Merge pull request 'ADR 0152: the operator's surface is a module, the console (group 3)' (#222) from feat/the-console into main

Reviewed-on: #222

00-META/checks/README.md

@@ -54,4 +54,6 @@ whose failure has never been observed is a guess about its own correctness.
 The development cycle, checked ([ADR 0080](../../02-DECISIONS/0080-the-development-cycle-is-checked.md)):
 a to-be design names a decision, an in-progress/implemented design names its owning code, a
 located/fixed issue names its owner, a fixed/resolved issue says what fixed it, a graduated
-research overview says what it became. `python3 00-META/checks/cycle.py`
+research overview says what it became, and no two issue records share a number (issue 155 — the
+number is how a record is cited, and `main` lags every open pull request, so two people reading it
+allocate the same one). `python3 00-META/checks/cycle.py`

00-META/checks/cycle.py

@@ -14,7 +14,8 @@ What is enforced:
           its owning code (`code:`) -- no development without a design that says where.
   issues  a known `status:`; once `located`, `located-in:` names the owner;
           once `resolved`, `fixed-by:` says what fixed it (prose counts --
-          "nothing, the capability existed" is an answer).
+          "nothing, the capability existed" is an answer). And no two records share a
+          number -- the number is how a record is cited.
   research a known `status:`; a `graduated` overview says what it `became:`, and every
           target it names exists.
   decisions every accepted record is REACHABLE from the cycle: cited by a design doc's
@@ -109,6 +110,27 @@ def main():
                               "without a design that says where" % status)
 
     # ---- issues ------------------------------------------------------------------------
+    # Two records may not share a number. Numbers are taken as "next free after main", and work
+    # sits on unmerged branches for days -- so two people reading the same main allocate the same
+    # number, and nothing said so. It happened twice in one evening between two machines, and the
+    # second collision landed on main with all three checks passing (issue 155). An issue number is
+    # how every other record cites this one; two records answering to it means a pointer that
+    # resolves to whichever the reader happened to open.
+    seen = {}
+    for folder in sorted(glob.glob(os.path.join(ROOT, "04-ISSUES", "*", ""))):
+        name = os.path.basename(os.path.normpath(folder))
+        number = name.split("-", 1)[0]
+        if not number.isdigit():
+            continue
+        if number in seen:
+            bad(os.path.join("04-ISSUES", name),
+                "is numbered %s, and so is %s -- an issue number is how it is cited, and two "
+                "records answering to one means a citation that resolves to whichever the reader "
+                "opened. Take the next free number across main AND every open pull request"
+                % (number, seen[number]))
+        else:
+            seen[number] = name
+
     for path in sorted(glob.glob(os.path.join(ROOT, "04-ISSUES", "*", "00-report.md"))):
         front = frontmatter(path)
         if front is None:

00-META/glossary.md

@@ -75,6 +75,18 @@ another — and a mesh you cannot name precisely is a mesh two people describe d
   is a role you occupy, and the two meet where a seat delivers a provision: occupying the seat is
   what makes a module *the* provider of it.
 
+## The surfaces
+
+- **console** — the module (`mesh-console`) that puts the mesh's tools in front of whoever is on a
+  machine: an MCP endpoint on the machine's loopback for an agent, the same endpoint for a person. It
+  is assigned like any module, holds a credential the mesh minted, and calls tools under a grant its
+  manifest declares (`invokes`). Loopback is the authority boundary: whoever is on the machine owns the
+  mesh there ([ADR 0152](../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md)).
+  Not "the tool bridge", "the brain" or "the MCP server" — those name the predecessor's program or a
+  protocol, and the console is a module.
+- **invokes** — the manifest word for the tools a module calls, `<module>.<tool>` each or `*` for
+  every one. A grant on the publish side and nothing else; a module that declares none calls nothing.
+
 ## How this page is kept
 
 A new name for an existing thing lands here first, in the same change that introduces it in code. A

00-META/process/03-issues.md

@@ -21,7 +21,12 @@ incident someone must **clear**.
 
 ## Steps
 
-1. Take the next free number. Create `04-ISSUES/NNN-short-name/00-report.md`:
+1. Take the next free number — **across `main` and every open pull request**, not `main` alone.
+   Work sits on unmerged branches for days, so two people both reading `main` allocate the same
+   number; it happened twice in one hour between two machines, and the second collision reached
+   `main` with every check passing (issue 155). `cycle.py` now refuses two records sharing a number,
+   which catches a collision but does not prevent one. Create
+   `04-ISSUES/NNN-short-name/00-report.md`:
 
    ```yaml
    ---
@@ -42,6 +47,14 @@ incident someone must **clear**.
 ## Rules
 
 - Closed issues are never deleted — they are the mesh's symptom-to-component memory.
+- `fixed-by:` names something that will still exist: a commit or a pull request, never a branch. A
+  branch is deleted when it merges, so a branch name there is a pointer that resolves to nothing by
+  the time anybody follows it.
+- A fix that turns out to have broken something else is written back into the record that asked for
+  it, pointing at the new issue. Somebody arriving at a record to learn why the code is the way it
+  is must not have to already know there was a sequel.
+- Renumbering a collision happens once, in the branch that lands last. Renumbering a branch whose
+  author is still pushing only moves the race.
 - An issue whose answer is a general lesson should also be written to the knowledge base, so
   the next person searching a symptom finds it. Both, not either.
 - `status: wontfix` is legitimate and requires a sentence saying why.

02-DECISIONS/0016-the-lab.md

@@ -13,6 +13,14 @@ decisions taken over three days; the reasoning is kept, the fragmentation is not
 
 The environment a change is run against before it reaches real machines.
 
+> **Still the lab, no longer the test bed — 2026-09-30, by [ADR 0149](0149-the-live-mesh-is-the-test-bed.md).**
+> Everything here stands. What changed is what the lab is *for*: a change is verified against the mesh
+> that is running, because the faults that cost the most are faults of a mesh that already exists —
+> bound consumers, containers made against an older roster, an adopted machine — and a bed is by
+> construction a mesh that does not. Raising a mesh from bare is now the lab's whole job, which is the
+> one thing the live mesh cannot be asked to do. 0149 also supersedes
+> [ADR 0068](0068-the-lab-takes-requests.md), which extended this one and was never built.
+
 ## A node in the lab is a virtual machine
 
 It boots a stock Linux image, runs the real install, and becomes a node. **It is not a model of

02-DECISIONS/0037-where-a-module-lives.md

@@ -1,6 +1,6 @@
 ---
 topic: building it
-status: proposed
+status: accepted
 date: 2026-09-01
 deciders: jochen
 reconstructed: false
@@ -101,3 +101,19 @@ the digest down after building.
 **Whether kind 4 deserves a module at all.** Thirty-five descriptions that say *install this and
 write these files* may be better as one module with settings than as thirty-five modules. Left
 open deliberately; it is a question about the shape of the catalogue, not about whether to have one.
+
+## Accepted, 2026-09-29, against what was built
+
+*Marked in a grooming pass: the mesh was built to this record and the record still said `proposed`.*
+
+The proposal is the arrangement that exists. `mesh-catalog` holds descriptions of software we did
+not write and the programs that provision it, and holds neither the mesh's own components nor an
+application's own module. The mesh's list of modules is a table in the control plane, filled by
+`module add`, and every module records the source it came from with the commit it was read at.
+
+**One half is not built: `module check` as a command on the control plane's binary.** A manifest is
+still validated by a test that reaches into the control plane's internals — which works for this
+catalogue and gives nothing at all to somebody describing their own application in their own
+repository, which this record says is the case that matters most. That is
+[issue 148](../04-ISSUES/148-a-manifest-outside-this-catalogue-has-no-check/00-report.md).
+

02-DECISIONS/0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md

@@ -26,6 +26,14 @@ runtime is per-module, not per-node, and treating the audit-logger as special le
 modules' code with nothing to run it: the conversion produced tools and events that, as it stands,
 never execute.
 
+> **The hosting form is settled elsewhere — 2026-09-30.** Where this record says "a container", read
+> [ADR 0150](0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md): a module's own
+> code runs as supervised processes under this record's one account. Nothing else here changes — the
+> per-module runtime, the per-tool key and the single scoped account are the argument this record made
+> and they are why 0150 goes the way it does. The note is here because two design documents chose the
+> other form without knowing this record existed
+> ([issue 117](../04-ISSUES/117-a-modules-own-code-is-a-container-and-a-process/00-report.md)).
+
 ## Decision
 
 ### A module with tools or events runs a process of its own

02-DECISIONS/0066-public-routing-is-name-agnostic.md

@@ -80,6 +80,15 @@ reaching the routed name, which the clause above has just made resolvable inside
 three are one decision: **compose the name, propagate it, certify it** — each is meaningless without
 the one before it.
 
+> **The mechanism changed — 2026-09-30, by [ADR 0148](0148-the-meshs-names-are-resolved-not-copied-into-containers.md).**
+> A routed name still reaches every asker in the mesh, which is what this record decided and it stands.
+> It no longer reaches them by being written into each declared container: copying the roster in made the
+> roster part of every container's identity, so one name moving replaced every container in the mesh
+> ([issue 151](../04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)). A
+> container resolves through its machine's resolver instead. The consequence below — that an internal
+> issuer's challenge needs the routed name resolvable inside the mesh — holds unchanged, by the means the
+> machine itself already uses.
+
 ## Consequences
 
 - **Lab-versus-production is one node-level `public-domain` setting**, not an override on every

02-DECISIONS/0068-the-lab-takes-requests.md

@@ -1,14 +1,21 @@
 ---
 topic: building it
-status: proposed
+status: superseded
 date: 2026-09-12
 deciders: jochen
 reconstructed: false
 extends: 0016-the-lab.md
+superseded-by: 02-DECISIONS/0149-the-live-mesh-is-the-test-bed.md
 ---
 
 # 68. The lab takes requests, one at a time, and runs each from its own copy
 
+> **Superseded — 2026-09-30, by [ADR 0149](0149-the-live-mesh-is-the-test-bed.md).** Never built. The
+> live mesh became the test bed, because the faults that cost the most are faults of a mesh that
+> already exists — bound consumers, containers made against an older roster, an adopted machine — and a
+> bed is by construction a mesh that does not. The rule worth keeping from below is that a run reads a
+> copy that is not anybody's working tree.
+
 ## Context
 
 **The lab is exclusive hardware, and today a person holds it.** Raising a scenario takes over

02-DECISIONS/0080-the-development-cycle-is-checked.md

@@ -35,6 +35,16 @@ The development cycle is enforced mechanically, to the extent frontmatter can ca
   capability existed" is an answer).
 - **No silent graduation** — a `graduated` research overview says what it `became:`, and the
   targets exist.
+- **No two records answering to one number** — added 2026-09-30; see the insight below.
+
+> **Progressive insight — 2026-09-30.** The list above named four things `cycle.py` enforces, and
+> now names five. Nothing enforced that two issue records hold different numbers: two machines
+> filing issues within one hour both read `main`, both took "the next free number", and collided
+> twice — the second collision reaching `main` with `records.py`, `cycle.py` and `index.py` all
+> reporting success (04-ISSUES/155). A number is how every other record cites one, so two records
+> answering to it is a citation that resolves to whichever folder the reader opened. `cycle.py`
+> refuses it now. The decision here stands exactly as written: this is one more thing frontmatter
+> and file names can carry, found by its absence rather than by reasoning.
 
 [`00-META/checks/cycle.py`](../00-META/checks/cycle.py) refuses violations, beside `records.py`
 and `index.py`; all three run before any merge here. What frontmatter cannot see — that code work

02-DECISIONS/0095-the-control-plane-is-the-way-to-ask-a-module.md

@@ -47,6 +47,14 @@ Anything with the control plane in reach can ask any module anything it serves.
 harder: nothing outside the control plane can, and the control plane's connection is one more
 thing on the path of every question — a cost accepted for the audit it buys.
 
+> **The mechanism changed — 2026-09-30, by ADR 0152.** What stands: `ask` on the control plane, and
+> that every call passes an account whose permission list says what it may ask. What moved: "nothing
+> outside the control plane can" stopped being true when a person's account gained a publish grant per
+> tool (design 25 §7, 2026-09-28), and [ADR 0152](0152-the-operators-surface-is-a-module-the-console.md)
+> takes the first option above for a module as well — a manifest declares `invokes`, and the bus grants
+> exactly that publish side. The audit the second option bought is the bus's permission list, which
+> derives both.
+
 ## How it is checked
 
 A tools-only bed asks a served tool through the control plane and asserts an answer arrived —

02-DECISIONS/0113-the-vault-makes-every-secret.md

@@ -1,6 +1,6 @@
 ---
 topic: what runs on it
-status: proposed
+status: accepted
 date: 2026-09-25
 deciders: jochen
 reconstructed: false
@@ -275,3 +275,11 @@ On acceptance, each of these is superseded or amended by this record, not edited
   everything a module needs is a requirement
 - [Issue 095](../04-ISSUES/095-a-module-assigned-after-genesis-has-no-broker-account/00-report.md),
   [issue 103](../04-ISSUES/103-a-container-is-not-recreated-when-a-file-it-reads-changes/00-report.md): what fails today
+
+## Accepted, 2026-09-29, against what was built
+
+*Marked in a grooming pass.* The vault is a module providing `secret` at mesh scope, and six
+modules in the catalogue require it — so a shared secret is a requirement answered by the vault,
+which is what this record asks for. Private keys are still made where they are used and never
+travel, which is the other half and was never in question.
+

02-DECISIONS/0114-a-shared-credential-rotates-over-two-credentials.md

@@ -1,6 +1,6 @@
 ---
 topic: what runs on it
-status: proposed
+status: accepted
 date: 2026-09-26
 deciders: jochen
 reconstructed: false
@@ -189,6 +189,23 @@ On acceptance, each of these is amended by this record, not edited:
   credential is no longer all-or-nothing with a window. It overlaps, with each step confirmed.
   A single-party credential is staged, not replaced.
 
