, and `TestAnIssueIsOpenedNotedLinkedMovedAndListedOverTheBus` asserts one log entry per change. - [ ] Changed existing tests, and why: - `tracker_test.go` and `import_test.go` used to make the bucket fail to stand for a lost write. The record is now the log, so they make the log fail instead. - The size-refusal test now names 252 KiB. - [ ] module.json keeps `history: 8` for `issues`; the manifest test holds it above 1 until the follow-up. ## Checks `MESH_GATE=<mesh-controller at main 3bd984c> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh` passes locally at head d413da2: words, agent-home, paths, every manifest through `module check` (69 long-running resources without health, as kept), and `go test -race` of the module. Do not merge without review.
mesh-admin added 1 commit 2026-10-12 00:42:29 +00:00
mesh-issues: keep every issue in the log changes and the bucket as its fold, so the record outlives the bucket's cap and a week of events (hq ADR 0300, issues 501, 502)
mesh/merge-gate fail: builds mesh-issues → novox; no bus step; a manifest the change touches fails the module check: modules/mesh-issues/module.json: this …
mesh/repo-check fail: its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25
mesh/delivery superseded: a newer head of the same pull request
983d434623
Every write appends one operation at the issue's last sequence before the
bucket is written; a refused append writes nothing. The start carries what
the bucket holds, refolds what is behind and indexes every issue in memory,
so list answers from memory with a ranked search and a before cursor.
Author
Contributor

Delivery novox/mesh-catalog@d413da203c3a — delivered since 2026-10-12T09:07:49Z

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

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

Machines

  • media-importer on ace: passed — healthy 1 times over 41s
  • mesh-sdk on every machine running it: sent
  • nzbgeek on ace: passed — judged with media-importer: healthy 1 times over 41s
  • nzbget on ace: passed — judged with media-importer: healthy 1 times over 41s
  • youtube on ace: passed — judged with media-importer: healthy 1 times over 41s
  • mesh-issues on novox: passed — healthy 1 times over 24s
  • media-importer on the rest running it: sent
  • mesh-issues on the rest running it: sent
  • nzbgeek on the rest running it: sent
  • nzbget on the rest running it: sent
  • youtube on the rest running it: sent

Transitions

  • 2026-10-12 08:36 (new) → proposed (announced): novox/mesh-catalog#228's head announced
  • 2026-10-12 08:38 proposed → checked (checked): the verdict names this commit
  • 2026-10-12 08:38 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail
  • 2026-10-12 08:48 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 09:04 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 09:07 delivering → delivered (done): its walk is done, carrying its own commit: every machine of its deploy plan passed, or was left as its policy says

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

<!-- mesh-delivery:view --> **Delivery** `novox/mesh-catalog@d413da203c3a` — **delivered** since 2026-10-12T09:07:49Z **Delivery plan** — builds mesh-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues **Machines** - media-importer on ace: passed — healthy 1 times over 41s - mesh-sdk on every machine running it: sent - nzbgeek on ace: passed — judged with media-importer: healthy 1 times over 41s - nzbget on ace: passed — judged with media-importer: healthy 1 times over 41s - youtube on ace: passed — judged with media-importer: healthy 1 times over 41s - mesh-issues on novox: passed — healthy 1 times over 24s - media-importer on the rest running it: sent - mesh-issues on the rest running it: sent - nzbgeek on the rest running it: sent - nzbget on the rest running it: sent - youtube on the rest running it: sent **Transitions** - 2026-10-12 08:36 (new) → proposed (announced): novox/mesh-catalog#228's head announced - 2026-10-12 08:38 proposed → checked (checked): the verdict names this commit - 2026-10-12 08:38 checked → ready (accepted): the gate passed or warned, and the repository's own check did not fail - 2026-10-12 08:48 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 09:04 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 09:07 delivering → delivered (done): its walk is done, carrying its own commit: every machine of its deploy plan passed, or was left as its policy says The commit's note under `refs/notes/mesh-plan` keeps every transition: `git log --notes=mesh-plan`.
Author
Contributor

Merge check at 983d4346

  • mesh/merge-gate: FAIL — a manifest the change touches fails the module check: modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) (modules: mesh-issues)
  • mesh/repo-check: FAIL — its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25

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

  • tier 0: mesh-issues
  • novox: receives mesh-issues
  • read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go

Run by the build seat on shanks as build-1791765751673943708 (builds --log build-1791765751673943708).

--- the gate: mesh-issues
modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs)
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
mesh-controller: 1 problem(s) in 1 manifest(s)
--- the same manifests on main
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
1 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
--- 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 42280 strings of 682 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.012s
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)
mesh-controller: 1 problem(s) in 126 manifest(s)
modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs)
adwaita: ok, 4 tool(s)
artifact-store-tools: ok, 4 tool(s), invokes seat:mesh-controller.artifacts, seat:mesh-controller.collect
asus-zephyrus-g14: ok, 11 tool(s); WARNING: asusd, supergfxd, upower stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
audit-logger: ok, keeps trail (valuable), spool (valuable)
avahi: ok, 4 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
baserow: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
betterbird: ok, 7 tool(s); WARNING: user-js ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
blueman: ok, 3 tool(s)
bluetooth: ok, 9 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
build-agent: ok, keeps workspace (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
ca-trust: ok; WARNING: trust stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
claude-code: ok, 45 tool(s), keeps state servers, holdings, config, proposals, reads claude-licence-manager.bindings, keeps agent-home (valuable); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
claude-licence-manager: ok, keeps state bindings, reads claude-code.holdings; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
clipmenu: ok, 3 tool(s)
confluence: ok
cups: ok, 7 tool(s); WARNING: socket, scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
dbus: ok, 5 tool(s); WARNING: system-bus stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
de-spiegel: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
desk-channel: ok, 1 tool(s), invokes seat:node-notifier.send, keeps state shown; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
disk-load: ok, 4 tool(s)
distribution: ok, keeps registry (rebuildable); WARNING: store stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
dmenu: ok, 2 tool(s)
dnsmasq: ok, 3 tool(s), ready: service by unit every 30s; watch by its tool dnsmasq_health every 20s
docker: ok, 22 tool(s), invokes seat:mesh-controller.images; WARNING: socket, runtime, prune stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
docker-compose: ok, 9 tool(s)
dunst: ok, 6 tool(s)
fail2ban: ok, 1 tool(s); WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
feh: ok, 2 tool(s)
flatpak: ok, 11 tool(s)
fonts: ok, 6 tool(s)
forticlient: ok, 3 tool(s), ready: split-dns by unit every 30s; WARNING: scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
gitea: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
gitlab: ok
gnome-keyring: ok, 4 tool(s)
grafana: ok, keeps data (valuable), ready: server by http / on web every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
hello-web: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
home-assistant: ok, keeps config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
hostname: ok; WARNING: name ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
i3: ok, 10 tool(s)
i3status-rust: ok, 4 tool(s)
icecast: ok, keeps logs (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
influxdb: ok, keeps data (valuable), config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
invoicing: ok; WARNING: app, api stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
jira: ok
keycloak: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
lab: ok, keeps work (cache); WARNING: runtime-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
lemurs: ok, 2 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
letta: ok; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
local-model-consumer: ok
localization: ok, 4 tool(s)
logrotate: ok, 7 tool(s); WARNING: timer stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
mailu: ok, keeps mail (valuable), dkim (valuable), data (valuable), dav (valuable), webmail (valuable), queue (valuable), filter (rebuildable), certs (rebuildable), automx (rebuildable), fetchmail (rebuildable), clamav (cache), redis (cache), ready: resolver by its image's own check every 30s; admin by its image's own check every 30s; imap by its image's own check every 30s; smtp by its image's own check every 30s; antispam by its image's own check every 30s; webmail by its image's own check every 30s; webdav by its image's own check every 30s; fetchmail by its image's own check every 30s; front by its image's own check every 30s; WARNING: config-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: redis, antivirus, automx stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
marrytts: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
matrix: ok, keeps db (valuable), ready: element by its image's own check every 30s; WARNING: element-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: homeserver stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
memory-pressure: ok, 6 tool(s); WARNING: oomd stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
mesh-catalog: ok
mesh-cli: ok
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-knowledge: ok, 2 tool(s), invokes seat:issue-tracker.reserve, claude-code.claude_code_judge, claude-code.claude_code_judged, records.records_status, records.records_list, records.records_read, seat:operator-channel.notify, ready: keep by unit every 30s; watch by its tool knowledge_health every 1m0s
mesh-vault: ok, keeps state (valuable), ledger (valuable), root (valuable)
mesh-watcher: ok, 4 tool(s), invokes mesh-watcher.watcher_ping, seat:mesh-controller.seats; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
messenger: ok, 7 tool(s), invokes seat:mesh-controller.conditions, seat:mesh-controller.root-free, seat:issue-tracker.tracking, keeps state open, sent, asks, identities; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
minio: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
model-usage: ok, keeps state (cache), spool (valuable)
mongodb: ok, keeps data (valuable), dumps (rebuildable), ready: server by tcp on database every 30s
mosquitto: ok, keeps data (rebuildable), ready: server by tcp on mqtt every 30s
mounts: ok, 3 tool(s), invokes seat:mesh-controller.data, keeps state shares, reads nfs-server.exports, ready: apply by unit every 30s; watch by its tool mounts_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
mssql: ok, keeps data (valuable), dumps (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
n8n: ok, keeps data (valuable), cache (cache); WARNING: server, cache-server, browser stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nats: ok, 9 tool(s), keeps jetstream (valuable), snapshots (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
netcheck: ok, 5 tool(s)
networkmanager: ok
nextcloud: ok, keeps html (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nextcloud-client: ok, 4 tool(s)
nfs-server: ok, 1 tool(s), invokes seat:mesh-controller.seats, seat:mesh-controller.data, keeps state exports, reads mounts.shares, ready: server by unit every 30s; exports by its tool nfs_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
nftables: ok, 1 tool(s); WARNING: load stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nm-applet: ok, 3 tool(s)
node-env: ok
nodered: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
ollama: ok, keeps models (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
only-office: ok, keeps data (rebuildable), lib (rebuildable), db (rebuildable), fonts (rebuildable), logs (cache), rabbitmq (cache), redis (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
openai-consumer: ok
openrazer: ok, 3 tool(s)
pacman: ok, 15 tool(s); WARNING: mirror-refresh, cache-cleaning stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
photos-eef: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
photos-filip: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
picom: ok, 4 tool(s); WARNING: configuration ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
polychromatic: ok, 3 tool(s)
postgres: ok, keeps store (valuable), dumps (rebuildable), ready: server by tcp on database every 30s
power: ok, 5 tool(s), keeps state draw, ready: draw by unit every 30s; WARNING: logind-power ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: after-boot, before-shutdown stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
powerlevel10k: ok
records: ok, 6 tool(s), keeps checkout (rebuildable)
redis: ok, keeps data (cache), ready: server by tcp on cache every 30s
restic: ok, 3 tool(s), keeps repository (rebuildable), state (cache), ready: night by its tool backup_health every 1m0s; measure by unit every 30s
rofi: ok, 3 tool(s)
route-adapter: ok; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
route-proxy: ok, keeps acme (rebuildable), ca (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
screen-lock: ok, 3 tool(s)
searxng: ok, keeps valkey (cache); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: cache, server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
sensors: ok, 6 tool(s), ready: drives by unit every 30s; watch by its tool sensors_health every 1m0s
showcase: ok; WARNING: server, tools stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
slack: ok, 4 tool(s)
snapd: ok, 10 tool(s)
ssh-client: ok, 8 tool(s), keeps ssh (valuable)
sshd: ok; WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
step-ca: ok, keeps home (valuable), ready: server by its image's own check every 30s
sudo: ok, 4 tool(s)
supabase: ok, keeps db (valuable), dumps (rebuildable), storage (valuable), functions (valuable), db-config (rebuildable), ready: db by its image's own check every 30s; kong by its image's own check every 30s; meta by its image's own check every 30s; studio by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: vector, analytics, auth, rest, realtime, imgproxy, storage-api, edge-functions, pooler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
systemd: ok
systemd-networkd: ok
systemd-resolved: ok, ready: service by unit every 30s; guard by unit every 30s
telegram: ok, 1 tool(s), keeps state offset, messages
time-sync: ok, 3 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
triggerhappy: ok, 4 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
umami: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
unifi: ok, keeps data (valuable); WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
website: ok, ready: server by http / on web every 30s
xclip: ok, 4 tool(s)
xdg: ok, 5 tool(s); WARNING: user-dirs ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
xorg: ok, 9 tool(s)
xterm: ok, 2 tool(s)
zfs: ok, 15 tool(s), ready: watch by its tool zfs_health every 1m0s; events by unit every 30s; WARNING: scrub-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
zsh: ok, 1 tool(s); WARNING: tools-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
zsh-autosuggestions: ok
zsh-syntax-highlighting: ok
long-running resources without health: 69
files asking for a setting without saying whether it is trusted: 25
**Merge check** at `983d4346` - `mesh/merge-gate`: **FAIL** — a manifest the change touches fails the module check: modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) (modules: mesh-issues) - `mesh/repo-check`: **FAIL** — its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25 **Delivery plan** — builds mesh-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues - read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go Run by the build seat on shanks as `build-1791765751673943708` (`builds --log build-1791765751673943708`). ``` --- the gate: mesh-issues modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 mesh-controller: 1 problem(s) in 1 manifest(s) --- the same manifests on main mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 1 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 --- 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 42280 strings of 682 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.012s 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) mesh-controller: 1 problem(s) in 126 manifest(s) modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) adwaita: ok, 4 tool(s) artifact-store-tools: ok, 4 tool(s), invokes seat:mesh-controller.artifacts, seat:mesh-controller.collect asus-zephyrus-g14: ok, 11 tool(s); WARNING: asusd, supergfxd, upower stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) audit-logger: ok, keeps trail (valuable), spool (valuable) avahi: ok, 4 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) baserow: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) betterbird: ok, 7 tool(s); WARNING: user-js ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) blueman: ok, 3 tool(s) bluetooth: ok, 9 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) build-agent: ok, keeps workspace (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) ca-trust: ok; WARNING: trust stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) claude-code: ok, 45 tool(s), keeps state servers, holdings, config, proposals, reads claude-licence-manager.bindings, keeps agent-home (valuable); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) claude-licence-manager: ok, keeps state bindings, reads claude-code.holdings; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) clipmenu: ok, 3 tool(s) confluence: ok cups: ok, 7 tool(s); WARNING: socket, scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) dbus: ok, 5 tool(s); WARNING: system-bus stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) de-spiegel: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) desk-channel: ok, 1 tool(s), invokes seat:node-notifier.send, keeps state shown; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) disk-load: ok, 4 tool(s) distribution: ok, keeps registry (rebuildable); WARNING: store stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) dmenu: ok, 2 tool(s) dnsmasq: ok, 3 tool(s), ready: service by unit every 30s; watch by its tool dnsmasq_health every 20s docker: ok, 22 tool(s), invokes seat:mesh-controller.images; WARNING: socket, runtime, prune stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) docker-compose: ok, 9 tool(s) dunst: ok, 6 tool(s) fail2ban: ok, 1 tool(s); WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) feh: ok, 2 tool(s) flatpak: ok, 11 tool(s) fonts: ok, 6 tool(s) forticlient: ok, 3 tool(s), ready: split-dns by unit every 30s; WARNING: scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) gitea: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) gitlab: ok gnome-keyring: ok, 4 tool(s) grafana: ok, keeps data (valuable), ready: server by http / on web every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) hello-web: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) home-assistant: ok, keeps config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) hostname: ok; WARNING: name ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) i3: ok, 10 tool(s) i3status-rust: ok, 4 tool(s) icecast: ok, keeps logs (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) influxdb: ok, keeps data (valuable), config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) invoicing: ok; WARNING: app, api stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) jira: ok keycloak: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) lab: ok, keeps work (cache); WARNING: runtime-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) lemurs: ok, 2 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) letta: ok; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) local-model-consumer: ok localization: ok, 4 tool(s) logrotate: ok, 7 tool(s); WARNING: timer stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) mailu: ok, keeps mail (valuable), dkim (valuable), data (valuable), dav (valuable), webmail (valuable), queue (valuable), filter (rebuildable), certs (rebuildable), automx (rebuildable), fetchmail (rebuildable), clamav (cache), redis (cache), ready: resolver by its image's own check every 30s; admin by its image's own check every 30s; imap by its image's own check every 30s; smtp by its image's own check every 30s; antispam by its image's own check every 30s; webmail by its image's own check every 30s; webdav by its image's own check every 30s; fetchmail by its image's own check every 30s; front by its image's own check every 30s; WARNING: config-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: redis, antivirus, automx stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) marrytts: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) matrix: ok, keeps db (valuable), ready: element by its image's own check every 30s; WARNING: element-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: homeserver stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) memory-pressure: ok, 6 tool(s); WARNING: oomd stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) mesh-catalog: ok mesh-cli: ok 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-knowledge: ok, 2 tool(s), invokes seat:issue-tracker.reserve, claude-code.claude_code_judge, claude-code.claude_code_judged, records.records_status, records.records_list, records.records_read, seat:operator-channel.notify, ready: keep by unit every 30s; watch by its tool knowledge_health every 1m0s mesh-vault: ok, keeps state (valuable), ledger (valuable), root (valuable) mesh-watcher: ok, 4 tool(s), invokes mesh-watcher.watcher_ping, seat:mesh-controller.seats; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) messenger: ok, 7 tool(s), invokes seat:mesh-controller.conditions, seat:mesh-controller.root-free, seat:issue-tracker.tracking, keeps state open, sent, asks, identities; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) minio: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) model-usage: ok, keeps state (cache), spool (valuable) mongodb: ok, keeps data (valuable), dumps (rebuildable), ready: server by tcp on database every 30s mosquitto: ok, keeps data (rebuildable), ready: server by tcp on mqtt every 30s mounts: ok, 3 tool(s), invokes seat:mesh-controller.data, keeps state shares, reads nfs-server.exports, ready: apply by unit every 30s; watch by its tool mounts_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) mssql: ok, keeps data (valuable), dumps (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) n8n: ok, keeps data (valuable), cache (cache); WARNING: server, cache-server, browser stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nats: ok, 9 tool(s), keeps jetstream (valuable), snapshots (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) netcheck: ok, 5 tool(s) networkmanager: ok nextcloud: ok, keeps html (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nextcloud-client: ok, 4 tool(s) nfs-server: ok, 1 tool(s), invokes seat:mesh-controller.seats, seat:mesh-controller.data, keeps state exports, reads mounts.shares, ready: server by unit every 30s; exports by its tool nfs_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) nftables: ok, 1 tool(s); WARNING: load stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nm-applet: ok, 3 tool(s) node-env: ok nodered: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) ollama: ok, keeps models (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) only-office: ok, keeps data (rebuildable), lib (rebuildable), db (rebuildable), fonts (rebuildable), logs (cache), rabbitmq (cache), redis (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) openai-consumer: ok openrazer: ok, 3 tool(s) pacman: ok, 15 tool(s); WARNING: mirror-refresh, cache-cleaning stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) photos-eef: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) photos-filip: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) picom: ok, 4 tool(s); WARNING: configuration ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) polychromatic: ok, 3 tool(s) postgres: ok, keeps store (valuable), dumps (rebuildable), ready: server by tcp on database every 30s power: ok, 5 tool(s), keeps state draw, ready: draw by unit every 30s; WARNING: logind-power ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: after-boot, before-shutdown stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) powerlevel10k: ok records: ok, 6 tool(s), keeps checkout (rebuildable) redis: ok, keeps data (cache), ready: server by tcp on cache every 30s restic: ok, 3 tool(s), keeps repository (rebuildable), state (cache), ready: night by its tool backup_health every 1m0s; measure by unit every 30s rofi: ok, 3 tool(s) route-adapter: ok; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) route-proxy: ok, keeps acme (rebuildable), ca (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) screen-lock: ok, 3 tool(s) searxng: ok, keeps valkey (cache); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: cache, server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) sensors: ok, 6 tool(s), ready: drives by unit every 30s; watch by its tool sensors_health every 1m0s showcase: ok; WARNING: server, tools stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) slack: ok, 4 tool(s) snapd: ok, 10 tool(s) ssh-client: ok, 8 tool(s), keeps ssh (valuable) sshd: ok; WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) step-ca: ok, keeps home (valuable), ready: server by its image's own check every 30s sudo: ok, 4 tool(s) supabase: ok, keeps db (valuable), dumps (rebuildable), storage (valuable), functions (valuable), db-config (rebuildable), ready: db by its image's own check every 30s; kong by its image's own check every 30s; meta by its image's own check every 30s; studio by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: vector, analytics, auth, rest, realtime, imgproxy, storage-api, edge-functions, pooler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) systemd: ok systemd-networkd: ok systemd-resolved: ok, ready: service by unit every 30s; guard by unit every 30s telegram: ok, 1 tool(s), keeps state offset, messages time-sync: ok, 3 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) triggerhappy: ok, 4 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) umami: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) unifi: ok, keeps data (valuable); WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) website: ok, ready: server by http / on web every 30s xclip: ok, 4 tool(s) xdg: ok, 5 tool(s); WARNING: user-dirs ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) xorg: ok, 9 tool(s) xterm: ok, 2 tool(s) zfs: ok, 15 tool(s), ready: watch by its tool zfs_health every 1m0s; events by unit every 30s; WARNING: scrub-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) zsh: ok, 1 tool(s); WARNING: tools-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) zsh-autosuggestions: ok zsh-syntax-highlighting: ok long-running resources without health: 69 files asking for a setting without saying whether it is trusted: 25 ```
jschoubben added 1 commit 2026-10-12 00:44:01 +00:00
mesh-issues: keep the issues bucket at 8 values until the carry can read them, so the bus drops nothing before it is carried (hq ADR 0300 §6)
mesh/merge-gate fail: builds mesh-issues → novox; no bus step; a manifest the change touches fails the module check: modules/mesh-issues/module.json: this …
mesh/repo-check fail: its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25
mesh/delivery superseded: a newer head of the same pull request
39da2cd793
Author
Contributor