+## Accepted, 2026-09-30, and not scheduled
+
+Accepted as written. The separation it draws — a consumer's *resource* and a *credential that reaches
+it* are different things with different lifecycles — is the part that had to be settled, because the
+alternative is what the record was written against: retiring a credential taking the data it reached
+with it. That is a data-loss shape, and a record that names it should not sit unresolved while the
+code that could hit it is being written.
+
+**It is not built, and accepting it does not schedule it.** The SDK's provisioner adapter is still
+`create` / `remove` / `holds` rather than the four operations above, and no provider implements the
+two-credential rotation. Accepted-and-not-built is an ordinary state here — 0141 and 0142 are both in
+it — and it is the honest one: leaving this `proposed` made it invisible to anyone reading what the
+mesh has decided, while changing nothing about what runs.
+
+The work it implies belongs with the provisioner contract, beside
+[issue 124](../04-ISSUES/124-a-consumer-cannot-be-told-what-its-provider-derived/00-report.md).
+
 ## Consequences
 
 - **Every credential provider's adapter changes**, in two steps. The first separates *retire a

02-DECISIONS/0115-one-assignment-of-a-module-per-node.md

@@ -1,6 +1,6 @@
 ---
 topic: what runs on it
-status: proposed
+status: accepted
 date: 2026-09-26
 deciders: jochen
 extends: 0112-a-module-definition-names-no-node-mesh-or-path.md
@@ -47,3 +47,10 @@ other boundary already is: the module name.
   node runs one of each (ADR 0115)" — instead of failing on whichever name collides first.
 - Multi-tenant asks are answered in the catalogue (a second module definition), not in the
   control plane.
+
+## Accepted, 2026-09-29, against what was built
+
+*Marked in a grooming pass.* The rule is enforced where it cannot be forgotten: `assignment`'s
+primary key is `(node, module)`, so a second assignment of one module to one machine is not a thing
+the mesh can hold. The record read `proposed` while the schema had already settled it.
+

02-DECISIONS/0118-undeclaring-gives-a-unit-back-the-state-it-was-found-in.md

@@ -12,7 +12,7 @@ extends: 0102-the-mesh-writes-into-a-shared-file-never-over-it.md
 ## Context
 
 When a resource stops being declared — its module unassigned, the node sent a
-deliberately-empty declaration ([issue 127](../04-ISSUES/127-a-declaration-that-shrinks-to-empty-is-skipped-not-sent/00-report.md)),
+deliberately-empty declaration ([issue 149](../04-ISSUES/149-a-declaration-that-shrinks-to-empty-is-skipped-not-sent/00-report.md)),
 or a new catalogue version renaming its id — the host undoes it. The host's own code states
 the rule it means to follow: **it removes what it made and leaves what it merely configured.**
 For almost every resource it does exactly that:

02-DECISIONS/0141-the-host-delivers-its-own-successor.md

@@ -109,6 +109,23 @@ understated as none. The remaining work is a way to build the host and a way to
 path, and until both exist nothing delivers a version and every machine takes the fallback — which is
 what every machine does today.
 
+> **Progressive insight — 2026-09-30. Both of those exist now.** The paragraph above named two missing
+> things and they are built: a Go toolchain, based on a new `mesh-tools-go` module so the compiler is
+> named and not pinned, and `${version}` in any value of a resource that uses an archive or a bundle.
+> The mesh compiles its own host and publishes it to its own registry, measured — a statically linked
+> stripped binary, fetched back out and run. **The cost was larger again than this note said**: three
+> more things in the path assumed one language or one shape, and a fourth was in the base image.
+> The account is [issue 142](../04-ISSUES/142-the-host-is-the-one-thing-the-mesh-does-not-deliver/01-progress.md).
+>
+> The version in a path is the artifact's **digest**, not the commit this note's own wording would
+> suggest. Two builds of one commit are meant to be the same bytes, so a content-addressed version
+> means an unchanged build keeps the path it had; a commit-named one would move for an identical binary
+> and recreate everything reading it.
+>
+> **Still nothing delivers a version to a machine.** The host is a module and builds, and declares no
+> resources, so the bundle sits in the registry and no machine is asked to take it. That is the next
+> piece, and the decision above is unchanged by any of this.
+
 ## Consequences
 
 - **The host becomes a build target and a module** — a module whose resource is the next host, applied

02-DECISIONS/0145-a-module-checks-what-the-mesh-claims-is-reachable.md

@@ -0,0 +1,120 @@
+---
+topic: what runs on it
+status: superseded
+date: 2026-09-29
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0144-anything-on-a-machine-may-call-anything-on-it.md
+superseded-by: 02-DECISIONS/0146-connectivity-is-checked-by-name-per-hosting-form.md
+---
+
+# 145. A module checks what the mesh claims is reachable, and it checks itself
+
+## Context
+
+The mesh asserts three things are callable ([ADR 0144](0144-anything-on-a-machine-may-call-anything-on-it.md)):
+what runs on the same machine, another machine's service exposed to the private network, and another
+machine's service exposed publicly. It has never checked any of them.
+
+[Issue 145](../04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md):
+the first of the three was broken for eleven hours and the mesh answered *all heard from, every module
+current with its source* throughout. Every check it makes is about the relationship between the mesh and
+a machine — applied, current, containers running — and none about whether anything can reach anything.
+
+**A first answer was drafted and withdrawn.** [ADR 0143](0143-a-consumer-verifies-the-grant-it-is-given.md)
+put the check inside the host, verifying each grant from the consumer's network position. It was
+superseded because the difference it worked so hard to reproduce — whether a caller sat in a container —
+was the bug itself. What survives from it is the part that was right: a check run from the wrong place
+proves nothing, and the mesh's own reports are not evidence about the network.
+
+**The mesh already has the shape for this and it is a module.** A module can declare a container that
+runs on a cadence ([ADR 0053](0053-a-step-that-runs-on-a-schedule.md), and three modules already use
+`*/5 * * * *`), can be given the mesh's roster as a rendered fact — every machine's name, address and
+this node's own identity, the same mechanism the resolver and the operator's ssh configuration use — and
+can emit what it found on the bus. Nothing new is needed to build this except the module.
+
+**What it must not check is the trap.** The obvious probe target is ssh: present on every machine, never
+closed by design. Dialling it would have passed throughout the outage, because ssh is admitted
+unconditionally and the thing that broke was a service exposed to the private network. A checker whose
+probe is unconditionally open measures the one path that cannot fail, which is the failure this whole
+sequence keeps producing — a check that reads as verification and verifies nothing.
+
+## Considered Options
+
+1. **The host verifies each grant** ([ADR 0143](0143-a-consumer-verifies-the-grant-it-is-given.md)).
+   Superseded. It needed the host to act from another network position, which is machinery that exists
+   only while local calls are filtered wrongly.
+2. **The control plane dials every node.** Rejected: it sits on one machine and reaches the others by a
+   path no ordinary caller uses. It would have passed throughout the outage.
+3. **Probe an existing service.** Rejected for the target problem above: the services guaranteed on every
+   machine are the ones that are never closed, so they cannot fail the way the mesh fails.
+4. **A module on every machine that serves its own probe and dials the others'.** Adopted.
+
+## Decision
+
+**A module runs on every machine, serves an endpoint of its own, and dials every other machine's.** The
+probe is the module's own endpoint, declared reachable over the private network — so the thing being
+dialled is admitted by exactly the rule that governs every other internally-exposed service, and fails
+when that rule is wrong. A second endpoint, declared public, does the same for the public path where a
+machine has one.
+
+**It checks the three cases the mesh claims, by name:**
+
+- its **own machine**, by dialling its own machine's address — the case that broke, and the only one that
+  distinguishes a caller on the machine from a caller in one of its containers;
+- **each other machine over the private network**;
+- **each machine's public path**, where one is recorded.
+
+**It resolves before it dials, and says which failed.** A name that does not resolve and a port that does
+not answer are different faults with different owners, and a checker that reports one sentence for both
+sends a reader to the wrong place.
+
+**It runs where the callers run.** The module's own code in its own container, on the cadence the mesh
+already has, from the same position as every other module on that machine. It is not the host and not the
+control plane, and that is the whole point.
+
+**It says what it found and nothing else.** It emits results; it repairs nothing, opens nothing and holds
+no credential beyond its own. A checker that fixes things is a second control plane.
+
+**One failure is not a fault.** A machine rebooting is ordinary. A path is reported broken after it has
+failed on consecutive runs, and the count travels with the result so a reader can tell "briefly away"
+from "never worked" — the one thing [ADR 0143](0143-a-consumer-verifies-the-grant-it-is-given.md) got
+right and worth keeping.
+
+## Consequences
+
+- **The mesh gains the ability to be wrong out loud about the network.** Eleven hours becomes two runs.
+- **It is a module, so it is assigned, built, pushed and reported on like everything else** — no new host
+  capability, no new vocabulary, nothing in the control plane that has to know about checking.
+- **Its own endpoint is the instrument.** That is what makes it able to fail; it also means the checker
+  must be assigned to a machine before that machine can be checked, and a machine without it is
+  unchecked rather than healthy.
+- **It cannot check what it cannot be told.** The roster gives it machines; it does not give it every
+  module's endpoints, so this checks the paths the mesh claims and not every grant in the mesh. That is
+  the honest scope of a first one, and the difference is worth saying rather than growing quietly.
+- **What got harder:** one more module on every machine, and a module whose whole purpose is to fail
+  visibly when something else is wrong. Its own failures will be read as the mesh's, which is the cost of
+  an instrument.
+
+## How it is checked
+
+- **It catches the measured outage.** A bed closes the path from a container to a service exposed to the
+  private network on its own machine — issue 145's shape — and the checker reports its own machine
+  unreachable while every other path still reads reachable. This fails against a probe on a port that is
+  never closed, which is the wrong target this record exists to name.
+- **A machine rebooting is not a fault**: one failed run reports nothing, the count rises and falls.
+- **A name that does not resolve is reported as that**, not as a port that did not answer.
+- **It reports and does not act**: asserted by giving it a broken path and checking nothing on the machine
+  changed.
+- **A machine without the module reads unchecked**, never healthy — asserted on what the mesh says about
+  a machine it is not assigned to.
+
+## References
+
+- [ADR 0144](0144-anything-on-a-machine-may-call-anything-on-it.md) — the three things that must be callable
+- [ADR 0143](0143-a-consumer-verifies-the-grant-it-is-given.md) — superseded; what survives is that the
+  position matters
+- [ADR 0053](0053-a-step-that-runs-on-a-schedule.md) — the cadence
+- [ADR 0138](0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md) — internal and public,
+  which the probe endpoints declare
+- [issue 145](../04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md)