Merge check at 39da2cd7

  • mesh/merge-gate: FAIL — a manifest the change touches fails the module check: modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) (modules: mesh-issues)
  • mesh/repo-check: FAIL — its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25

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

  • tier 0: mesh-issues
  • novox: receives mesh-issues
  • read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go

Run by the build seat on novox as build-1791765842817338043 (builds --log build-1791765842817338043).

--- the gate: mesh-issues
modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs)
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
mesh-controller: 1 problem(s) in 1 manifest(s)
--- the same manifests on main
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations
long-running resources without health: 0
files asking for a setting without saying whether it is trusted: 0
1 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
--- its merge-check.sh (go toolchain)
ok  	git.novox.be/novox/mesh-catalog/checks/words	0.007s
words: 12 retired words and 70 daemons' lines, none in the 42280 strings of 682 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.012s
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)
mesh-controller: 1 problem(s) in 126 manifest(s)
modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs)
adwaita: ok, 4 tool(s)
artifact-store-tools: ok, 4 tool(s), invokes seat:mesh-controller.artifacts, seat:mesh-controller.collect
asus-zephyrus-g14: ok, 11 tool(s); WARNING: asusd, supergfxd, upower stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
audit-logger: ok, keeps trail (valuable), spool (valuable)
avahi: ok, 4 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
baserow: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
betterbird: ok, 7 tool(s); WARNING: user-js ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
blueman: ok, 3 tool(s)
bluetooth: ok, 9 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
build-agent: ok, keeps workspace (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
ca-trust: ok; WARNING: trust stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
claude-code: ok, 45 tool(s), keeps state servers, holdings, config, proposals, reads claude-licence-manager.bindings, keeps agent-home (valuable); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
claude-licence-manager: ok, keeps state bindings, reads claude-code.holdings; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
clipmenu: ok, 3 tool(s)
confluence: ok
cups: ok, 7 tool(s); WARNING: socket, scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
dbus: ok, 5 tool(s); WARNING: system-bus stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
de-spiegel: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
desk-channel: ok, 1 tool(s), invokes seat:node-notifier.send, keeps state shown; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
disk-load: ok, 4 tool(s)
distribution: ok, keeps registry (rebuildable); WARNING: store stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
dmenu: ok, 2 tool(s)
dnsmasq: ok, 3 tool(s), ready: service by unit every 30s; watch by its tool dnsmasq_health every 20s
docker: ok, 22 tool(s), invokes seat:mesh-controller.images; WARNING: socket, runtime, prune stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
docker-compose: ok, 9 tool(s)
dunst: ok, 6 tool(s)
fail2ban: ok, 1 tool(s); WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
feh: ok, 2 tool(s)
flatpak: ok, 11 tool(s)
fonts: ok, 6 tool(s)
forticlient: ok, 3 tool(s), ready: split-dns by unit every 30s; WARNING: scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
gitea: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
gitlab: ok
gnome-keyring: ok, 4 tool(s)
grafana: ok, keeps data (valuable), ready: server by http / on web every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
hello-web: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
home-assistant: ok, keeps config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
hostname: ok; WARNING: name ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
i3: ok, 10 tool(s)
i3status-rust: ok, 4 tool(s)
icecast: ok, keeps logs (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
influxdb: ok, keeps data (valuable), config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
invoicing: ok; WARNING: app, api stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
jira: ok
keycloak: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
lab: ok, keeps work (cache); WARNING: runtime-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
lemurs: ok, 2 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
letta: ok; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
local-model-consumer: ok
localization: ok, 4 tool(s)
logrotate: ok, 7 tool(s); WARNING: timer stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
mailu: ok, keeps mail (valuable), dkim (valuable), data (valuable), dav (valuable), webmail (valuable), queue (valuable), filter (rebuildable), certs (rebuildable), automx (rebuildable), fetchmail (rebuildable), clamav (cache), redis (cache), ready: resolver by its image's own check every 30s; admin by its image's own check every 30s; imap by its image's own check every 30s; smtp by its image's own check every 30s; antispam by its image's own check every 30s; webmail by its image's own check every 30s; webdav by its image's own check every 30s; fetchmail by its image's own check every 30s; front by its image's own check every 30s; WARNING: config-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: redis, antivirus, automx stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
marrytts: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
matrix: ok, keeps db (valuable), ready: element by its image's own check every 30s; WARNING: element-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: homeserver stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
memory-pressure: ok, 6 tool(s); WARNING: oomd stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
mesh-catalog: ok
mesh-cli: ok
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-knowledge: ok, 2 tool(s), invokes seat:issue-tracker.reserve, claude-code.claude_code_judge, claude-code.claude_code_judged, records.records_status, records.records_list, records.records_read, seat:operator-channel.notify, ready: keep by unit every 30s; watch by its tool knowledge_health every 1m0s
mesh-vault: ok, keeps state (valuable), ledger (valuable), root (valuable)
mesh-watcher: ok, 4 tool(s), invokes mesh-watcher.watcher_ping, seat:mesh-controller.seats; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
messenger: ok, 7 tool(s), invokes seat:mesh-controller.conditions, seat:mesh-controller.root-free, seat:issue-tracker.tracking, keeps state open, sent, asks, identities; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
minio: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
model-usage: ok, keeps state (cache), spool (valuable)
mongodb: ok, keeps data (valuable), dumps (rebuildable), ready: server by tcp on database every 30s
mosquitto: ok, keeps data (rebuildable), ready: server by tcp on mqtt every 30s
mounts: ok, 3 tool(s), invokes seat:mesh-controller.data, keeps state shares, reads nfs-server.exports, ready: apply by unit every 30s; watch by its tool mounts_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
mssql: ok, keeps data (valuable), dumps (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
n8n: ok, keeps data (valuable), cache (cache); WARNING: server, cache-server, browser stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nats: ok, 9 tool(s), keeps jetstream (valuable), snapshots (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
netcheck: ok, 5 tool(s)
networkmanager: ok
nextcloud: ok, keeps html (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nextcloud-client: ok, 4 tool(s)
nfs-server: ok, 1 tool(s), invokes seat:mesh-controller.seats, seat:mesh-controller.data, keeps state exports, reads mounts.shares, ready: server by unit every 30s; exports by its tool nfs_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
nftables: ok, 1 tool(s); WARNING: load stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
nm-applet: ok, 3 tool(s)
node-env: ok
nodered: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
ollama: ok, keeps models (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
only-office: ok, keeps data (rebuildable), lib (rebuildable), db (rebuildable), fonts (rebuildable), logs (cache), rabbitmq (cache), redis (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
openai-consumer: ok
openrazer: ok, 3 tool(s)
pacman: ok, 15 tool(s); WARNING: mirror-refresh, cache-cleaning stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
photos-eef: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
photos-filip: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
picom: ok, 4 tool(s); WARNING: configuration ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
polychromatic: ok, 3 tool(s)
postgres: ok, keeps store (valuable), dumps (rebuildable), ready: server by tcp on database every 30s
power: ok, 5 tool(s), keeps state draw, ready: draw by unit every 30s; WARNING: logind-power ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: after-boot, before-shutdown stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
powerlevel10k: ok
records: ok, 6 tool(s), keeps checkout (rebuildable)
redis: ok, keeps data (cache), ready: server by tcp on cache every 30s
restic: ok, 3 tool(s), keeps repository (rebuildable), state (cache), ready: night by its tool backup_health every 1m0s; measure by unit every 30s
rofi: ok, 3 tool(s)
route-adapter: ok; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
route-proxy: ok, keeps acme (rebuildable), ca (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
screen-lock: ok, 3 tool(s)
searxng: ok, keeps valkey (cache); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: cache, server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
sensors: ok, 6 tool(s), ready: drives by unit every 30s; watch by its tool sensors_health every 1m0s
showcase: ok; WARNING: server, tools stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
slack: ok, 4 tool(s)
snapd: ok, 10 tool(s)
ssh-client: ok, 8 tool(s), keeps ssh (valuable)
sshd: ok; WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
step-ca: ok, keeps home (valuable), ready: server by its image's own check every 30s
sudo: ok, 4 tool(s)
supabase: ok, keeps db (valuable), dumps (rebuildable), storage (valuable), functions (valuable), db-config (rebuildable), ready: db by its image's own check every 30s; kong by its image's own check every 30s; meta by its image's own check every 30s; studio by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: vector, analytics, auth, rest, realtime, imgproxy, storage-api, edge-functions, pooler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
systemd: ok
systemd-networkd: ok
systemd-resolved: ok, ready: service by unit every 30s; guard by unit every 30s
telegram: ok, 1 tool(s), keeps state offset, messages
time-sync: ok, 3 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
triggerhappy: ok, 4 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
umami: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
unifi: ok, keeps data (valuable); WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8)
website: ok, ready: server by http / on web every 30s
xclip: ok, 4 tool(s)
xdg: ok, 5 tool(s); WARNING: user-dirs ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
xorg: ok, 9 tool(s)
xterm: ok, 2 tool(s)
zfs: ok, 15 tool(s), ready: watch by its tool zfs_health every 1m0s; events by unit every 30s; WARNING: scrub-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
zsh: ok, 1 tool(s); WARNING: tools-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339)
zsh-autosuggestions: ok
zsh-syntax-highlighting: ok
long-running resources without health: 69
files asking for a setting without saying whether it is trusted: 25
**Merge check** at `39da2cd7` - `mesh/merge-gate`: **FAIL** — a manifest the change touches fails the module check: modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) (modules: mesh-issues) - `mesh/repo-check`: **FAIL** — its merge-check.sh failed: files asking for a setting without saying whether it is trusted: 25 **Delivery plan** — builds mesh-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues - read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go Run by the build seat on novox as `build-1791765842817338043` (`builds --log build-1791765842817338043`). ``` --- the gate: mesh-issues modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 mesh-controller: 1 problem(s) in 1 manifest(s) --- the same manifests on main mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, records.records_status, records.records_list, records.records_read, mesh-knowledge.knowledge_highest, keeps state issues, counters, grooming, reservations long-running resources without health: 0 files asking for a setting without saying whether it is trusted: 0 1 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 --- its merge-check.sh (go toolchain) ok git.novox.be/novox/mesh-catalog/checks/words 0.007s words: 12 retired words and 70 daemons' lines, none in the 42280 strings of 682 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.012s 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) mesh-controller: 1 problem(s) in 126 manifest(s) modules/mesh-issues/module.json: this is not a module manifest: json: unknown field "logs" (read without logs) adwaita: ok, 4 tool(s) artifact-store-tools: ok, 4 tool(s), invokes seat:mesh-controller.artifacts, seat:mesh-controller.collect asus-zephyrus-g14: ok, 11 tool(s); WARNING: asusd, supergfxd, upower stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) audit-logger: ok, keeps trail (valuable), spool (valuable) avahi: ok, 4 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) baserow: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) betterbird: ok, 7 tool(s); WARNING: user-js ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) blueman: ok, 3 tool(s) bluetooth: ok, 9 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) build-agent: ok, keeps workspace (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) ca-trust: ok; WARNING: trust stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) claude-code: ok, 45 tool(s), keeps state servers, holdings, config, proposals, reads claude-licence-manager.bindings, keeps agent-home (valuable); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) claude-licence-manager: ok, keeps state bindings, reads claude-code.holdings; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) clipmenu: ok, 3 tool(s) confluence: ok cups: ok, 7 tool(s); WARNING: socket, scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) dbus: ok, 5 tool(s); WARNING: system-bus stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) de-spiegel: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) desk-channel: ok, 1 tool(s), invokes seat:node-notifier.send, keeps state shown; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) disk-load: ok, 4 tool(s) distribution: ok, keeps registry (rebuildable); WARNING: store stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) dmenu: ok, 2 tool(s) dnsmasq: ok, 3 tool(s), ready: service by unit every 30s; watch by its tool dnsmasq_health every 20s docker: ok, 22 tool(s), invokes seat:mesh-controller.images; WARNING: socket, runtime, prune stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) docker-compose: ok, 9 tool(s) dunst: ok, 6 tool(s) fail2ban: ok, 1 tool(s); WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) feh: ok, 2 tool(s) flatpak: ok, 11 tool(s) fonts: ok, 6 tool(s) forticlient: ok, 3 tool(s), ready: split-dns by unit every 30s; WARNING: scheduler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) gitea: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) gitlab: ok gnome-keyring: ok, 4 tool(s) grafana: ok, keeps data (valuable), ready: server by http / on web every 30s; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) hello-web: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) home-assistant: ok, keeps config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) hostname: ok; WARNING: name ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) i3: ok, 10 tool(s) i3status-rust: ok, 4 tool(s) icecast: ok, keeps logs (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) influxdb: ok, keeps data (valuable), config (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) invoicing: ok; WARNING: app, api stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) jira: ok keycloak: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) lab: ok, keeps work (cache); WARNING: runtime-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) lemurs: ok, 2 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) letta: ok; WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) local-model-consumer: ok localization: ok, 4 tool(s) logrotate: ok, 7 tool(s); WARNING: timer stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) mailu: ok, keeps mail (valuable), dkim (valuable), data (valuable), dav (valuable), webmail (valuable), queue (valuable), filter (rebuildable), certs (rebuildable), automx (rebuildable), fetchmail (rebuildable), clamav (cache), redis (cache), ready: resolver by its image's own check every 30s; admin by its image's own check every 30s; imap by its image's own check every 30s; smtp by its image's own check every 30s; antispam by its image's own check every 30s; webmail by its image's own check every 30s; webdav by its image's own check every 30s; fetchmail by its image's own check every 30s; front by its image's own check every 30s; WARNING: config-env ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: redis, antivirus, automx stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) marrytts: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) matrix: ok, keeps db (valuable), ready: element by its image's own check every 30s; WARNING: element-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: homeserver stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) memory-pressure: ok, 6 tool(s); WARNING: oomd stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) mesh-catalog: ok mesh-cli: ok 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-knowledge: ok, 2 tool(s), invokes seat:issue-tracker.reserve, claude-code.claude_code_judge, claude-code.claude_code_judged, records.records_status, records.records_list, records.records_read, seat:operator-channel.notify, ready: keep by unit every 30s; watch by its tool knowledge_health every 1m0s mesh-vault: ok, keeps state (valuable), ledger (valuable), root (valuable) mesh-watcher: ok, 4 tool(s), invokes mesh-watcher.watcher_ping, seat:mesh-controller.seats; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) messenger: ok, 7 tool(s), invokes seat:mesh-controller.conditions, seat:mesh-controller.root-free, seat:issue-tracker.tracking, keeps state open, sent, asks, identities; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) minio: ok, keeps data (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) model-usage: ok, keeps state (cache), spool (valuable) mongodb: ok, keeps data (valuable), dumps (rebuildable), ready: server by tcp on database every 30s mosquitto: ok, keeps data (rebuildable), ready: server by tcp on mqtt every 30s mounts: ok, 3 tool(s), invokes seat:mesh-controller.data, keeps state shares, reads nfs-server.exports, ready: apply by unit every 30s; watch by its tool mounts_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) mssql: ok, keeps data (valuable), dumps (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) n8n: ok, keeps data (valuable), cache (cache); WARNING: server, cache-server, browser stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nats: ok, 9 tool(s), keeps jetstream (valuable), snapshots (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) netcheck: ok, 5 tool(s) networkmanager: ok nextcloud: ok, keeps html (valuable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nextcloud-client: ok, 4 tool(s) nfs-server: ok, 1 tool(s), invokes seat:mesh-controller.seats, seat:mesh-controller.data, keeps state exports, reads mounts.shares, ready: server by unit every 30s; exports by its tool nfs_health every 1m0s; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) nftables: ok, 1 tool(s); WARNING: load stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) nm-applet: ok, 3 tool(s) node-env: ok nodered: ok, keeps data (valuable), ready: server by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) ollama: ok, keeps models (rebuildable); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) only-office: ok, keeps data (rebuildable), lib (rebuildable), db (rebuildable), fonts (rebuildable), logs (cache), rabbitmq (cache), redis (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) openai-consumer: ok openrazer: ok, 3 tool(s) pacman: ok, 15 tool(s); WARNING: mirror-refresh, cache-cleaning stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) photos-eef: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) photos-filip: ok; WARNING: client stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) picom: ok, 4 tool(s); WARNING: configuration ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) polychromatic: ok, 3 tool(s) postgres: ok, keeps store (valuable), dumps (rebuildable), ready: server by tcp on database every 30s power: ok, 5 tool(s), keeps state draw, ready: draw by unit every 30s; WARNING: logind-power ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: after-boot, before-shutdown stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) powerlevel10k: ok records: ok, 6 tool(s), keeps checkout (rebuildable) redis: ok, keeps data (cache), ready: server by tcp on cache every 30s restic: ok, 3 tool(s), keeps repository (rebuildable), state (cache), ready: night by its tool backup_health every 1m0s; measure by unit every 30s rofi: ok, 3 tool(s) route-adapter: ok; WARNING: config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) route-proxy: ok, keeps acme (rebuildable), ca (cache); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) screen-lock: ok, 3 tool(s) searxng: ok, keeps valkey (cache); WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: cache, server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) sensors: ok, 6 tool(s), ready: drives by unit every 30s; watch by its tool sensors_health every 1m0s showcase: ok; WARNING: server, tools stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) slack: ok, 4 tool(s) snapd: ok, 10 tool(s) ssh-client: ok, 8 tool(s), keeps ssh (valuable) sshd: ok; WARNING: run stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) step-ca: ok, keeps home (valuable), ready: server by its image's own check every 30s sudo: ok, 4 tool(s) supabase: ok, keeps db (valuable), dumps (rebuildable), storage (valuable), functions (valuable), db-config (rebuildable), ready: db by its image's own check every 30s; kong by its image's own check every 30s; meta by its image's own check every 30s; studio by its image's own check every 30s; WARNING: settings ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: vector, analytics, auth, rest, realtime, imgproxy, storage-api, edge-functions, pooler stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) systemd: ok systemd-networkd: ok systemd-resolved: ok, ready: service by unit every 30s; guard by unit every 30s telegram: ok, 1 tool(s), keeps state offset, messages time-sync: ok, 3 tool(s); WARNING: daemon stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) triggerhappy: ok, 4 tool(s); WARNING: service stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) umami: ok; WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) unifi: ok, keeps data (valuable); WARNING: runtime-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339); WARNING: server stay(s) up and say(s) not how it is ready — judged by liveness alone, refused from 2026-11-18 (ADR 0240 rule 8) website: ok, ready: server by http / on web every 30s xclip: ok, 4 tool(s) xdg: ok, 5 tool(s); WARNING: user-dirs ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) xorg: ok, 9 tool(s) xterm: ok, 2 tool(s) zfs: ok, 15 tool(s), ready: watch by its tool zfs_health every 1m0s; events by unit every 30s; WARNING: scrub-config ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) zsh: ok, 1 tool(s); WARNING: tools-conf ask(s) for a setting and do(es) not say whether it is trusted, so it counts as trusted: set at the terminal alone; say "trusted" false where nothing trusts it (novox/hq issue 339) zsh-autosuggestions: ok zsh-syntax-highlighting: ok long-running resources without health: 69 files asking for a setting without saying whether it is trusted: 25 ```
jschoubben added 1 commit 2026-10-12 08:20:52 +00:00
mesh-issues: carry each value the issues bucket kept, read through the runtime's history, so the log begins with all the bucket still knew (hq ADR 0300 §6, ADR 0307)
mesh/merge-gate pass: builds 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 superseded: a newer head of the same pull request
dcc7510cd8
jschoubben added 1 commit 2026-10-12 08:22:23 +00:00
mesh-issues: pin mesh-sdk at main 9747f3e, where the log and the history are both merged
mesh/merge-gate pass: builds 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 superseded: a newer head of the same pull request
d33f3e3f71
Author
Contributor