02-DECISIONS/0146-connectivity-is-checked-by-name-per-hosting-form.md

@@ -0,0 +1,125 @@
+---
+topic: what runs on it
+status: accepted
+date: 2026-09-29
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0145-a-module-checks-what-the-mesh-claims-is-reachable.md
+supersedes: 02-DECISIONS/0145-a-module-checks-what-the-mesh-claims-is-reachable.md
+---
+
+# 146. Connectivity is checked by name, per hosting form, with a valid certificate
+
+## Context
+
+[ADR 0145](0145-a-module-checks-what-the-mesh-claims-is-reachable.md) decided that a module checks what
+the mesh claims is reachable, from where the callers are, because the mesh reported four machines healthy
+for eleven hours while a module could not reach its database
+([issue 145](../04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md)).
+That decision stands. What it got wrong is everything about *what* is dialled.
+
+It dialled a raw port on each machine's address. Three things are wrong with that:
+
+- **A raw port is not how anything in this mesh is reached.** A real caller resolves a name, the proxy
+  answers it, and the proxy reaches the service. A check that dials a port tests the last hop of a path
+  with four hops in it, and the three it skips — resolution, the proxy, the certificate — are where most
+  of the mesh's connectivity actually lives.
+- **It tested one hosting form.** A module is software that delivers services, and it may deliver them
+  from a container, from a unit the mesh writes for its own code, or from a unit a package ships. Those
+  are three different paths to the same machine, and the outage that produced this was two of them
+  disagreeing. A probe served one way measures one way.
+- **It said nothing about certificates.** An internal name that resolves, routes and answers over TLS
+  that nothing can verify is not a working path; it is a working path for whoever holds the proxy's
+  trust and nobody else.
+
+## Decision
+
+**Each hosting form gets its own endpoint, its own route and therefore its own name.** On every machine:
+
+| name | what serves it |
+|---|---|
+| `connect-docker.<node>.internal` | a container |
+| `connect-process.<node>.internal` | the mesh's own code, in a unit the mesh writes |
+| `connect-unit.<node>.internal` | a unit a package ships |
+
+and the same set under each machine's public domain where it has one — `connect-docker.<domain>` and its
+siblings. The names are the instrument: a failure reads as *`connect-docker.g14.internal` did not answer*,
+which says which machine and which hosting form without anybody interpreting anything.
+
+**Every machine checks every machine, by name, over TLS, verifying the certificate.** Not a port, not an
+address: resolve the name, connect, complete the handshake, check the certificate against the authority
+that should have issued it — the mesh's own for an internal name, a public one for a public name. That is
+the whole path a real caller takes, and each step failing is reported as itself.
+
+**No name is written anywhere.** The machines come from the roster the mesh already renders as a fact, and
+the labels are the module's. A machine that joins appears in every other machine's roster on the next
+push, and they begin checking it without an edit.
+
+**And the module arrives on a machine because the machine exists, not because somebody assigned it.** A
+machine that joins and does not have it is worse than unchecked: every other machine is already dialling
+its names, so it reads as broken everywhere until someone notices. This is the part the mesh cannot
+currently express — see below — and it is the part that makes the rest safe.
+
+**What survives from 0145**, unchanged: it reports and repairs nothing; one failure is not a fault and a
+path is broken after consecutive runs with the count travelling with the result; findings are said on the
+bus, because a finding in a file on the machine is what this exists to end; and the bus is the one path
+that cannot report its own failure, so an emit that does not land is written locally and nowhere else.
+
+## What this needs that the mesh does not have
+
+Named here rather than assumed, because each is a decision of its own and this record is not the place to
+make them:
+
+1. **A module that every machine has.** `ScopeNode` means *at most one holder per node* — an exclusivity
+   rule, not an obligation — and nothing assigns a module at enrolment. Today the resolver, the packet
+   filter, ssh and intrusion prevention are each assigned per machine by hand, which is the same gap
+   wearing different clothes.
+2. **A container running a module's own bundle.** A `process` runs the mesh's own compiled code with no
+   image; a `container` needs an image of the module's own, which means a Dockerfile — the thing the
+   `bundle` artifact exists to abolish. Nothing in the catalogue runs a bundle in a container, so
+   `connect-docker` has no shape yet.
+3. **A unit a package ships, for `connect-unit`.** The `service` resource puts an existing unit into a
+   state and deliberately installs none, so this form needs a package that serves a port — and naming a
+   program the machine may not have is
+   [issue 136](../04-ISSUES/136-a-module-may-name-a-program-the-machine-does-not-have/00-report.md).
+4. **A machine's public domain in the roster fact.** The fact carries each machine's name, mesh name,
+   address and operator account. The public names cannot be composed without the domain.
+5. **Something that installs the mesh's own root on a machine.** This is
+   [issue 129](../04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md), open
+   since before any of this. Until it is closed, every internal name will fail certificate verification
+   from every machine — correctly, because nothing can verify it. That is the checker working, and it is
+   worth saying in advance so the first run is not read as the checker being broken.
+
+## Consequences
+
+- **A failure names the machine and the hosting form.** That is the whole gain over a port: eleven hours
+  became two runs under 0145, and under this it also becomes one line that says where to look.
+- **The checker surfaces issue 129 immediately**, and will report every internal name unverifiable until
+  it is fixed. A reader must be told that before the first run rather than after.
+- **Five things must be built before this is what it says it is**, and until they are, what exists is a
+  port dial from one position — useful, and not this.
+- **What got harder:** a module with three hosting forms of the same trivial service is a strange thing to
+  read. It is justified only because those three forms are how the mesh actually runs software, and a
+  checker that tested one of them would keep the class of outage it exists to catch.
+
+## How it is checked
+
+- **A name per hosting form answers from every machine**, asserted by name and not by port.
+- **A certificate that does not verify is reported as that**, distinctly from a name that does not resolve
+  and a port that does not answer — three faults, three owners.
+- **A machine that joins is checked by every other machine without an edit**, asserted by adding one to a
+  bed and looking at what the others dial on their next run.
+- **A machine that joins has the module**, which is gap 1 above and is the assertion that cannot be
+  written yet.
+- **The measured outage is still caught**: the path from a container to a service on its own machine is
+  closed and `connect-docker.<that node>.internal` fails from that machine while the others still pass.
+
+## References
+
+- [ADR 0145](0145-a-module-checks-what-the-mesh-claims-is-reachable.md) — superseded; its core stands
+- [ADR 0138](0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md) — the two reaches these
+  names come from
+- [ADR 0066](0066-public-routing-is-name-agnostic.md) — a label plus a domain, which is why no name is written
+- [issue 129](../04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md) — what the
+  internal names will fail on until it is closed
+- [issue 145](../04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md)

02-DECISIONS/0147-a-module-anchors-the-meshs-authority.md

@@ -0,0 +1,169 @@
+---
+topic: what runs on it
+status: accepted
+date: 2026-09-29
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0098-a-fact-a-provider-makes-at-first-start-is-fetched-from-it.md
+---
+
+# 147. A module anchors the mesh's authority on a machine, and takes it away again
+
+## Context
+
+The mesh runs its own certificate authority and every internal name is served with a certificate
+from it. No machine trusts it. On an enrolled, adopted workstation — on the private network,
+resolving through the mesh's resolver — every internal HTTPS name fails verification with
+*unable to get local issuer certificate*
+([issue 129](../04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md)).
+The certificates are genuine; nothing on the machine has ever been told what issued them.
+
+The authority's only consumer today is a proxy, which fetches the root into a directory of its own
+and hands it to one program ([ADR 0098](0098-a-fact-a-provider-makes-at-first-start-is-fetched-from-it.md)).
+That is enough for the proxy and for nothing else: a browser, `git` over HTTPS, `curl`, a package
+manager and every module that calls another module by an internal name read the machine's trust
+store, which holds the predecessor's authority and a developer tool's local root, and nothing of
+the mesh's.
+
+The predecessor wrote its root into every machine it set up. Removing it was deliberate — an
+honest failure beats a name that verifies for the wrong reason — and it leaves the mesh with no
+answer at all until this one lands. It is also what keeps the predecessor alive on the machines
+that still speak TLS to a mesh name.
+
+**What makes this a decision rather than a patch** is where the knowledge goes. Two mechanisms in
+the mesh already write things onto a machine because it is on the private network: `/etc/hosts`
+and the registry's plaintext trust ([ADR 0082](0082-the-registry-is-reached-by-name-and-trusted-by-the-overlay.md)).
+Following that precedent, the controller would inject an anchor into every such machine's
+declaration, and issue 129 proposed exactly that. It would work. It would also put *where this
+operating system keeps trust anchors* and *which command refreshes its bundles* into the control
+plane, for a fact the control plane does not have (the root does not exist until the authority has
+run) and a machine that may have no reason to verify a mesh name at all.
+
+## Considered Options
+
+1. **The controller injects the anchor into every machine on the private network**, the
+   `/etc/hosts` and insecure-registry shape. Rejected: being on the network is what makes the
+   registry reachable, and that is why network presence is the right trigger *there* — the trust
+   and the reachability are the same fact. Trusting an authority is not the same fact as being
+   able to reach it, and the anchor's path and the bundle refresh are a property of the machine's
+   operating system, which is the host's half of the mesh, not the controller's.
+2. **A new host primitive — a `trust-anchor` resource type.** Rejected for now, not on principle.
+   The host's vocabulary should grow when a shape cannot be said with what exists, and this one
+   can: a file and a service already express it, as the packet filter proves
+   ([ADR 0140](0140-the-filter-constrains-what-arrives-from-outside.md), whose module writes a
+   unit file and a service and nothing else). The primitive becomes right the moment a second
+   operating system is in play, because the anchor directory and the refresh command are exactly
+   the difference `internal/system` exists to hold. Until then it would be a vocabulary word with
+   one speaker.
+3. **A module that requires the authority, fetches its root, installs it as an anchor and
+   refreshes the machine's bundles — and removes both when it is no longer assigned.** Adopted.
+
+## Decision
+
+**A machine trusts the mesh's authority because a module put its root there, and stops trusting it
+when that module is taken away.**
+
+1. **The module requires `internal-acme-ca`** and reads the provider's bound address and the path
+   it serves its root at. It requires nothing else and provides nothing: it is a consumer of the
+   authority like any other.
+2. **It fetches the root over the mesh's own network, without prior trust**, because there is no
+   prior trust to have — this is the module that establishes it — and the network is what
+   authenticates the fetch ([ADR 0098](0098-a-fact-a-provider-makes-at-first-start-is-fetched-from-it.md),
+   the same reasoning that lets the proxy fetch it). What it accepts is checked: a body that is
+   not a certificate fails, and the failure is the module's, not a later handshake's.
+3. **It installs the root where this machine's TLS clients look, and refreshes the extracted
+   bundles** — the command that does the refresh is an ordinary part of the unit that places the
+   anchor, not a new thing the mesh can be asked to do.
+4. **Removal is symmetric and is the same unit's business.** Undeclared, the host stops the unit;
+   stopping it removes the anchor and refreshes the bundles again. A machine that leaves the mesh
+   stops trusting the mesh, without anybody remembering to go and look.
+5. **It is an ordinary assignment.** No machine is given it automatically. A machine that verifies
+   a mesh name is assigned it, and a machine that does not is not — which is the same statement
+   the mesh already makes about every other module, and is why this is not the controller's
+   business.
+
+**One operating system, said out loud.** The anchor directory and the refresh command in the
+module today are Arch's. On a machine that is not Arch the unit fails, visibly, rather than
+writing a file nothing reads. That is the accurate failure, and it is the signal that option 2
+above has become right.
+
+## How this is checked
+
+- **The verification that could not succeed before.** On a machine holding the module, a plain
+  client fetches an internal HTTPS name with no `-k` and no bundle argument and verifies. On a
+  machine without it, the same fetch fails with *unable to get local issuer certificate*. Both
+  halves, because only the pair distinguishes "the anchor works" from "something else already
+  trusted it".
+- **The removal half, in the same bed:** unassign the module, refetch, and the failure returns.
+  Checking only the arrival is how a trust store fills up with authorities nobody can account for.
+- **What is deliberately not checked here:** that the authority issues, that a name resolves, that
+  the proxy serves. Those have their own beds, and this module's bed passing for those reasons is
+  the failure mode this record is most exposed to — which is why the negative half is not optional.
+
+**What this bed is dialled at, and why it is the authority itself.** The authority serves its own
+API with a certificate it issued, so the handshake under test needs nothing else in the mesh to be
+right. A trust bed that reached for a routed name through the proxy would be passing or failing for
+the proxy's reasons and the resolver's.
+
+**Written, and not yet run** *(2026-09-29)*. The bed is `trust-anchor` in the lab, and it cannot
+execute: raising a foundation fails before any module is reached, in both bundles that exist
+([issue 146](../04-ISSUES/146-the-foundation-cannot-be-raised-on-the-bus-the-mesh-runs-on/00-report.md)).
+So what stands behind this record today is the rendering — the script the machine would run names
+the authority it was bound to, checked in the control plane's own test suite — and **not** a machine
+that verified anything. That is a weaker thing than the paragraph above describes, and it stays
+written this way until the bed runs.
+
+> **Progressive insight — 2026-09-30. It has now been run, on the live mesh rather than in the bed.**
+> The paragraph above said nothing had verified anything, and something has. The module was registered
+> from the catalogue, assigned to a workstation, and checked in the form this section prescribes — the
+> authority's own API, so the handshake needs nothing else in the mesh to be right:
+>
+> ```
+> $ curl -sS -o /dev/null -w '%{http_code}' https://<the authority>:9000/health
+> 200
+> subject=CN=Step Online CA
+> issuer=O=Mesh Internal CA, CN=Mesh Internal CA Intermediate CA
+> Verify return code: 0 (ok)
+> ```
+>
+> **Both halves.** Unassigning and pushing removed the anchor, emptied the trust store of the mesh's
+> authority, and returned the plain client to *unable to get local issuer certificate* — then assigning
+> again restored it. The negative half is what distinguishes the anchor working from something else
+> having trusted it, and it is the half nothing had ever exercised.
+>
+> **One thing this found that is not in the module.** The removal only works because the *host* removes
+> the service before the script: stopping the unit is what deletes the certificate and refreshes the
+> bundles, and it needs the script it calls to still exist. Nothing in the module states that ordering;
+> the symmetry this record claims rests on it.
+>
+> Run on the live mesh because that is where a change is verified now
+> ([ADR 0149](0149-the-live-mesh-is-the-test-bed.md)), and the bed still cannot raise a foundation. The
+> evidence is [issue 129](../04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/02-resolution.md).
+> Extended the same day to every converged machine — `novox`, `g14` and `shanks` each hold the anchor
+> and verify with a plain client. `ace` is excluded on purpose: it is adopted, so a module assigned
+> there is held rather than run, which is right and is not trust.
+
+## Consequences
+
+The predecessor's authority can be retired from a machine once this module is assigned to it,
+which is the first time that has been true. `git` over HTTPS to the mesh's forge starts working,
+so the ssh-only clone URL stops being a rule. A module on any machine can call another module's
+internal name and verify it.
+
+What got harder: one more module to assign to a machine that needs it, and the machine's trust
+store now changes when an assignment changes — which is the point, and is also a thing an operator
+can be surprised by. The fetch without prior trust is the same exposure ADR 0098 accepted, now on
+every machine that holds the module rather than only where a proxy runs: anything that can stand
+in the middle of the mesh's own network at the moment of the fetch can be believed. The mesh
+already treats that network as the thing it authenticates.
+
+## References
+
+- [issue 129](../04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md) —
+  the symptom and the evidence.
+- [ADR 0098](0098-a-fact-a-provider-makes-at-first-start-is-fetched-from-it.md) — a fact made at
+  first start is fetched from its provider; this extends it from one program to the machine.
+- [ADR 0082](0082-the-registry-is-reached-by-name-and-trusted-by-the-overlay.md) — the precedent
+  this deliberately does not follow, and why it is right where it is.
+- [ADR 0005](0005-the-node-host.md) — the host is where one operating system's difference lives.
+- [`03-DESIGN/01-to-be/08-connectivity.md`](../03-DESIGN/01-to-be/08-connectivity.md).