Merge check at dcc7510c

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step

  • tier 0: mesh-issues
  • novox: receives mesh-issues
  • read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go

Run by the build seat on g14 as build-1791793276168022861 (builds --log build-1791793276168022861).

--- the gate: mesh-issues
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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
1 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 2026-10-12T00:55:24Z 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 2026-10-12T00:55:24Z

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 1 module(s) in 1 tier(s): mesh-issues

notes:
  - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42284 strings of 682 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
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	10.907s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `dcc7510c` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues - read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go Run by the build seat on g14 as `build-1791793276168022861` (`builds --log build-1791793276168022861`). ``` --- the gate: mesh-issues mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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 1 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 2026-10-12T00:55:24Z 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 2026-10-12T00:55:24Z 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 1 module(s) in 1 tier(s): mesh-issues notes: - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42284 strings of 682 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 testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 10.907s --- its merge-check-typescript.sh (typescript toolchain) ```
Author
Contributor

Merge check at d33f3e3f

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step

  • tier 0: mesh-issues
  • novox: receives mesh-issues
  • read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go

Run by the build seat on shanks as build-1791793373315764927 (builds --log build-1791793373315764927).

--- the gate: mesh-issues
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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
1 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 2026-10-12T00:55:24Z 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 2026-10-12T00:55:24Z

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 1 module(s) in 1 tier(s): mesh-issues

notes:
  - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42284 strings of 682 files
ok  	git.novox.be/novox/mesh-catalog/checks/agenthome	0.013s
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
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	11.118s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `d33f3e3f` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues - read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go Run by the build seat on shanks as `build-1791793373315764927` (`builds --log build-1791793373315764927`). ``` --- the gate: mesh-issues mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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 1 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 2026-10-12T00:55:24Z 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 2026-10-12T00:55:24Z 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 1 module(s) in 1 tier(s): mesh-issues notes: - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42284 strings of 682 files ok git.novox.be/novox/mesh-catalog/checks/agenthome 0.013s 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 testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 11.118s --- its merge-check-typescript.sh (typescript toolchain) ```
Author
Contributor

Review (independent): CHANGES

Head reviewed: d33f3e3f71. Read against ADR 0300 (every clause and check), ADR 0297, ADR 0307 and ADR 0275 §3, with mesh-sdk main 9747f3e (go/log.go, go/state.go).

Blocking

B1. An interrupted carry loses the bucket's kept values for good, and the start line still says every kept value was carried (ADR 0300 §6, ADR 0307; checklist item 5).

The cause is in changes.go:510-521. Once the log begins an issue, carryOne no longer reads History. It only compares the log's fold with the bucket's current value, and if they differ it appends that current value alone. So when a carry stops partway, for example because the bus fails between two AppendAt calls (changes.go:534), the next carry skips every value between the last one carried and the current one. Those values are still in the bucket, which keeps history: 8 so they can be read. ADR 0307 exists so these values are kept; the operator rejected carrying the current value alone.

I reproduced it on a copy of the head:

keptBefore(300, "v1","v2","v3","v4"); the second append fails; Carry() errs; the failure is cleared; Carry()
-> log of 300: [carried:v1 carried:v4]
-> Said: "Carried 1 values of 1 issues into the log, each value the bucket kept, oldest first"

v2 and v3 are gone, and the answer reports success. That breaks the rule that failure must be loud.

The fix: when the log begins the issue with nothing but carried operations and the bucket still has folded 0, read History again. Then append the kept values that come after the last one the log holds. Match by value, or record the bucket revision inside each carried operation and resume from it. Add a test with a failure partway through the carry.

B2. One issue that cannot be carried stops the whole start (changes.go:637-641, 690-694).

Carry returns at the first carryOne error, and Start then returns before the refold, the counter raise and view().load. main.start retries every minute and hits the same failure each time. I reproduced it: a log that refuses every append leaves Start failing and view().ready() false. A permanent per-issue refusal would cause this. For example, an issue near MostIssueBytes (256 KiB) that overflows the log's 256 KiB value limit once the carried wrapper is added. In that case §3 (refold, counter) and §5 (memory, index) never run for any issue, and nothing raises a condition; there are only log lines.

The fix: record the issue's failure in Started.Faults, carry the other issues, and continue the start. That issue's writes are already refused one by one through change. Also check whether any live issue comes within a few hundred bytes of 256 KiB.

Checklist

  1. Pass. Every write to the bucket is in changes.go: 389, 391 (store), 443, 445 (catchUp), 548 (carryOne), 730, 740 (Start). No other file writes IssuesState. All the writes go through change or begin: open (tracker.go:189-212), move (416), note (575), link (645), set and triage set (ready.go:83), ready (ready.go:355), delivery (delivery.go:170, 245), warnings (claims.go:94, 108; delivery.go:293, 517), groom apply (groom.go:517), and import create, fill and fill-open (import.go:603-623, 638, 679). reserve.go writes only counters and reservations.
  2. Pass. AppendAt(key, op, is.Folded) is at changes.go:347. When it answers ErrChanged, the change folds the log (catchUp) and tries again with no bucket write, and the retries stop after issueAttempts. Any other error is answered as an error with "get N before changing it again", so the change is never retried and cannot be applied twice.
  3. Pass. The open validates first (tracker.go:141-173), then calls next(), then begin with expect 0. An append failure is answered with "N is spent". A lost answer is recognised as the open's own only when Last holds this open's random Opening, and the counter's compare-and-set means two callers never hold one number. Non-blocking: mine checks the key's last entry. If another operation landed on the issue right after a lost first append, the open is answered errBegun ("counter behind") although its issue exists. Reading the key's first entry (Read(1, key, 1)) would avoid that.
  4. Pass with a note. The fold is deterministic. Start refolds an issue the bucket holds behind its log and rebuilds a missing one, and the counter is raised to the log's highest number. TestTheFoldOnStartRestoresAnEmptiedBucketExactly compares whole JSON values. I added a property check to a copy of the head: after every test, fold each issue's whole log from nothing and compare it with the bucket byte for byte. That made 809 comparisons across about 110 tests, covering delivery, triage, ready, import, worker and claims, and found 0 differences. Non-blocking: in Start (changes.go:710-715), when an entry fails to fold, the later entries of that issue are still folded on top of the earlier state, and the result can be written to the bucket. After a fault, skip that issue.
  5. Fail (B1). The carry runs before any write to the log: on start, and through change while an issue's folded is 0. History is read oldest first. Passing over deletes is sound, because the tracker never deletes an issue and each value after a delete is the whole issue. Two carriers racing are safe through expect, so nothing is carried twice. When the runtime does not serve mesh/state.history, kept returns the error and the issue's writes are refused, which is loud and loses nothing. The !ok branch (current value only) cannot run in production, because runtimeStore always has History. For an issue changed in the bucket between History and the first append: a writer on the new code carries the issue itself and loses the expect race, which is safe. A writer from before the switch can leave […, newer, older] in the log for a while, and the next carry fixes it. But Start's refold (changes.go:739) sees folded 0 below the log's fold and overwrites that writer's value with the log's fold. Non-blocking, because it needs an old instance running alongside: skip folded == 0 there.
  6. Pass. module.json keeps history: 8 and has logs: ["changes"]. manifest_test.go refuses a history below 2 and pins logs. The one value per key that ADR 0300 §3 and the §6 live check require waits for the follow-up, as the PR says.
  7. Pass. The callers emit as before. TestEachWriteSaysItsEventAlongsideItsOperation covers open, note, link, set and move.
  8. Pass. The bucket watch and remember keep memory current, so a list right after a write on the same instance sees the write. before and next page correctly; the 530-issue test is real. A search together with before is refused. The BM25 scoring is correct (the Lucene idf, k1 1.2, b 0.75, title words weighted 3), and the index is rebuilt in load. Non-blocking: memoryView.all() copies issues shallowly. No caller changes the shared slices today, but the comment there promises more than the code does.
  9. Pass. go test -race is clean. The fifty concurrent opens over the real bus are a real test (the counter losses are asserted), and two instances on one log are covered by TestAnOperationInBetweenIsNotLost. Missing: a test with two instances carrying the same issue at once.
  10. Pass. The tests in tracker_test.go and import_test.go now make the log fail instead of the bucket, and their assertions are as strict as before or stricter ("N is spent", "did not answer"). None was turned into a test that passes without checking anything.
  11. Pass with a note. go.mod pins exactly v0.1.14-0.20261012082113-9747f3e1b1ac, which is the tip of mesh-sdk main. No other dependency changed, and go mod verify succeeds. go.sum keeps three stale SDK lines (4cf45b5, 5794ccd, d41d255); run go mod tidy.
  12. Pass. go test -race ./... passes at the head. I built mesh-controller at main 3bd984c, and MESH_GATE=<it> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh exits 0.
  13. The required checks at d33f3e3, read once: mesh/merge-gate success, "pass: builds mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)". mesh/repo-check success, "pass: its merge-check.sh passed, each part in its toolchain". mesh/delivery success, "ready: it delivers once merged".
  14. Pass. The glossary words are used, and the diff never says "ticket".