02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md

@@ -0,0 +1,176 @@
+---
+topic: the tiers
+status: accepted
+date: 2026-09-30
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0066-public-routing-is-name-agnostic.md
+---
+
+# 148. The mesh's names are resolved, not copied into every container
+
+## Context
+
+The mesh gives every container it declares the whole roster of mesh names as entries written into
+the container's own hosts file at creation
+([design 08 §2](../03-DESIGN/01-to-be/08-connectivity.md),
+[ADR 0066](0066-public-routing-is-name-agnostic.md)). A container takes those entries once and never
+looks again.
+
+Three issues are the same fact arriving three times.
+
+**A container keeps the address it was made with.** Adopting the predecessor's tunnel moved the hub's
+private address; the declaration followed it within one push and nothing on the machine did. The
+forge's container held the old address, lost its database, reported healthy while its existing
+connections lasted, and then the public name went down
+([issue 109](../04-ISSUES/109-a-container-keeps-the-address-it-was-made-with/00-report.md)).
+
+**The same fault, four days later, undetected for five days.** One container had restarted 2286 times
+against a database it could no longer find, while the mesh reported the machine as doing what it was
+told. Inside it, `novox.internal` was an address that had not existed for five days
+([issue 135](../04-ISSUES/135-a-containers-mesh-names-are-not-compared/00-report.md)). Forty-eight
+other containers were current, none of them corrected — each had been recreated for some other
+reason and picked up the roster on the way.
+
+135 was fixed by putting the roster into the digest the host compares a container against, so a
+container whose names moved is recreated like one whose image moved. **That made the roster part of
+every container's identity**, which is the third arrival:
+
+**One name moving replaces every container in the mesh.** Migrating one small module on one machine
+took four routine actions; each changed the roster, and each replaced every container on the control
+node — its own store, the registry, the edge proxy, the forge, the directory, mail. The control plane
+was unreachable twice while its own store came back through crash recovery
+([issue 151](../04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)). None of
+the replaced containers had anything to do with the module being migrated, or with its machine.
+
+The blast radius of a name is now every container that carries the list, which is all of them. The
+node-by-node migration ahead adds names one module at a time — on one machine alone that is around
+twenty-five — and each would be a full restart of every service on the hub.
+
+## Considered Options
+
+**1. Keep the roster in every container and accept the churn.** Rejected. It is not a cost that can
+be paid down: the mesh gets more names as it grows, and every name costs a restart of everything.
+A rollback costs another.
+
+**2. Scope each container's entries to the names it actually binds.** A container is given the names
+of the things it declared a requirement on, so a name's blast radius is its consumers. Tidy, needs no
+new mechanism, and keeps 135's guarantee exactly.
+
+Rejected, and this is the close one. It contradicts the standing intent that **anything on the mesh
+can call anything on it** — three cases, same machine, the private network, the public network, and
+no fourth. Scoping resolution to declared couplings makes a name reachable only where the mesh was
+told in advance that it would be wanted, and a person debugging inside a container would find names
+missing that exist everywhere else on the machine. It also leaves the roster in the digest, so the
+churn returns the moment a widely-bound name moves — smaller, not gone.
+
+**3. Resolve at lookup time through the machine's resolver, and copy nothing.** Chosen.
+
+## Decision
+
+**A container resolves the mesh's names through its machine's resolver, at the moment it asks. No
+mesh name and no mesh address is written into a container, and none is part of a container's
+identity.**
+
+The three consequences that make this worth doing:
+
+- **Staleness stops being possible**, rather than being detected. 109 and 135 are not bugs that were
+  fixed; they are a shape that no longer exists. A name that moves is answered differently by the next
+  lookup, in every container, with nothing recreated and nothing restarted.
+- **A name's blast radius becomes nothing.** Assigning a module on one machine does not touch a
+  container on another.
+- **Anything can still call anything**, which option 2 gave up. The resolver answers every mesh name to
+  every asker on the machine, exactly as it answers the machine itself.
+
+**The resolver is a machine-level process, not a container** — one of the modules that is not a
+container at all — so a container depending on it is not the circularity it would be if the mesh's
+own store had to resolve a name through something the store's own runtime had to start first.
+
+**What a module declares for itself is untouched.** Entries a manifest asks for are the module's own,
+stay in the container, and stay in its identity: they are part of what the module *is*, they do not
+move when the mesh's roster does, and the mesh does not know what they mean.
+
+**The machine's own roster file is untouched.** It is a file, rewritten in place, read by processes and
+people; nothing restarts when it changes. It is only the *copy into each container* that this ends.
+
+### The order this lands in, which is not a preference
+
+**Nothing may stop copying names until resolution works from a container.** Removing the copy first
+reintroduces 109 and 135 — silently, and on a live mesh, which is exactly how both were found.
+
+1. **A container on any network can reach the resolver.** Today a container on the runtime's default
+   network asks from an address the converged filter drops, so it has no DNS at all
+   ([issue 110](../04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md));
+   and on two of four machines the resolver binds loopback only, so the runtime hands containers a
+   public resolver instead. Both are prerequisites, not related work.
+
+   > **Progressive insight — 2026-09-30, later the same day. The loopback claim was wrong.** The
+   > resolver bound the private address on all four machines; on two the runtime had never been told
+   > to use it, and on all four the resolver discarded a query that arrived on the runtime's bridge.
+   > The step stands; the facts under it were those. Both fixed the same day
+   > ([issue 110's resolution](../04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/01-resolution.md)),
+   > and step 3 landed after them.
+2. **The runtime is told which resolver to use, per machine, as a file** — not per container as a
+   creation-time argument, or the resolver's address is back in every container's identity and the
+   problem has only got smaller.
+3. **Then, and only then, the roster leaves the declaration and the digest.**
+
+Until step 3 the mesh keeps copying, and keeps comparing. 151 stays open until step 3 lands; it is not
+closed by this record, only answered by it.
+
+## How this is checked
+
+- **A container resolves a name that moved, without being recreated.** Move a name the mesh serves;
+  from a container that was running before the move and has not been touched since, the name answers
+  with the new address. This is the one 109 and 135 would both have failed.
+- **A name's blast radius is nothing.** Add a routed name on one machine; no container on any other
+  machine is recreated. The apply report on each machine says nothing changed. This is 151.
+- **Anything calls anything.** From a container on any machine, every `<node>.internal` name and every
+  routed name the mesh serves resolves — including names the module never declared a requirement on,
+  which is the guarantee option 2 would have given up.
+- **On every network the runtime offers.** The first three hold for a container on the runtime's
+  default network as well as one on a declared network, because the default network is the case that
+  has no DNS today.
+- **No mesh name is in a container's spec.** A test asserts the digest a host computes for a container
+  does not move when the mesh's roster does, and does move when the module's own declared entries do.
+
+## Consequences
+
+- **The resolver becomes load-bearing for every container**, where before it was load-bearing for the
+  machine. This is a real cost and is accepted: a resolver that is down is a machine that cannot
+  resolve, which is already true of the machine itself, and is a smaller event than a roster change
+  destroying and recreating every container on the machine.
+- **Design 08's "a file rather than a resolver" no longer describes containers.** It was written when
+  the mesh had no resolver and it gave the right answer then. The reasoning it rested on — every Linux
+  has a hosts file, no package needed — was already overtaken by names a hosts file cannot express:
+  service names and wildcards under `<node>.internal`, which is why the resolver was built.
+- **ADR 0066's mesh-wide propagation is kept and its mechanism changes.** A routed name still reaches
+  every asker in the mesh; it reaches them through the resolver rather than by being written into each
+  container. The consequence 0066 records — that an internal issuer's challenge needs the routed name
+  resolvable inside the mesh — holds unchanged and by the same means the machine already uses.
+- **Issue 110 stops being a container-DNS inconvenience and becomes a prerequisite** for the mesh not
+  restarting itself whenever it learns a name.
+- **A container that names a resolver of its own has opted out of the machine's**, and the copy this
+  record removes was the only reason such a container could reach anything by a mesh name.
+
+  > **Progressive insight — 2026-09-30, the afternoon this landed. Found the hard way.** The mail
+  > system's admin, behind Mailu's own resolver, lost its database the moment the copy went
+  > ([issue 171](../04-ISSUES/171-a-modules-own-resolver-knows-no-mesh-name/00-report.md)). A `dns` on
+  > a container is a decision about whether mesh names exist inside it, not a preference; the module
+  > was corrected, and whether the controller should refuse the contradiction is that issue's open
+  > question.
+
+- **A container started by hand gets the mesh's names too**, where before only declared containers did.
+  Design 08 drew that boundary deliberately, on the grounds that reaching into every container is what
+  a nameserver would be for. This record accepts that consequence rather than working around it: a
+  person debugging in a hand-started container resolving the same names as everything else is the
+  behaviour worth having, and it is what "anything can call anything" means.
+
+## References
+
+- [issue 151](../04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md) — one name replaces every container; the question this answers
+- [issue 135](../04-ISSUES/135-a-containers-mesh-names-are-not-compared/00-report.md) — the roster put into the digest
+- [issue 109](../04-ISSUES/109-a-container-keeps-the-address-it-was-made-with/00-report.md) — the first arrival
+- [issue 110](../04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md) — the prerequisite
+- [ADR 0066](0066-public-routing-is-name-agnostic.md) — routed names propagate mesh-wide; extended here
+- [design 08 §2](../03-DESIGN/01-to-be/08-connectivity.md) — the file-not-resolver reasoning this narrows

02-DECISIONS/0149-the-live-mesh-is-the-test-bed.md

@@ -0,0 +1,106 @@
+---
+topic: building it
+status: accepted
+date: 2026-09-30
+deciders: jochen
+reconstructed: false
+supersedes: 02-DECISIONS/0068-the-lab-takes-requests.md
+---
+
+# 149. The live mesh is the test bed
+
+## Context
+
+[ADR 0068](0068-the-lab-takes-requests.md) proposed that the lab accept queued requests
+— a bed and a commit — answer them one at a time from a copy it owns, and expose that through tools so
+an agent could start a run and come back to it. It has been `proposed` since 2026-09-12 and nothing was
+built.
+
+What happened instead is that the mesh became the thing under test. It runs on four machines; every
+fault worth finding in the last month was found on them, and none was found in a bed:
+
+- a container holding an address that had not existed for five days, on the control node
+  ([issue 135](../04-ISSUES/135-a-containers-mesh-names-are-not-compared/00-report.md));
+- a machine reading healthy for eleven hours while no module could reach another
+  ([issue 145](../04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md));
+- one name replacing every container on the hub
+  ([issue 151](../04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md));
+- a consumer assertion that is correct on a mesh being raised and fatal on one that is running
+  ([issue 156](../04-ISSUES/156-moving-a-consumers-delivery-subject-stops-the-control-plane/00-report.md)).
+
+The last is the one that settles it. That change was exercised on the raise path — which is what a bed
+*is* — and the raise path is the only path on which the fault cannot appear. A bed raises a mesh; it
+does not have a mesh that has been running for weeks, with consumers already bound, containers created
+against an older roster, and an adopted machine carrying a predecessor's configuration. **The faults
+that cost the most were all faults of a mesh that already exists**, and a bed is by construction a mesh
+that does not.
+
+Lab runs are also expensive in a way that changed the behaviour around them: each costs a build and
+several minutes, so they were batched, and a batched test is one whose result arrives after the next
+three changes were already written.
+
+## Considered Options
+
+**1. Build 0068 as proposed.** Rejected. It answers a question nobody is asking: the bottleneck was
+never that a person had to sit at the lab, it was that a bed cannot hold the state the faults live in.
+Queueing and tooling a mechanism that finds the wrong class of fault faster is not an improvement.
+
+**2. Leave 0068 `proposed`.** Rejected, and it is why this record exists rather than nothing. A record
+that contradicts current practice and sits unresolved is worse than either answer: it reads as intent
+to anyone who finds it, and the practice it contradicts is written down nowhere but a handoff note.
+
+**3. Record that the live mesh is the test bed, and supersede 0068.** Chosen.
+
+## Decision
+
+**A change is verified against the mesh that is running.** Not because a bed would be unwelcome, but
+because the state that breaks things is state a bed does not have: containers made against an older
+roster, consumers already bound, an adopted machine, a store with weeks of history.
+
+**A change that can only be exercised on the raise path is not verified.** If the only test available
+raises a fresh mesh, the record says so, and says which case was therefore not covered. The words
+"exercised on a fresh mesh" are a statement about coverage, not a pass.
+
+**The lab is not retired**, and [ADR 0016](0016-the-lab.md) stands. It remains the place to raise a
+mesh from bare, which is the one thing the live mesh cannot be asked to do and the one thing a bed does
+better than anything else. What this record removes is the lab as the *default* answer to "is this
+change good", and with it 0068's queue, tools and request protocol.
+
+**Accuracy over a green run.** An honest failure on the live mesh beats a pass in a bed that could not
+have failed — and a change that is risky on the running mesh is a reason to make the change smaller,
+not a reason to test it somewhere it cannot break.
+
+**What 0068 got right is kept as a rule, not a mechanism:** a run reads a copy that is not anybody's
+working tree. Every run of the lab that mattered was pinned to a checkout rather than a worktree, and
+the ones that were not produced results about code nobody had written down.
+
+## How this is checked
+
+- **A record that says a change was verified says on what.** Where it was a fresh mesh, it says which
+  case is uncovered. This is the clause that would have caught 156: its change was verified, honestly,
+  on the only path where it works.
+- **The lab is not in the path of a merge.** No check, playbook or handoff requires a bed to have run.
+- **0068 is unreachable as intent.** Its status is `superseded` and it names this record, so a reader
+  arriving at the queue design finds out immediately that it was not built and why.
+
+## Consequences
+
+- **A fault can be introduced on the machines that serve.** This is the cost, it is real, and it was
+  paid twice in one evening — a control plane crash-looping for half an hour, and every container on the
+  hub recreated five times. Both were found in minutes because they were live, and both would have
+  passed a bed.
+- **There is no pre-merge gate beyond the repositories' own suites.** `make check` and the three hq
+  checks are what stands between a change and the machines, which raises what those suites are worth
+  and makes a test that cannot fail a genuine defect rather than an untidiness.
+- **Raising a mesh from bare is now the lab's whole job**, and is exercised deliberately rather than
+  as a side effect of testing something else. The foundation work
+  ([issue 146](../04-ISSUES/146-the-foundation-cannot-be-raised-on-the-bus-the-mesh-runs-on/00-report.md))
+  is that job, and it is also the proof that the mesh can make another of itself.
+- **An agent cannot hand a run to a queue and come back**, which 0068 would have given. In practice it
+  watches a push and reads the machines, which is what happened anyway.
+
+## References
+
+- [ADR 0068](0068-the-lab-takes-requests.md) — superseded by this
+- [ADR 0016](0016-the-lab.md) — the lab, which stands
+- [issue 156](../04-ISSUES/156-moving-a-consumers-delivery-subject-stops-the-control-plane/00-report.md) — correct on the raise path, fatal on a running mesh

02-DECISIONS/0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md

@@ -0,0 +1,113 @@
+---
+topic: what runs on it
+status: accepted
+date: 2026-09-30
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md
+---
+
+# 150. A module's own code runs as supervised processes under the module's one account
+
+## Context
+
+The repository answers "what runs a module's own code" two ways and reconciles them nowhere
+([issue 117](../04-ISSUES/117-a-modules-own-code-is-a-container-and-a-process/00-report.md)).
+
+[ADR 0047](0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md) is accepted and
+says **a container** — "the tool runtime carrying that module's compiled code" — and "one module, one
+process, one account". Two `proposed` design documents say a **`process`** resource running an argv,
+supervised by the machine, and one of them declares *four* of them for a single module and presents
+four as the point. Neither design document names 0047 in its `decisions:`, and the string `process`
+as a resource type appears in no decision record at all. The thing as built is the container.
+
+Two things have happened since 0047 was written that bear on it directly.
+
+[**ADR 0142**](0142-the-mesh-delivers-its-own-components-as-binaries.md) decided that the mesh's own
+components are binaries on the machine rather than container images, and
+[issue 114](../04-ISSUES/114-should-the-controller-be-a-container-or-a-process/00-report.md) was closed
+by it. That settled the mesh's components and deliberately said nothing about a module's.
+
+And the standing definition of a module hardened: **a module is software that delivers one or more
+services, and a module is not a container.** It may deliver them as a container, an installed package
+with a unit, a binary, or configuration files; 61 of 73 happen to use a container and 11 do not,
+including the resolver, sshd and fail2ban. A rule that a module's *own code* must be a container makes
+the one kind of module the mesh writes itself the only kind that has no choice.
+
+## Considered Options
+
+**1. Hold 0047 as written: a container.** Rejected. Its own reasoning does not require one. What 0047
+argued for was a runtime **per module** rather than one for the whole node, because a node-wide runtime
+could not hold a per-module broker account and per-module runtimes competing on one tool key would each
+be handed calls for tools they do not have. A supervised unit per module satisfies that argument
+exactly — it is per module, and a unit runs as an account. The container was the mechanism to hand, not
+the conclusion.
+
+**2. Let each design document choose.** Rejected; that is the present state and it is what issue 117
+reports. A module author reading the guide writes four processes; a module author reading the record
+writes a container; nothing tells either that the other exists.
+
+**3. Settle the hosting form as a supervised process, and settle the count separately.** Chosen.
+
+## Decision
+
+**A module's own code runs as one or more supervised processes on the machine, under the module's single
+account.** Where [ADR 0047](0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md) says
+"a container, the tool runtime carrying that module's compiled code", read this record. Everything else
+0047 decided stands untouched: a tool is served on its own key, only the module that serves it answers,
+and the module's account is scoped to exactly its tool keys.
+
+**The invariant is the account, not the process count.** 0047's "one module, one process, one account"
+carried its weight in the last clause. Its stated worry about a second process was "not a second one to
+scope and seal" — a second *identity* to grant, seal a secret to, and scope on the bus. Several
+processes sharing the module's one account create no second identity, so nothing further is scoped or
+sealed, and a module may therefore declare as many as its work has shapes: events, tools, a
+provisioner, a scheduled ingest. **What a module may not have is two accounts.**
+
+**A module that delivers its service as a container still does.** This record is about the code the
+module itself carries — its tools, its events, its provisioner — and not about the software it delivers.
+A module wrapping a third-party image wraps a third-party image.
+
+**Why supervised by the machine rather than by the mesh:** it is the same answer ADR 0142 gave for the
+mesh's own components, for the same reason. A unit the machine restarts needs no image, no registry
+pull and no runtime to be up before the mesh's own code can run — which matters most for exactly the
+modules whose code the mesh cannot start any other way.
+
+## How this is checked
+
+- **No design document describes a hosting form for a module's own code without citing this record.**
+  Designs 18 and 20 name it in `decisions:`; this is the gap issue 117's third point reports, and
+  `cycle.py` already enforces that a to-be design names its decisions.
+- **A module declaring several processes resolves to one account.** A test composes a module with more
+  than one process resource and asserts the mesh mints exactly one broker account for it, scoped to that
+  module's tool keys and nothing else — which is 0047's invariant stated as an assertion rather than a
+  sentence.
+- **A module's own code does not require the container runtime.** A machine with no container runtime
+  can still run a module whose code is its own, which is the claim that separates this from option 1 and
+  is checkable on a machine that has one by asserting the declaration names no image for it.
+
+## Consequences
+
+- **The sidecar port stops being needed.** [ADR 0029](0029-a-network-is-a-shape-because-an-action-cannot-be-undone.md)
+  records that "anything that is a service plus a sidecar currently has to publish a port to talk to
+  itself", and the host's `network` shape exists partly for it. A process beside the service on the same
+  machine reaches it without publishing anything, so that pressure goes.
+- **Something must supervise, and it is the machine.** This adds a unit per module's code to what the
+  host writes and owns. The mesh already writes and owns units — `nftables` proves a module can write one
+  and run it — so the mechanism exists; the count grows.
+- **A module's code is delivered, not pulled**, which puts it behind the same gap as the host's own
+  delivery ([ADR 0141](0141-the-host-delivers-its-own-successor.md), not built): nothing yet delivers a
+  version of a module's binary to a machine. A container's code arrives by `docker pull`, and this does
+  not. **This is the cost of the decision and it is not paid**; until delivery exists, a module whose code
+  is its own is a module somebody places by hand.
+- **Issue 117 is answered and its three disagreements close differently:** container-or-unit is decided
+  here; one-process-or-several is decided here as several under one account; and whether the record was
+  consulted is fixed by designs 18 and 20 naming this one.
+
+## References
+
+- [issue 117](../04-ISSUES/117-a-modules-own-code-is-a-container-and-a-process/00-report.md) — the contradiction this answers
+- [ADR 0047](0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md) — extended; its "a container" clause is settled here
+- [ADR 0142](0142-the-mesh-delivers-its-own-components-as-binaries.md) — the same answer for the mesh's own components
+- [ADR 0141](0141-the-host-delivers-its-own-successor.md) — the delivery this depends on and which is not built
+- [ADR 0029](0029-a-network-is-a-shape-because-an-action-cannot-be-undone.md) — the sidecar port this relieves

02-DECISIONS/0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md

@@ -0,0 +1,99 @@
+---
+topic: the tiers
+status: accepted
+date: 2026-09-30
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0066-public-routing-is-name-agnostic.md
+---
+
+# 151. A route's internal name is composed under the node that serves it
+
+## Context
+
+A module that requires a route is given two names from one label: a public one, `<label>.<public
+domain>`, and an internal one, `<label>.<node>.internal`
+([ADR 0138](0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md)). Both were composed
+from the node the module runs on.
+
+The two are answered differently. The public name is published into every machine's roster at the
+address of the node whose proxy serves it ([ADR 0066](0066-public-routing-is-name-agnostic.md)), so
+it reaches the proxy from anywhere in the mesh. The internal name is answered by every machine's
+resolver as *anything under a node's name goes to that node*
+([design 08 §2](../03-DESIGN/01-to-be/08-connectivity.md)) — the node it was composed from, which is
+the consumer's. Where the proxy runs on another machine, that name sends a client to a machine with
+nothing listening, while the public name works
+([issue 139](../04-ISSUES/139-an-internal-route-name-resolves-to-the-consumers-node/00-report.md)).
+Every route on this mesh today is served beside its module, so it has not been seen; `route` is
+provided mesh-wide precisely so that stops being true.
+
+Beside it, the roster gave every routed name a second entry with the mesh's suffix appended —
+`<name>.<public domain>.internal` — because it composed a full name for every entry as it does for a
+machine. That name resolved on every machine, was served by nothing, and was refused by the proxy at
+the handshake; the first three names tried while reproducing an unrelated issue were those, and the
+evidence pointed at a regression that had not happened
+([issue 157](../04-ISSUES/157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md)).
+
+## Considered Options
+
+**1. Keep the consumer's name and publish it at the serving node's address**, as the public name is.
+The name stays `<label>.<consumer>.internal` and an exact roster entry overrides the wildcard.
+Rejected: it makes `<x>.<node>.internal` mean *goes to that node* except when it does not, which is
+the one rule the resolver design states; it needs an entry per route where the wildcard needed none;
+and which of an exact entry and a wildcard a resolver answers first is the resolver's business, which
+the mesh deliberately does not know.
+
+**2. A proxy on every machine, so the serving node is always the consumer's.** Rejected for this
+question: it is a different decision about what `route` is — a node-scoped seat with a mesh-wide
+fallback — and this mesh runs one proxy on the hub today. Whatever is decided there, a route served
+from another machine must have a name that reaches it.
+
+**3. Compose the internal name under the node that serves the route.** Chosen.
+
+## Decision
+
+**A route's internal name is `<label>.<serving node>.internal` — composed under the node whose proxy
+answers the route, which is the machine the request arrives at.** The public name is unchanged:
+`<label>.<public domain>` of the node the module runs on, which is where the operator put it.
+
+Where the proxy runs beside the module — every route on this mesh today — the two nodes are one and
+nothing changes. Where it does not, the name says where the request goes, which is what a name under
+a node's name has always meant.
+
+**A routed name has no mesh form.** The roster publishes it as itself, once, at the serving node's
+address. Only a machine has a bare name beside its full one.
+
+What certifies the internal name is unchanged by this: the proxy that terminates it obtains a
+certificate from the mesh's authority for the names it is given, and it is given this one.
+
+Taken on the operator's standing instruction to answer the open design questions in the work order.
+
+## How this is checked
+
+- **Composition.** A controller test contributes a route from a module on one node to a proxy offered
+  from another, gathered the way the controller gathers a consumer's contribution for a provider on
+  another machine, and asserts the internal name carries the serving node.
+- **Publication.** A controller test renders a roster with a machine and a routed name and asserts
+  the routed name appears as itself, once, and never with the suffix appended.
+- **On the mesh.** After the change no machine's roster carries a `<domain>.internal` entry, and a
+  route's internal name still answers from a container with a certificate from the mesh's authority.
+
+## Consequences
+
+- **A route served from another machine now has a usable internal name.** The first module assigned
+  that way will resolve, where before it would have resolved to the wrong machine with no error.
+- **The internal name of a route can change when its proxy moves.** A route re-homed from one proxy
+  to another gets a new internal name, as the design's rule implies; clients that dialled the old one
+  reach the old machine. The public name does not move with the proxy and is the stable one.
+- **The roster is one line shorter per routed name**, and a person reading a hosts file no longer
+  finds names that resolve to a refusal.
+- **Issue 139's second question — a per-node route holder — is left open**, and is a decision about
+  what a seat is rather than about a name.
+
+## References
+
+- [issue 139](../04-ISSUES/139-an-internal-route-name-resolves-to-the-consumers-node/00-report.md) — the question
+- [issue 157](../04-ISSUES/157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md) — the alias
+- [ADR 0066](0066-public-routing-is-name-agnostic.md) — routed names propagate mesh-wide; extended here
+- [ADR 0138](0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md) — how the two names are composed and how far each reaches
+- [design 08 §2](../03-DESIGN/01-to-be/08-connectivity.md) — anything under a node's name goes to that node