Before delivery (not a code finding)

The merge order in the description must be proven live before merging. If the runtime on novox does not serve mesh/log.*, every open spends a number and fails, and every write is refused. If it does not serve mesh/state.history, every write to an issue not yet carried is refused.

Review (independent): CHANGES Head reviewed: d33f3e3f7109ba75bdabf1831a37f977ab128d2e. Read against ADR 0300 (every clause and check), ADR 0297, ADR 0307 and ADR 0275 §3, with mesh-sdk main 9747f3e (go/log.go, go/state.go). ## Blocking **B1. An interrupted carry loses the bucket's kept values for good, and the start line still says every kept value was carried (ADR 0300 §6, ADR 0307; checklist item 5).** The cause is in `changes.go:510-521`. Once the log begins an issue, `carryOne` no longer reads `History`. It only compares the log's fold with the bucket's current value, and if they differ it appends that current value alone. So when a carry stops partway, for example because the bus fails between two `AppendAt` calls (`changes.go:534`), the next carry skips every value between the last one carried and the current one. Those values are still in the bucket, which keeps `history: 8` so they can be read. ADR 0307 exists so these values are kept; the operator rejected carrying the current value alone. I reproduced it on a copy of the head: ``` keptBefore(300, "v1","v2","v3","v4"); the second append fails; Carry() errs; the failure is cleared; Carry() -> log of 300: [carried:v1 carried:v4] -> Said: "Carried 1 values of 1 issues into the log, each value the bucket kept, oldest first" ``` v2 and v3 are gone, and the answer reports success. That breaks the rule that failure must be loud. The fix: when the log begins the issue with nothing but `carried` operations and the bucket still has `folded` 0, read `History` again. Then append the kept values that come after the last one the log holds. Match by value, or record the bucket revision inside each `carried` operation and resume from it. Add a test with a failure partway through the carry. **B2. One issue that cannot be carried stops the whole start (`changes.go:637-641`, `690-694`).** `Carry` returns at the first `carryOne` error, and `Start` then returns before the refold, the counter raise and `view().load`. `main.start` retries every minute and hits the same failure each time. I reproduced it: a log that refuses every append leaves `Start` failing and `view().ready()` false. A permanent per-issue refusal would cause this. For example, an issue near `MostIssueBytes` (256 KiB) that overflows the log's 256 KiB value limit once the `carried` wrapper is added. In that case §3 (refold, counter) and §5 (memory, index) never run for any issue, and nothing raises a condition; there are only log lines. The fix: record the issue's failure in `Started.Faults`, carry the other issues, and continue the start. That issue's writes are already refused one by one through `change`. Also check whether any live issue comes within a few hundred bytes of 256 KiB. ## Checklist 1. **Pass.** Every write to the bucket is in `changes.go`: 389, 391 (`store`), 443, 445 (`catchUp`), 548 (`carryOne`), 730, 740 (`Start`). No other file writes `IssuesState`. All the writes go through `change` or `begin`: open (`tracker.go:189-212`), move (416), note (575), link (645), set and triage set (`ready.go:83`), ready (`ready.go:355`), delivery (`delivery.go:170`, 245), warnings (`claims.go:94`, 108; `delivery.go:293`, 517), groom apply (`groom.go:517`), and import create, fill and fill-open (`import.go:603-623`, 638, 679). `reserve.go` writes only `counters` and `reservations`. 2. **Pass.** `AppendAt(key, op, is.Folded)` is at `changes.go:347`. When it answers `ErrChanged`, the change folds the log (`catchUp`) and tries again with no bucket write, and the retries stop after `issueAttempts`. Any other error is answered as an error with "get N before changing it again", so the change is never retried and cannot be applied twice. 3. **Pass.** The open validates first (`tracker.go:141-173`), then calls `next()`, then `begin` with expect 0. An append failure is answered with "N is spent". A lost answer is recognised as the open's own only when `Last` holds this open's random `Opening`, and the counter's compare-and-set means two callers never hold one number. Non-blocking: `mine` checks the key's *last* entry. If another operation landed on the issue right after a lost first append, the open is answered `errBegun` ("counter behind") although its issue exists. Reading the key's first entry (`Read(1, key, 1)`) would avoid that. 4. **Pass with a note.** The fold is deterministic. `Start` refolds an issue the bucket holds behind its log and rebuilds a missing one, and the counter is raised to the log's highest number. `TestTheFoldOnStartRestoresAnEmptiedBucketExactly` compares whole JSON values. I added a property check to a copy of the head: after every test, fold each issue's whole log from nothing and compare it with the bucket byte for byte. That made 809 comparisons across about 110 tests, covering delivery, triage, ready, import, worker and claims, and found 0 differences. Non-blocking: in `Start` (`changes.go:710-715`), when an entry fails to fold, the later entries of that issue are still folded on top of the earlier state, and the result can be written to the bucket. After a fault, skip that issue. 5. **Fail (B1).** The carry runs before any write to the log: on start, and through `change` while an issue's `folded` is 0. History is read oldest first. Passing over deletes is sound, because the tracker never deletes an issue and each value after a delete is the whole issue. Two carriers racing are safe through `expect`, so nothing is carried twice. When the runtime does not serve `mesh/state.history`, `kept` returns the error and the issue's writes are refused, which is loud and loses nothing. The `!ok` branch (current value only) cannot run in production, because `runtimeStore` always has `History`. For an issue changed in the bucket between `History` and the first append: a writer on the new code carries the issue itself and loses the `expect` race, which is safe. A writer from before the switch can leave `[…, newer, older]` in the log for a while, and the next carry fixes it. But `Start`'s refold (`changes.go:739`) sees `folded` 0 below the log's fold and overwrites that writer's value with the log's fold. Non-blocking, because it needs an old instance running alongside: skip `folded == 0` there. 6. **Pass.** `module.json` keeps `history: 8` and has `logs: ["changes"]`. `manifest_test.go` refuses a history below 2 and pins `logs`. The one value per key that ADR 0300 §3 and the §6 live check require waits for the follow-up, as the PR says. 7. **Pass.** The callers emit as before. `TestEachWriteSaysItsEventAlongsideItsOperation` covers open, note, link, set and move. 8. **Pass.** The bucket watch and `remember` keep memory current, so a list right after a write on the same instance sees the write. `before` and `next` page correctly; the 530-issue test is real. A search together with `before` is refused. The BM25 scoring is correct (the Lucene idf, k1 1.2, b 0.75, title words weighted 3), and the index is rebuilt in `load`. Non-blocking: `memoryView.all()` copies issues shallowly. No caller changes the shared slices today, but the comment there promises more than the code does. 9. **Pass.** `go test -race` is clean. The fifty concurrent opens over the real bus are a real test (the counter losses are asserted), and two instances on one log are covered by `TestAnOperationInBetweenIsNotLost`. Missing: a test with two instances carrying the same issue at once. 10. **Pass.** The tests in `tracker_test.go` and `import_test.go` now make the log fail instead of the bucket, and their assertions are as strict as before or stricter ("N is spent", "did not answer"). None was turned into a test that passes without checking anything. 11. **Pass with a note.** `go.mod` pins exactly `v0.1.14-0.20261012082113-9747f3e1b1ac`, which is the tip of mesh-sdk main. No other dependency changed, and `go mod verify` succeeds. `go.sum` keeps three stale SDK lines (4cf45b5, 5794ccd, d41d255); run `go mod tidy`. 12. **Pass.** `go test -race ./...` passes at the head. I built mesh-controller at main 3bd984c, and `MESH_GATE=<it> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh` exits 0. 13. The required checks at d33f3e3, read once: `mesh/merge-gate` success, "pass: builds mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)". `mesh/repo-check` success, "pass: its merge-check.sh passed, each part in its toolchain". `mesh/delivery` success, "ready: it delivers once merged". 14. **Pass.** The glossary words are used, and the diff never says "ticket". ## Before delivery (not a code finding) The merge order in the description must be proven live before merging. If the runtime on novox does not serve `mesh/log.*`, every open spends a number and fails, and every write is refused. If it does not serve `mesh/state.history`, every write to an issue not yet carried is refused.
jschoubben added 1 commit 2026-10-12 08:35:59 +00:00
mesh-issues: resume an interrupted carry by revision and fault one issue instead of the whole start, so no kept value is lost and the tracker always starts (review of PR 228)
mesh/delivery delivered
mesh/merge-gate pass: builds 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)
d413da203c
An open's lost answer is judged by the issue's first entry; an entry that does not
fold faults its issue on start; a value written beside the log is never folded over;
one issue is kept in at most 252 KiB so its whole operation fits a log entry.
Author
Contributor

Answer to the review (comment 9798), at head d413da2.