02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md

@@ -0,0 +1,160 @@
+---
+topic: what runs on it
+status: accepted
+date: 2026-09-30
+deciders: jochen
+reconstructed: false
+extends: 02-DECISIONS/0095-the-control-plane-is-the-way-to-ask-a-module.md
+---
+
+# 152. The operator's surface is a module the mesh assigns: the console
+
+## Context
+
+**Since the bus moved, nobody can ask the mesh anything without opening a shell on a machine.** Every
+tool call an operator's assistant makes fails, on every machine including the one the operator sits
+at, with *AMQP not connected*
+([issue 147](../04-ISSUES/147-the-operators-tools-still-dial-the-bus-that-was-removed/00-report.md)).
+The program answering is the predecessor's tool server, started on the workstation by hand, with the
+predecessor's broker address written into the assistant's own configuration. It has no manifest, no
+assignment, no account on the bus, and the mesh has never known it exists. Nothing regressed: the
+mesh removed a transport that a program outside the mesh still dials.
+
+**The mesh has a tool model, and it works.** A module serves each tool on its own subject and its
+account may serve nothing else ([ADR 0047](0047-a-module-runs-its-code-as-its-own-process-with-its-own-account.md));
+a person is issued an account whose only permission is to publish the tool subjects named at issue
+([25 — The bus on NATS](../03-DESIGN/01-to-be/25-the-bus-on-nats.md) §7); a client on the
+runtime repository's main branch speaks that account as a command line and as an MCP server. Measured
+on the live mesh on 2026-09-28: a call to the forge's `gitea_list_repos` answered with real
+repositories; the tool list came back empty, because it asks the catalogue for a tool nothing serves
+([ADR 0132](0132-a-seat-carries-the-tools-its-holder-must-serve.md)).
+
+**Two records have already said where the surface belongs.** ADR 0132's consequences: *the MCP
+surface belongs inside the mesh — a module the mesh assigns to the machine where the agent sits, with
+a credential the mesh minted and authority derived from what it may call, not a program started by
+hand with a credential printed to a terminal.* Design [33](../03-DESIGN/01-to-be/33-the-tools-the-mesh-answers.md)
+§6 says the same. Neither is a decision about the surface: 0132 decided where a seat's tools live,
+and named the surface in passing.
+
+**And one record says the opposite, in the letter.** [ADR 0095](0095-the-control-plane-is-the-way-to-ask-a-module.md)
+made the control plane *the* way to ask a module, deferred "calling is a grant" until something asked
+for it, and recorded that *nothing outside the control plane can*. A person's account (design 25 §7,
+built 2026-09-28) is exactly that grant, minted for a person. So 0095's exclusivity has already been
+widened once without a record saying so; a module that calls tools widens it a second time, and this
+record is where that is said.
+
+**What a module's account may do today, counted from the composition** (`internal/broker`): publish
+its own events, publish the accept subjects of seats it uses, subscribe its own tools and what it
+consumes. No module principal may publish another module's tool subject. Of 72 modules in the
+catalogue, 45 serve tools and 0 may call one.
+
+The question the work order asks before any code: **does the mesh grow its own operator surface, or is
+the surface an ordinary module that happens to serve tools?**
+
+## Considered Options
+
+**1. The control plane serves the agent protocol itself** — a listener on the controller, or a verb
+its binary runs. Rejected. It puts a surface for tools the control plane does not implement on the one
+component that must stay answerable while it is itself being replaced, which is the reason 0132
+rejected the control plane as the answer to discovery. A person on a workstation would reach it over
+the network, and the networked surface [ADR 0035](0035-one-implementation-several-surfaces.md)
+reserves for that authenticates through an OAuth2 provider that is not configured — so the controller's
+`api` verb correctly serves nothing today, and this option would either wait for it or bypass it.
+
+**2. A program a person installs and starts by hand with a printed credential** — what exists on the
+runtime repository's main. Rejected as the end state. It is outside the mesh in every way issue 147
+names: no assignment, no declaration, no seat, no check that it reaches anything, revoked only by a
+person remembering to. It is the predecessor's arrangement one bus later, and it fails the same way
+the next time an address moves. It stays as the recovery path, the way the command line does
+(ADR 0035): a credential from `operator issue` and the `mesh` client work with no console assigned.
+
+**3. An ordinary module, assigned to the machine the person sits at, holding a credential the mesh
+minted, serving the mesh's tools on that machine's loopback.** Chosen.
+
+## Decision
+
+**The console is a module.** `mesh-console` is built by the mesh, registered like any module,
+assigned to a machine, and given a bus credential sealed to that machine. It serves the mesh's tools to
+whoever is on that machine: to an agent over MCP, and to a person through the same endpoint. Assigning
+it to a machine is what makes the mesh reachable from there; unassigning it revokes that, at the next
+composition, with nothing on the machine to remember to remove.
+
+**A grant to call is a manifest word: `invokes`.** A module declares the tools it calls, each as
+`<module>.<tool>`, or the single entry `*` for every tool on the mesh. The bus grants exactly that
+publish side and nothing beside it — no event, no subscription, no seat. This is ADR 0095's deferred
+first option, taken now that a consumer asks for it; a person's account already has this shape, and
+the same composition derives both. The control plane's `ask` stands, and 0095's audit point with it:
+every call still passes one account whose permission list says what it may ask.
+
+**Authority is the machine's login.** The console listens on the machine's loopback only, declared
+`from: machine`, so whoever can open a socket on the machine is whoever owns the machine, and *the
+account that installed the host owns the mesh on that node*
+([ADR 0034](0034-the-local-account-owns-the-mesh.md)). Anything on a machine may call anything on it,
+and that is the whole of local ([ADR 0144](0144-anything-on-a-machine-may-call-anything-on-it.md)).
+The mesh knows no person: what the audit sees is which console asked, under the account
+`<node>.mesh-console`. How a person's identity reaches a session is the question design 15 leaves
+open, and this record does not close it.
+
+**What the console lists is asked of the modules.** Design 33 §5: a module's own tools are answered by
+the module, from the code that defines them. The tool runtime therefore answers one reserved verb for
+every module it serves — `tools`, the module's tool names, descriptions and argument schemas — and the
+console assembles its list by asking the catalogue which modules the mesh holds and each module what
+it answers. A module that is not running is absent from the list and says so; a tool an agent already
+knows the name of can be called whether or not it was listed. A module may not name a tool of its own
+`tools`; the runtime refuses the collision at load rather than letting one shadow the other. A
+role's tools, and the mesh's own verbs, join the list when the `mesh-controller` seat serves them
+(design 33 §1, third family) — the console reads whatever the mesh can say about itself, and grows as
+that does.
+
+**The console holds one credential and one grant: `*`.** It is the operator's surface on a machine the
+operator owns; narrowing what it may call is a setting on its assignment, which
+[ADR 0046](0046-a-module-configuration-is-its-assignments-not-its-manifest.md) already provides for
+and nothing here builds.
+
+## Consequences
+
+- **The way a person drives the mesh is inside the mesh.** It is declared, delivered, replaced and
+  revoked by the same machinery as everything else, and `status` says whether the machine carrying it
+  has applied. Issue 147's shape — a surface kept alive by an address in a file — cannot recur, because
+  there is no file: the console's credential names the bus the mesh is on, and moves when it does.
+- **A module may now call tools, which ADR 0095 had reserved to the control plane.** The grant is
+  explicit, per tool or `*`, and derived by the same composition that grants everything else. A module
+  that declares no `invokes` gains nothing. What got harder: a manifest reviewer has one more field to
+  read for authority, and `*` in it deserves the reader's attention every time.
+- **A new manifest word ships one release before any manifest uses it**, and must reach both parsers:
+  the build machine's and the running controller's. The console's manifest cannot be registered until
+  the controller and the builder that packages it have been rebuilt with the word.
+- **Discovery costs a fan-out per list.** One request per module the mesh holds, answered at once by
+  the bus for every module nothing serves, so the cost is bounded by the modules that are up. The list
+  is cached briefly in the console; a module assigned a moment ago appears at the next refresh.
+- **Every tool runtime must be rebuilt once** to answer `tools`. Until a module is, it is callable and
+  not listed, and the console says which modules did not answer.
+- **The mesh's own verbs are not in the console yet.** `status`, `push`, `assign` are the
+  `mesh-controller` seat's tools under 0132, and the three prerequisites 0132 names are still not in
+  place. A person asking what a node runs still opens a shell for that question, and that gap is design
+  33's to close, not this record's — recorded here so nobody reads the console as the whole of 147.
+- **The person's credential is not retired.** `operator issue` and the `mesh` client remain the path
+  when no console is assigned, and the path an operator uses to bring a mesh up far enough to assign
+  one.
+
+## How this is checked
+
+| Rule | Checked by |
+|---|---|
+| A module's `invokes` becomes exactly that publish grant, and nothing else | the bus user composition test: a module invoking `shop.price` may publish that subject and no other tool's; `*` may publish every tool subject; neither may publish an event or subscribe anything it did not consume |
+| A malformed `invokes` entry is refused at registration | a parser test: an entry naming no tool is a problem named in the manifest's words |
+| The runtime answers `tools` for every module it serves | the runtime's test: a module registering two tools answers three names, and a module naming one of its own `tools` is refused at load |
+| The console's list is what the modules answer | the client's test against a real bus: two modules up, a third the catalogue holds and nothing serves, and the list carries the two and names the third as not answering |
+| A call from the console reaches a module over the bus | the same test, and the live mesh: the console assigned to a workstation answers `tools/list` on its loopback and a call to the forge returns repositories |
+| The console listens on loopback and nowhere else | its manifest declares `from: machine`, and the filter composed for the machine opens nothing for it |
+
+## References
+
+- [issue 147](../04-ISSUES/147-the-operators-tools-still-dial-the-bus-that-was-removed/00-report.md) — the surface outside the mesh
+- [ADR 0095](0095-the-control-plane-is-the-way-to-ask-a-module.md) — extended: a grant to call, for a module as for a person
+- [ADR 0132](0132-a-seat-carries-the-tools-its-holder-must-serve.md) — where a seat's tools live, and the sentence that named the surface
+- [ADR 0035](0035-one-implementation-several-surfaces.md) — three surfaces over one implementation
+- [ADR 0034](0034-the-local-account-owns-the-mesh.md), [ADR 0144](0144-anything-on-a-machine-may-call-anything-on-it.md) — why loopback is the authority boundary
+- [33 — The tools the mesh answers](../03-DESIGN/01-to-be/33-the-tools-the-mesh-answers.md) §5, §6 — discovery, and what serves it to an agent
+- [34 — The console](../03-DESIGN/01-to-be/34-the-console.md) — the design this record authorises
+- mesh-tools `src/client.ts`, `src/mcp.ts`, `src/mesh.ts` — the client this makes a module of

02-DECISIONS/README.md

@@ -48,6 +48,31 @@ form above and dated no earlier than the record's own `date:` — an unmarked ed
 violation the reviewer looks for in the diff, and a marked one is legible in the record itself.
 The git history is the backstop, not the record of intent; the note is the record of intent.
 
+## A pointer back from what a record changes
+
+A new record naming an old one is not enough. **Where a record changes a mechanism an older record
+states — without reversing the decision, so no supersession — the older record gets a dated note
+saying where its mechanism now lives.** A reader arrives at the old record by following a citation,
+and finds text that is still the decision and no longer the method; nothing in it says a later record
+moved the method, and the new record is not in their hands.
+
+> **The mechanism changed — YYYY-MM-DD, by ADR NNNN.** What still stands, what moved,
+> and why.
+
+Three examples of the shape, all found by being missed: ADR 0066 still described a routed name being
+written into every container after 0148 replaced that with resolution; ADR 0047 still said a module's
+code runs in a container after 0150 made it a supervised process; and ADR 0016 still read as though the
+lab were the test bed after 0149 said the live mesh is. Each was a citation leading to the wrong
+answer, in a record that was not wrong about anything it decided.
+
+**This is not machine-checked, and it cannot be from `extends:` alone.** 102 records extend another and
+87 name a parent that does not mention them, which is correct: extending usually means building on a
+context, and a one-directional pointer is the right shape for that. What needs a note is the narrower
+case where the parent's own text has gone stale, and which case that is, is a judgement — so it is a
+rule for the author and the reviewer, and the diff is where it is caught. Making it mechanical would
+mean a record declaring the relationship in its frontmatter, which is a change to the record schema and
+has not been decided.
+
 The records run in the order the decisions were taken, oldest first.
 
 **Every decision is a record.** There is no ledger and no index file — if a decision is worth
@@ -174,6 +199,8 @@ python3 00-META/checks/index.py            fail if stale
 - **0108** — [A route carries the policy applied to a request, and names a secret rather than holding one](0108-a-route-carries-the-policy-applied-to-a-request.md)
 - **0109** — [A package registry seat is one per ecosystem, not one for all of them](0109-a-package-registry-seat-is-one-per-ecosystem.md)
 - **0126** — [A module declares its own seats; the mesh reserves its own](0126-a-module-declares-its-own-seats.md)
+- **0148** — [The mesh's names are resolved, not copied into every container](0148-the-meshs-names-are-resolved-not-copied-into-containers.md)
+- **0151** — [A route's internal name is composed under the node that serves it](0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md)
 
 ### What runs on them, and how it gets there
 
@@ -207,9 +234,9 @@ python3 00-META/checks/index.py            fail if stale
 - **0099** — [A step that runs once names what it reads, and runs again when it changed](0099-a-step-that-runs-once-names-what-it-reads.md)
 - **0110** — [A seat is held by one assignment, from a closed set, and it may deliver a provision](0110-a-seat-is-a-module-assignment-from-a-closed-set.md)
 - **0112** — [A module definition names no node, no mesh and no path: everything it needs is a requirement the mesh resolves](0112-a-module-definition-names-no-node-mesh-or-path.md)
-- **0113** — [The vault makes every shared secret, a provider makes resources and data, and the mesh carries both](0113-the-vault-makes-every-secret.md)  *(proposed)*
+- **0113** — [The vault makes every shared secret, a provider makes resources and data, and the mesh carries both](0113-the-vault-makes-every-secret.md)
-- **0114** — [A credential two parties hold rotates over two credentials; one a single party holds rotates in place, staged; and retiring a credential never removes what it reached](0114-a-shared-credential-rotates-over-two-credentials.md)  *(proposed)*
+- **0114** — [A credential two parties hold rotates over two credentials; one a single party holds rotates in place, staged; and retiring a credential never removes what it reached](0114-a-shared-credential-rotates-over-two-credentials.md)
-- **0115** — [One assignment of a module per node: the module's name is the assignment's identity](0115-one-assignment-of-a-module-per-node.md)  *(proposed)*
+- **0115** — [One assignment of a module per node: the module's name is the assignment's identity](0115-one-assignment-of-a-module-per-node.md)
 - **0117** — [A machine's uplink is a seat: the mesh configures the manager, never the link](0117-a-machines-uplink-is-a-seat.md)
 - **0118** — [Undeclaring removes what the mesh made, and gives a unit back the state it was found in](0118-undeclaring-gives-a-unit-back-the-state-it-was-found-in.md)
 - **0120** — [A roster fact carries its format as a template: the mesh owns the data, the module owns the format](0120-a-roster-fact-carries-its-format-as-a-template.md)
@@ -225,6 +252,11 @@ python3 00-META/checks/index.py            fail if stale
 - **0141** — [The host delivers its own successor, and versions live side by side](0141-the-host-delivers-its-own-successor.md)
 - **0143** — [A consumer verifies the grant it is given](0143-a-consumer-verifies-the-grant-it-is-given.md)  *(superseded)*
 - **0144** — [Anything on a machine may call anything on it, and that is the whole of "local"](0144-anything-on-a-machine-may-call-anything-on-it.md)
+- **0145** — [A module checks what the mesh claims is reachable, and it checks itself](0145-a-module-checks-what-the-mesh-claims-is-reachable.md)  *(superseded)*
+- **0146** — [Connectivity is checked by name, per hosting form, with a valid certificate](0146-connectivity-is-checked-by-name-per-hosting-form.md)
+- **0147** — [A module anchors the mesh's authority on a machine, and takes it away again](0147-a-module-anchors-the-meshs-authority.md)
+- **0150** — [A module's own code runs as supervised processes under the module's one account](0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md)
+- **0152** — [The operator's surface is a module the mesh assigns: the console](0152-the-operators-surface-is-a-module-the-console.md)
 
 ### How it is built
 
@@ -234,9 +266,9 @@ python3 00-META/checks/index.py            fail if stale
 - **0014** — [No workspace — each module is a standalone package consuming published dependencies](0014-no-npm-workspace.md)
 - **0015** — [Applications live in their own repository; the monorepo is for the mesh](0015-applications-live-in-their-own-repository.md)
 - **0016** — [The lab](0016-the-lab.md)
-- **0037** — [Where a module lives](0037-where-a-module-lives.md)  *(proposed)*
+- **0037** — [Where a module lives](0037-where-a-module-lives.md)
 - **0039** — [What the SDK holds, and what it refuses](0039-what-the-sdk-holds-and-refuses.md)
-- **0068** — [The lab takes requests, one at a time, and runs each from its own copy](0068-the-lab-takes-requests.md)  *(proposed)*
+- **0068** — [The lab takes requests, one at a time, and runs each from its own copy](0068-the-lab-takes-requests.md)  *(superseded)*
 - **0069** — [A module is a repository and a path within it](0069-a-module-is-a-repository-and-a-path.md)
 - **0076** — [The SDK is a published package, and the toolchain resolves it by version](0076-the-sdk-is-a-published-package.md)
 - **0082** — [The registry is reached by name, and the overlay is its security](0082-the-registry-is-reached-by-name-and-trusted-by-the-overlay.md)
@@ -245,6 +277,7 @@ python3 00-META/checks/index.py            fail if stale
 - **0097** — [A vendor image is a declared build input, and a recipe fetches nothing undeclared](0097-a-vendor-image-is-a-declared-build-input.md)
 - **0107** — [Persistent data is a directory bind, never a named volume](0107-persistent-data-is-a-directory-bind-never-a-named-volume.md)
 - **0111** — [A build source is on the mesh's git seat, or it is an external repository](0111-a-build-source-is-on-the-git-seat-or-external.md)
+- **0149** — [The live mesh is the test bed](0149-the-live-mesh-is-the-test-bed.md)
 
 ### How it is checked
 

03-DESIGN/01-to-be/08-connectivity.md

@@ -7,8 +7,11 @@ code:
   - mesh-controller internal/identity/authority.go
   - mesh-host internal/identity/serving.go
   - mesh-host internal/apply (the service that reflects a rule set)
-updated: 2026-09-28
+updated: 2026-09-30
 decisions:
+  - 02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md
+  - 02-DECISIONS/0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md
+  - 02-DECISIONS/0147-a-module-anchors-the-meshs-authority.md
   - 02-DECISIONS/0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md
   - 02-DECISIONS/0140-the-filter-constrains-what-arrives-from-outside.md
   - 02-DECISIONS/0104-a-provision-may-be-answered-by-an-adapter-to-the-predecessor.md
@@ -288,6 +291,31 @@ hosts file by the runtime. That extends the file decision rather than overturnin
 mesh and not chosen by a module: a module that listed the machines would go stale the day one
 joins, and a module that did not would be one whose containers cannot reach anything by name.
 
+**Superseded for containers by [ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md)
+(2026-09-30).** Everything above still describes the machine's own roster file, which is how it works
+and how it will keep working. It no longer describes containers.
+
+Copying the roster into each container made the roster part of each container's identity, so one name
+moving replaced every container in the mesh — a module assigned on one machine restarted the store, the
+registry, the edge and mail on another
+([issue 151](../../04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)). And it
+did not stop the staleness it was meant to: a copy taken at creation is stale the moment the roster
+moves, twice found as a container holding an address that had not existed for days
+([issues 109](../../04-ISSUES/109-a-container-keeps-the-address-it-was-made-with/00-report.md)
+and [135](../../04-ISSUES/135-a-containers-mesh-names-are-not-compared/00-report.md)).
+
+**A container resolves the mesh's names through its machine's resolver, at the moment it asks, and
+nothing is copied.** The resolver is a machine-level process rather than a container, so nothing
+circular is being asked for. It was gated on a container being able to reach the resolver from any of
+the runtime's networks
+([issue 110](../../04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md)),
+and landed the day that did, 2026-09-30: the controller writes no mesh name into a container and the
+host's digest carries only what the module declared for itself.
+
+The paragraph below states the old boundary, and 0148 deliberately gives it up: a container somebody
+started by hand resolves the same names as everything else, because the resolver answers the machine,
+not a list of containers.
+
 **The boundary, which is deliberate and worth stating:** *declared* containers. A container
 somebody starts by hand is not the mesh's to configure, and reaching into every container on a
 machine — declared or not — is what a nameserver in `resolv.conf` would be for.
@@ -298,6 +326,14 @@ machine — declared or not — is what a nameserver in `resolv.conf` would be f
 service, the rest is the node — so what resolves is *anything under a node's name*, going to that
 node. What routes it once it arrives is a proxy's, and stays separate.
 
+*2026-09-30.* **So the node in a route's internal name is the one whose proxy answers it**
+([ADR 0151](../../02-DECISIONS/0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md)).
+Composed from the node the module ran on, the name sent a client to a machine with nothing listening
+whenever the proxy ran elsewhere
+([issue 139](../../04-ISSUES/139-an-internal-route-name-resolves-to-the-consumers-node/00-report.md));
+composed from the serving node, the rule above holds without exception. The public name stays the
+module's node's, which is where the operator put it.
+
 **The mesh writes the data and runs no daemon.** One wildcard per machine, from the same set that
 writes the hosts file. A resolver is third-party software and runs *on* the mesh rather than being
 *of* it: the mesh has no business shipping one, choosing which one, or knowing its configuration
@@ -373,7 +409,9 @@ can reach from the outside but cannot resolve from the inside is a name it canno
 authority of its own.
 
 **So a granted route is published into internal resolution as well** — the routed name to the node
-that serves it, mesh-wide, by the same mechanism that writes the node names. It is *given by the
+that serves it, mesh-wide, by the same mechanism that writes the node names — and as itself: a routed
+name has no mesh form, and the suffixed alias the roster once added beside it resolved to a refusal
+([issue 157](../../04-ISSUES/157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md)). It is *given by the
 mesh, not chosen by a module*, for the same reason the node names are: a module listing the routes
 would go stale the day one changes. The mesh propagates the names it was told to serve and still
 knows nothing about what they mean
@@ -710,6 +748,22 @@ step, so when the authority moves the root is fetched again and the proxy is rec
 *How it is checked:* the route-forwarding bed installs the authority, the proxy and a consumer
 from the catalogue and asserts the routed name is served.
 