BLOCKING 1: an interrupted carry. carryOne now resumes instead of restarting. Each carried operation records revision, the bucket's revision of the value it holds.

  • A kept value counts as already in the log when its revision is at or below the highest revision any carried operation of that issue records, or when it is a value the tracker wrote itself (it has folded).
  • Values are matched by revision, not content. The same content can be kept twice (a write that changed nothing), and a revision names exactly one value.
  • The carry's answer counts only what it actually appended. When it fails partway it says how many values it carried, and the issue is listed in faults.
  • Test: TestAnInterruptedCarryResumesAfterTheLastValueCarried. The bus fails after the first value; a write in between is refused and changes nothing; the second carry appends the remaining three. The log ends carried:v1 carried:v2 carried:v3 carried:v4.

BLOCKING 2: one issue stopping the start. Carry records a failed issue in Carried.faults and goes on with the others. Start copies those into its own faults as not carried: …, so they appear in the start line, which is what the module logs on start. The start then still refolds, raises the counter and loads memory.

  • Writes to a refused issue are refused with the same reason (issue N is kept from before the tracker's log and is not carried into it yet, so nothing is changed: …) until it can be carried. Until then it is read and listed as before.
  • MostIssueBytes is now 252 KiB, so an issue's whole operation fits one log entry (256 KiB). The largest issue on record is 25 KiB.
  • Test: TestOneIssueThatCannotBeCarriedDoesNotStopTheStart. One issue at the bucket's old limit is refused by a log that enforces 256 KiB; the other two are carried; the view is loaded and list returns all 3; a note on the refused issue is refused and changes nothing.

Non-blocking:

  • (a) Lost answer on an open. The check now reads the issue's first entry (firstOf, Read(1, key, 1)) and no longer uses Last; Last is gone from the Log interface. Test: TestALostOpenIsKnownByItsFirstEntryNotItsLast. Another operation lands between the lost answer and the retry, and the open is still known as its own.
  • (b) An entry that does not fold. At start, once an entry of an issue fails to fold, that issue is faulted. No later entry is folded onto it, and its bucket value is left as it is. Test: TestAnEntryThatCannotBeFoldedLeavesItsIssueAsItIs.
  • (c) A value written beside the log. If the bucket holds a value with folded 0 (written by an instance from before the switch, after the carry), the refold step at start skips it and records it as a fault, so nothing overwrites it. The carry, at this start or the next one, or before its next change, appends that value as one more operation; the old instance's write was the newest bucket write, made at the revision it read.
    • Test: TestAValueWrittenBesideTheLogIsCarriedNotOverwritten. While its carry fails, the value is kept untouched (two faults). At the next start it is carried, and the log reads opened noted carried:written beside the log.
  • (d) memoryView.all(). The comment now says it is a shallow copy and the lists are shared and read-only. Every caller only reads, and writes go through change, which reads the bucket again. I kept the shallow copy because a deep copy would mean a JSON round trip of every issue on every list.
  • (e) Fallback removed. The "current value only" fallback is gone. History is now part of the Store interface: the runtime store reads mesh/state.history, and the terminal's dry-run store answers empty.
  • (f) Ran go mod tidy; it removed 6 stale lines from go.sum.
  • (g) Two carriers at once. Test: TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce. The second instance carries in full between the first one's append and its judgement; the first is refused, reads again and appends nothing, so each value appears once.

I also adopted your fold check: every test harness on the fake log now asserts, at cleanup, that each folded issue in the bucket matches the fold of its whole log byte for byte.

go test -race passes, and so does MESH_GATE=<mesh-controller main 3bd984c> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh at d413da2.

Answer to the review (comment 9798), at head d413da2. **BLOCKING 1: an interrupted carry.** `carryOne` now resumes instead of restarting. Each `carried` operation records `revision`, the bucket's revision of the value it holds. - A kept value counts as already in the log when its revision is at or below the highest revision any carried operation of that issue records, or when it is a value the tracker wrote itself (it has `folded`). - Values are matched by revision, not content. The same content can be kept twice (a write that changed nothing), and a revision names exactly one value. - The carry's answer counts only what it actually appended. When it fails partway it says how many values it carried, and the issue is listed in `faults`. - Test: `TestAnInterruptedCarryResumesAfterTheLastValueCarried`. The bus fails after the first value; a write in between is refused and changes nothing; the second carry appends the remaining three. The log ends `carried:v1 carried:v2 carried:v3 carried:v4`. **BLOCKING 2: one issue stopping the start.** `Carry` records a failed issue in `Carried.faults` and goes on with the others. `Start` copies those into its own faults as `not carried: …`, so they appear in the start line, which is what the module logs on start. The start then still refolds, raises the counter and loads memory. - Writes to a refused issue are refused with the same reason (`issue N is kept from before the tracker's log and is not carried into it yet, so nothing is changed: …`) until it can be carried. Until then it is read and listed as before. - `MostIssueBytes` is now 252 KiB, so an issue's whole operation fits one log entry (256 KiB). The largest issue on record is 25 KiB. - Test: `TestOneIssueThatCannotBeCarriedDoesNotStopTheStart`. One issue at the bucket's old limit is refused by a log that enforces 256 KiB; the other two are carried; the view is loaded and `list` returns all 3; a note on the refused issue is refused and changes nothing. **Non-blocking:** - **(a) Lost answer on an open.** The check now reads the issue's first entry (`firstOf`, `Read(1, key, 1)`) and no longer uses `Last`; `Last` is gone from the Log interface. Test: `TestALostOpenIsKnownByItsFirstEntryNotItsLast`. Another operation lands between the lost answer and the retry, and the open is still known as its own. - **(b) An entry that does not fold.** At start, once an entry of an issue fails to fold, that issue is faulted. No later entry is folded onto it, and its bucket value is left as it is. Test: `TestAnEntryThatCannotBeFoldedLeavesItsIssueAsItIs`. - **(c) A value written beside the log.** If the bucket holds a value with `folded` 0 (written by an instance from before the switch, after the carry), the refold step at start skips it and records it as a fault, so nothing overwrites it. The carry, at this start or the next one, or before its next change, appends that value as one more operation; the old instance's write was the newest bucket write, made at the revision it read. - Test: `TestAValueWrittenBesideTheLogIsCarriedNotOverwritten`. While its carry fails, the value is kept untouched (two faults). At the next start it is carried, and the log reads `opened noted carried:written beside the log`. - **(d) `memoryView.all()`.** The comment now says it is a shallow copy and the lists are shared and read-only. Every caller only reads, and writes go through `change`, which reads the bucket again. I kept the shallow copy because a deep copy would mean a JSON round trip of every issue on every list. - **(e) Fallback removed.** The "current value only" fallback is gone. `History` is now part of the `Store` interface: the runtime store reads `mesh/state.history`, and the terminal's dry-run store answers empty. - **(f)** Ran `go mod tidy`; it removed 6 stale lines from go.sum. - **(g) Two carriers at once.** Test: `TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce`. The second instance carries in full between the first one's append and its judgement; the first is refused, reads again and appends nothing, so each value appears once. I also adopted your fold check: every test harness on the fake log now asserts, at cleanup, that each folded issue in the bucket matches the fold of its whole log byte for byte. `go test -race` passes, and so does `MESH_GATE=<mesh-controller main 3bd984c> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh` at d413da2.
Author
Contributor

Merge check at d413da20

  • mesh/merge-gate: PASS — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step

  • tier 0: mesh-issues
  • novox: receives mesh-issues
  • read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go

Run by the build seat on g14 as build-1791794163029908393 (builds --log build-1791794163029908393).

--- the gate: mesh-issues
mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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
1 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 2026-10-12T08:35:24Z 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 2026-10-12T08:35:24Z

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 1 module(s) in 1 tier(s): mesh-issues

notes:
  - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42291 strings of 682 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
testing modules/mesh-issues
ok  	mesh-issues/cmd/mesh-issues	12.066s
--- its merge-check-typescript.sh (typescript toolchain)
**Merge check** at `d413da20` - `mesh/merge-gate`: **PASS** — every machine composes with the change as it did without (4 of 4 compose) (modules: 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-issues → novox; no bus step - tier 0: mesh-issues - novox: receives mesh-issues - read by no module's build: modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go, modules/mesh-issues/cmd/mesh-issues/triage_test.go Run by the build seat on g14 as `build-1791794163029908393` (`builds --log build-1791794163029908393`). ``` --- the gate: mesh-issues mesh-issues: ok, invokes claude-code.claude_code_judge, seat:mesh-delivery.checks, seat:operator-channel.notify, 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 1 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 2026-10-12T08:35:24Z 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 2026-10-12T08:35:24Z 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 1 module(s) in 1 tier(s): mesh-issues notes: - modules/mesh-issues/README.md, modules/mesh-issues/cmd/mesh-issues/bus_test.go, modules/mesh-issues/cmd/mesh-issues/import_test.go, modules/mesh-issues/cmd/mesh-issues/log_test.go, modules/mesh-issues/cmd/mesh-issues/manifest_test.go, modules/mesh-issues/cmd/mesh-issues/memory_test.go, modules/mesh-issues/cmd/mesh-issues/tracker_test.go and modules/mesh-issues/cmd/mesh-issues/triage_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 42291 strings of 682 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 testing modules/mesh-issues ok mesh-issues/cmd/mesh-issues 12.066s --- its merge-check-typescript.sh (typescript toolchain) ```
Author
Contributor

Review (independent): APPROVE

Head reviewed: d413da203c. I re-reviewed the diff since d33f3e3 against comment 9800, and re-ran my reproduction tests and new adversarial tests against a copy of this head.

B1 — an interrupted carry: fixed

carryOne (changes.go) now treats a kept value as already in the log when one of two things holds: its revision is at or below the highest revision that a carried operation of the issue holds, or it is a value with folded set (the tracker's own fold), which also resets everything before it.

I checked whether a value below the highest carried revision can be skipped without ever being carried. Carried operations are appended in increasing revision order, history oldest first and then the current value. A value that cannot be read now stops the carry with an error instead of being passed over. So the carried revisions form an unbroken run with no gap.

I checked the folded != 0 rule. A value with folded set is written over a value with folded 0 in only one place, carryOne's Update, and only after that value was carried. change carries first, catchUp returns early on folded 0, and the start skips folded 0. So the reset cannot drop an uncarried value.

The check kept[last].Revision < rev covers a read that is newer than the history. A history newer than the read makes the Update fail with ErrChanged, and the carry reads again, so the old […, newer, older] reordering is gone.

I re-ran my reproduction tests at this head:

  • Failure after v1: with values v1–v4 and the bus failing after the first append, the first carry says "1 of its 4 values" and records a fault. The second carry appends 3, and the log reads carried:v1 carried:v2 carried:v3 carried:v4.
  • Bucket write fails: every append lands but the bucket write fails, then an old instance writes v3, then the carry runs again. The log reads carried:v1 carried:v2 carried:v3-old-writer, and the bucket holds v3 folded.
  • Interrupted, then old writer: the carry is interrupted, an old instance writes v4, then the carry resumes. The log reads carried:v1 carried:v2 carried:v3 carried:v4-old, in order.

Two carriers at once are safe: the expect race plus matching by revision mean each value is carried once (TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce, which is a real interleave through between).

B2 — one issue stopping the start: fixed

A failing issue now becomes an entry in Carried.Faults, the other issues are carried, and Start still refolds, raises the counter and loads memory. An issue that was not carried stays readable and listed. Writes to it are refused with the reason, so nothing reaches the log before its carry.

MostIssueBytes is 252 KiB, and the fake log now enforces the 256 KiB entry limit. An existing issue between 252 and 256 KiB becomes a fault: it can be read but never written until someone shortens it outside the tracker. That is loud only in the start line. Before delivery, confirm that no live issue is near 252 KiB. nats.nats_buckets shows mesh-issues_issues at 12.4 MB over 2092 values kept, about 6 KB on average; I could not read the largest single value with the tools available.

(a)–(g)

  • (a) Pass. firstOf reads Read(1, key, 1), Last is gone, and the test interleaves an operation after a lost answer.
  • (b) Pass. At start, an issue whose entry does not fold is marked broken: none of its later entries is folded and its bucket value is not written.
  • (c) Pass. The start does not refold over a bucket value with folded 0, and the value is carried later as one more operation, after the operations already in the log. That order is correct, because the old instance's write was the newest bucket write.
  • (d) Pass. The comment matches the code, and every caller only reads.
  • (e) Pass. History is part of Store, and the fallback that carried only the current value is gone.
  • (f) Pass. go.sum keeps only the 9747f3e lines.
  • (g) Pass. See B1.

The fold check now runs at cleanup of every harness on the fake log.

Non-blocking

  1. An instance from before the switch can undo a note that was answered as made. It needs an old instance writing at the same time as a new one. Test on a copy: during the new note's append, an old instance writes the issue from the state before the note. The note is answered as made, store loses its compare-and-set, catchUp returns the folded 0 value, and the next carry appends the old writer's value. The log reads opened noted carried:old writer's title, and the current issue has 0 notes. The note survives only in the log's history.
    • Under the old code the loser read again and applied its change on top.
    • This cannot happen if one instance runs on novox and the old and new processes never overlap during the restart; confirm that is so.
    • Otherwise, when store loses to a folded 0 value, carry it and apply the change again instead of answering it as made.
  2. Matching by revision assumes revisions only grow. If the bucket's stream were ever recreated, its revisions would start again at 1, and a later value from an old instance could fall at or below the highest carried revision, be skipped, and be folded over. This is contrived: it needs a recreated bucket and an old writer. A comment would do.
  3. An issue that cannot be carried is visible only in the start line. Every write to it is refused, but nothing raises a condition. Consider one.

Runs

  • go test -race -count=1 ./... passes.
  • I built mesh-controller at main 3bd984c. MESH_GATE=<it> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh exits 0 at d413da2.
  • The SDK pin is still exactly v0.1.14-0.20261012082113-9747f3e1b1ac.

Required checks at d413da2 (read once)

  • mesh/merge-gate: success, "pass: builds mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)".
  • mesh/repo-check: success, "pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)".
  • mesh/delivery: success, "ready: it delivers once merged".

The merge order in the description still holds: mesh-tools #29 (mesh/state.history) and the mesh/log.* verbs must be live on novox before this is delivered.

Review (independent): APPROVE Head reviewed: d413da203c3aaa75580ec5f6e7b9205f341bdb87. I re-reviewed the diff since d33f3e3 against comment 9800, and re-ran my reproduction tests and new adversarial tests against a copy of this head. ## B1 — an interrupted carry: fixed `carryOne` (`changes.go`) now treats a kept value as already in the log when one of two things holds: its revision is at or below the highest `revision` that a `carried` operation of the issue holds, or it is a value with `folded` set (the tracker's own fold), which also resets everything before it. I checked whether a value below the highest carried revision can be skipped without ever being carried. Carried operations are appended in increasing revision order, history oldest first and then the current value. A value that cannot be read now stops the carry with an error instead of being passed over. So the carried revisions form an unbroken run with no gap. I checked the `folded != 0` rule. A value with `folded` set is written over a value with `folded` 0 in only one place, `carryOne`'s Update, and only after that value was carried. `change` carries first, `catchUp` returns early on `folded` 0, and the start skips `folded` 0. So the reset cannot drop an uncarried value. The check `kept[last].Revision < rev` covers a read that is newer than the history. A history newer than the read makes the Update fail with ErrChanged, and the carry reads again, so the old `[…, newer, older]` reordering is gone. I re-ran my reproduction tests at this head: - **Failure after v1:** with values v1–v4 and the bus failing after the first append, the first carry says "1 of its 4 values" and records a fault. The second carry appends 3, and the log reads `carried:v1 carried:v2 carried:v3 carried:v4`. - **Bucket write fails:** every append lands but the bucket write fails, then an old instance writes v3, then the carry runs again. The log reads `carried:v1 carried:v2 carried:v3-old-writer`, and the bucket holds v3 folded. - **Interrupted, then old writer:** the carry is interrupted, an old instance writes v4, then the carry resumes. The log reads `carried:v1 carried:v2 carried:v3 carried:v4-old`, in order. Two carriers at once are safe: the `expect` race plus matching by revision mean each value is carried once (`TestTwoInstancesCarryingOneIssueAtOnceCarryEachValueOnce`, which is a real interleave through `between`). ## B2 — one issue stopping the start: fixed A failing issue now becomes an entry in `Carried.Faults`, the other issues are carried, and `Start` still refolds, raises the counter and loads memory. An issue that was not carried stays readable and listed. Writes to it are refused with the reason, so nothing reaches the log before its carry. `MostIssueBytes` is 252 KiB, and the fake log now enforces the 256 KiB entry limit. An existing issue between 252 and 256 KiB becomes a fault: it can be read but never written until someone shortens it outside the tracker. That is loud only in the start line. Before delivery, confirm that no live issue is near 252 KiB. `nats.nats_buckets` shows `mesh-issues_issues` at 12.4 MB over 2092 values kept, about 6 KB on average; I could not read the largest single value with the tools available. ## (a)–(g) - **(a) Pass.** `firstOf` reads `Read(1, key, 1)`, `Last` is gone, and the test interleaves an operation after a lost answer. - **(b) Pass.** At start, an issue whose entry does not fold is marked broken: none of its later entries is folded and its bucket value is not written. - **(c) Pass.** The start does not refold over a bucket value with `folded` 0, and the value is carried later as one more operation, after the operations already in the log. That order is correct, because the old instance's write was the newest bucket write. - **(d) Pass.** The comment matches the code, and every caller only reads. - **(e) Pass.** `History` is part of `Store`, and the fallback that carried only the current value is gone. - **(f) Pass.** go.sum keeps only the 9747f3e lines. - **(g) Pass.** See B1. The fold check now runs at cleanup of every harness on the fake log. ## Non-blocking 1. **An instance from before the switch can undo a note that was answered as made.** It needs an old instance writing at the same time as a new one. Test on a copy: during the new note's append, an old instance writes the issue from the state before the note. The note is answered as made, `store` loses its compare-and-set, `catchUp` returns the `folded` 0 value, and the next carry appends the old writer's value. The log reads `opened noted carried:old writer's title`, and the current issue has 0 notes. The note survives only in the log's history. - Under the old code the loser read again and applied its change on top. - This cannot happen if one instance runs on novox and the old and new processes never overlap during the restart; confirm that is so. - Otherwise, when `store` loses to a `folded` 0 value, carry it and apply the change again instead of answering it as made. 2. **Matching by revision assumes revisions only grow.** If the bucket's stream were ever recreated, its revisions would start again at 1, and a later value from an old instance could fall at or below the highest carried revision, be skipped, and be folded over. This is contrived: it needs a recreated bucket and an old writer. A comment would do. 3. **An issue that cannot be carried is visible only in the start line.** Every write to it is refused, but nothing raises a condition. Consider one. ## Runs - `go test -race -count=1 ./...` passes. - I built mesh-controller at main 3bd984c. `MESH_GATE=<it> MESH_CHECK_MODULES=mesh-issues sh merge-check.sh` exits 0 at d413da2. - The SDK pin is still exactly `v0.1.14-0.20261012082113-9747f3e1b1ac`. ## Required checks at d413da2 (read once) - `mesh/merge-gate`: success, "pass: builds mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)". - `mesh/repo-check`: success, "pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-check-typescript.sh (typescript)". - `mesh/delivery`: success, "ready: it delivers once merged". The merge order in the description still holds: mesh-tools #29 (`mesh/state.history`) and the `mesh/log.*` verbs must be live on novox before this is delivered.
mesh-admin merged commit 6546c05f6e into main 2026-10-12 08:46:10 +00:00
mesh-admin deleted branch adr-0300-issues-on-a-log 2026-10-12 08:46:10 +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#228