+**And a machine trusts that authority because a module put its root in its trust store**
+([ADR 0147](../../02-DECISIONS/0147-a-module-anchors-the-meshs-authority.md)). The proxy's fetch
+answers for the proxy and for nothing else: a browser, `git` over HTTPS, a package manager and
+every module calling another by an internal name read the machine's own trust store, and the mesh
+had never written anything there. A module requiring the authority does the whole of it — fetch
+the root over the mesh network, place it where this machine's TLS clients look, refresh the
+extracted bundles — and stopping it, which is what being unassigned does, takes the anchor away
+and refreshes them again. Not the controller's business, because being on the private network is
+what makes the authority *reachable* and is not the same fact as having a reason to *verify* a
+mesh name; and because where anchors live and which command refreshes them is one operating
+system's difference, which is the host's half of the mesh
+([ADR 0005](../../02-DECISIONS/0005-the-node-host.md)).
+*How it is checked:* on a machine holding the module a plain client verifies an internal HTTPS
+name with no bundle argument, and on one without it the same fetch fails to find an issuer — both
+halves, because only the pair tells the anchor apart from something that already trusted it.
+
 ### What was built
 
 *2026-08-31.*

03-DESIGN/01-to-be/12-a-module-repository.md

@@ -7,9 +7,10 @@ code:
   - mesh-controller internal/catalogue/build.go
   - mesh-controller internal/inventory/secrets.go
   - mesh-controller cmd/mesh-builder
-updated: 2026-09-12
+updated: 2026-09-30
 decisions:
   - 02-DECISIONS/0069-a-module-is-a-repository-and-a-path.md
+  - 02-DECISIONS/0037-where-a-module-lives.md
   - 02-DECISIONS/0009-modules-and-the-graph.md
   - 02-DECISIONS/0010-delivery.md
   - 02-DECISIONS/0005-the-node-host.md
@@ -60,6 +61,32 @@ module from a repository and a path, and the root-only reading left every existi
 unbuildable — pointed at the catalogue the builder finds no manifest, pointed at a module's source
 it finds no manifest either.*
 
+## A manifest is checked where it is written
+
+*Added 2026-09-30, from [issue 148](../../04-ISSUES/148-a-manifest-outside-this-catalogue-has-no-check/00-report.md).*
+
+The check the mesh applies at registration — the manifest parses strictly, every name in it is a
+usable one, its routes and events and seats are well formed, and no two manifests given together
+declare one seat — is a verb on the controller's binary, `module check <manifest>…`, and it needs no
+mesh. It reads the files it is given, runs the same functions registration runs, prints every problem
+in the manifest's own words, and exits non-zero if there was one. Somebody describing their own
+application in their own repository — the case [ADR 0037](../../02-DECISIONS/0037-where-a-module-lives.md)
+calls the one that matters most — runs it before pushing, and finds out there rather than when a
+running mesh refuses the registration, or later, when a machine applies something that resolved and
+should not have.
+
+**What it cannot know, it says.** A seat another module declares elsewhere is unknown to a check that
+was not handed that module's manifest, and the output says so rather than refusing: pass the other
+manifest too. The mesh's own seats it knows from the binary, which is the one place that set may be
+read without a store ([ADR 0122](../../02-DECISIONS/0122-a-seat-is-data-a-rename-is-a-database-update.md)
+keeps the store authoritative, so a claim on a mesh seat is judged fully only at registration, and the
+check says that too).
+
+*How it is checked:* the controller's test runs the check over the catalogue checkout beside it and
+over a manifest with a known fault, and asserts the first passes and the second names the fault; the
+test that used to be the only check, `TestEveryCatalogueManifestParses`, now stands beside a command
+anybody can run.
+
 ## The manifest in the repository is not the manifest the mesh holds
 
 A resource names an artifact:

03-DESIGN/01-to-be/18-building-a-module.md

@@ -5,8 +5,9 @@ code:
   - mesh-controller cmd/mesh-builder
   - mesh-controller internal/builder
   - mesh-catalog modules/builder
-updated: 2026-09-29
+updated: 2026-09-30
 decisions:
+  - 02-DECISIONS/0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md
   - 02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md
   - 02-DECISIONS/0111-a-build-source-is-on-the-git-seat-or-external.md
   - 02-DECISIONS/0097-a-vendor-image-is-a-declared-build-input.md

03-DESIGN/01-to-be/20-writing-a-module.md

@@ -5,8 +5,9 @@ code:
   - mesh-catalog modules/showcase
   - mesh-controller internal/builder
   - mesh-sdk src
-updated: 2026-09-21
+updated: 2026-09-30
 decisions:
+  - 02-DECISIONS/0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md
   - 02-DECISIONS/0099-a-step-that-runs-once-names-what-it-reads.md
   - 02-DECISIONS/0053-a-step-that-runs-on-a-schedule.md
   - 02-DECISIONS/0074-the-wire-is-specified-not-the-types.md

03-DESIGN/01-to-be/25-the-bus-on-nats.md

@@ -361,6 +361,13 @@ bridged. It is three things:
 
 Nothing is built of this before §10's bed passes; the MCP surface is a thin adapter over (2).
 
+*Built, and then made a module — 2026-09-30.* (1) and (2) exist: `operator issue` and the `mesh`
+client. What (2) describes as a program on the workstation is now the recovery path; the surface an
+operator uses is a module the mesh assigns to the machine, holding a credential the mesh minted —
+[ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md),
+[34 — The console](34-the-console.md). The tool list it asks for is no longer
+`catalog_tools`, which nothing served: each runtime answers `tools` for its own module.
+
 ## 8. What a module sees, and what the wire does
 
 **The contract a module is written against does not change.** `publish` on an envelope becomes a

03-DESIGN/01-to-be/33-the-tools-the-mesh-answers.md

@@ -102,6 +102,12 @@ being something a person carries and becomes something the mesh runs, on a node,
 An agent's authority can then be role-shaped: *the forge's tools*, rather than a list of
 module-specific names that changes the day the forge is replaced.
 
+*Decided and designed on 2026-09-30:* the module is the console —
+[ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md),
+[34 — The console](34-the-console.md). It builds the second half of §5 now (a module's tools are
+asked of the module, through a `tools` verb every runtime answers) and lists a role's tools when the
+records carry them.
+
 ## 7. Versioning
 
 A seat's tools are an interface and change like one. Additive within a version. A change that would

03-DESIGN/01-to-be/34-the-console.md

@@ -0,0 +1,116 @@
+---
+layer: to-be
+status: in-progress
+code: [mesh-catalog, mesh-tools, mesh-controller]
+updated: 2026-09-30
+decisions:
+  - 02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md
+  - 02-DECISIONS/0132-a-seat-carries-the-tools-its-holder-must-serve.md
+  - 02-DECISIONS/0095-the-control-plane-is-the-way-to-ask-a-module.md
+  - 02-DECISIONS/0035-one-implementation-several-surfaces.md
+  - 02-DECISIONS/0034-the-local-account-owns-the-mesh.md
+---
+
+# 34 — The console
+
+**The mesh's tools, on the machine a person sits at, served by a module the mesh assigned there.**
+An agent reaches them over MCP on the machine's loopback; a person reaches the same endpoint. Nothing is
+installed by hand, nothing is configured with an address, and the mesh knows the surface exists because
+it put it there ([ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md)).
+
+## 1. What it is
+
+A module, `mesh-console`, in the catalogue. Its image is the tool runtime's own — the client that
+already speaks the bus as a command line and as an MCP server — started in a mode that reads the
+module's credential and listens on loopback. It has no state, no provision, no seat. What it needs is
+the bus, which it gets the way every module does: a credential the mesh minted for `<node>.mesh-console`,
+sealed to the machine, delivered as the module's own secret.
+
+Its manifest says three things nothing else in the catalogue says together:
+
+- `invokes: ["*"]` — it calls every tool on the mesh, and the bus grants exactly that publish side;
+- `listens` on a port `from: machine` — the filter opens nothing for it, because loopback is not outside;
+- no `emits`, no `consumes`, no `tools` — it answers nothing on the bus and nobody can address it there.
+
+## 2. What it serves, and to whom
+
+**One endpoint, `POST /mcp` on the machine's loopback**, speaking MCP over HTTP: `initialize`,
+`tools/list`, `tools/call`. An agent on the machine is pointed at it once — the address is the machine's
+own and never changes — and sees every tool the mesh can say it has. A person at a terminal uses the
+same endpoint through the `mesh` client, or through anything that can make an HTTP request; the client
+needs no credential, because the console holds it.
+
+**The endpoint is the machine's login.** It binds `127.0.0.1` and nothing else. Whoever can connect is
+on the machine, and whoever is on the machine is the account that owns the mesh there
+([ADR 0034](../../02-DECISIONS/0034-the-local-account-owns-the-mesh.md),
+[ADR 0144](../../02-DECISIONS/0144-anything-on-a-machine-may-call-anything-on-it.md)). There is no
+token, no login page and no second identity, on purpose: a credential a person had to carry to reach
+their own machine's console would be the arrangement this replaces, moved one hop.
+
+## 3. How it knows what the mesh can do
+
+Design [33](33-the-tools-the-mesh-answers.md) §5 splits discovery in two: a role's tools are read from
+the mesh's records, a module's own are asked of the module. The console builds the second half now and
+reads the first when it exists.
+
+**Every tool runtime answers `tools`.** The runtime that serves a module's tools also serves one verb of
+its own under that module's name, `mesh.mod.<module>.tool.tools`, answering the module's tool names,
+descriptions and argument schemas — the definitions from the code that answers them, and from nowhere
+else. A module may not name a tool of its own `tools`; the runtime refuses the collision at load.
+
+**The console asks the catalogue which modules the mesh holds, then asks each.** `catalog_modules`
+answers the roster; one `tools` request per module, in parallel, answers the list. The bus refuses at
+once a request nothing serves, so a module that is not running costs nothing and is named in the answer
+as not answering, rather than silently absent — *silence and success must never look alike*. The list is
+kept for a short while and refreshed, so an agent asking on every turn does not fan out on every turn.
+
+**A tool that was not listed can still be called.** Listing is discovery; calling is the grant. An agent
+that knows a tool's name asks for it by `<module>.<tool>` and the module answers or the bus says why not.
+
+**What is missing from the list, and until when.** A role's tools and the mesh's own verbs — `status`,
+`push`, `assign` — are the `mesh-controller` seat's under ADR 0132 and are not served yet; their three
+prerequisites are listed in that record. When the seat serves them, the console lists them beside the
+modules' own, and the person stops opening a shell for the mesh's own questions. Until then the console
+says so in its handshake.
+
+## 4. Where it runs
+
+On whichever machines an operator sits at, by assignment. It is not on the control node by default and
+does not need to be: it reaches the bus like any module, from anywhere in the mesh. A machine that is
+not a node cannot have it, which is the right refusal — the mesh reaches what it declares, and a
+workstation that wants the console joins first.
+
+The person's credential and the `mesh` client (design [25](25-the-bus-on-nats.md) §7) remain the path
+for a machine that is not a node, and the path to a mesh not yet far enough along to assign anything.
+
+## 5. Removing it
+
+Unassigning the console from a machine revokes its bus account at the next composition and stops the
+container; nothing is left on the machine that could still connect. An agent pointed at the loopback
+address gets a refused connection, which is the truthful answer.
+
+## How it is checked
+
+| Check | Defends |
+|---|---|
+| a module invoking one tool may publish that subject and no other tool's; `*` may publish every one; neither may publish an event or subscribe what it did not consume | ADR 0152, the grant |
+| a module registering two tools answers three names to `tools`, with schemas; a module naming its own `tools` is refused at load | ADR 0152, discovery |
+| against a real bus: two modules up, a third held and not running — the console lists the two and names the third as not answering | ADR 0152, silence is not success |
+| a call through the console's endpoint reaches a module over the bus and the answer is the module's own, unshaped | ADR 0035, a surface decides nothing |
+| on the live mesh: the console assigned to a workstation answers `tools/list` on loopback and a call to the forge returns repositories | the exit of work-order step 3 |
+| the composed filter for a machine carrying the console opens no port for it | ADR 0144 |
+
+## What this does not settle
+
+- Narrowing a console's grant per assignment. ADR 0046 makes it a setting; nothing reads one yet.
+- The mesh's own verbs on the bus. Design 33's third family; this document only says where they appear
+  once they exist.
+- A person's identity behind the console. The mesh sees the console's account; design 15 keeps the
+  question open.
+
+## References
+
+- [ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md) — the decision
+- [33 — The tools the mesh answers](33-the-tools-the-mesh-answers.md) — what the console lists
+- [25 — The bus on NATS](25-the-bus-on-nats.md) §7 — the person's client this makes a module of
+- [issue 147](../../04-ISSUES/147-the-operators-tools-still-dial-the-bus-that-was-removed/00-report.md) — the symptom

04-ISSUES/006-hq-is-not-indexed-into-the-knowledge-base/00-report.md

@@ -1,7 +1,7 @@
 ---
 status: located
 opened: 2026-08-23
-located-in: [hal, hq]
+located-in: [hq README.md, and the agent ADR 0025 names — not built]
 fixed-by:
 amended-design: 02-DECISIONS/0025-the-design-record-is-read-not-copied.md
 ---
@@ -155,3 +155,30 @@ design document here, and get it back. That check fails today by design.
 
 **What stands until then** is the signpost, and the honest description of it: reachable, not
 surfacing.
+
+## Where this stands, 2026-09-29
+
+*Added in a grooming pass.* The knowledge base this record is about is the **predecessor's**, and it
+is no longer reachable from anything: the surface that answered `recall_search` speaks the transport
+the mesh removed at the cut-over
+([issue 147](../147-the-operators-tools-still-dial-the-bus-that-was-removed/00-report.md)).
+
+So the sentence in `README.md` that this record catches — *these documents are still indexed into
+the knowledge base* — is now wrong twice over: nothing indexed them, and there is nothing to index
+them into. The record stays open, and its answer is no longer "index this repository somewhere"; it
+is whatever the mesh grows as its own knowledge surface, if it grows one. Until then the honest fix
+is the README, which should stop claiming a property nothing provides.
+
+
+## Where this stands, 2026-09-30
+
+The README no longer claims a property nothing provides: it says the indexing never existed, that the
+store it named is unreachable since the cut-over, and that ADR 0025's answer — read, not copied, by an
+agent that consults this repository — is decided and not built. That was the honest fix the previous
+note asked for, and it is done.
+
+The record stays open on ADR 0025's build, and on nothing else. The mesh's own operator surface is now
+a module ([ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md)); the
+reader ADR 0025 describes is the mesh session of design 15, which would answer through that surface
+like any tool. What closes this is still the check 0025 names: search for a phrase that appears only in
+a design document here, and get it back.

04-ISSUES/087-the-controller-cannot-tell-a-host-is-too-old/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-22
-located-in: []
+located-in: [mesh-controller internal/link/protocol.go (the report field that was missing), internal/inventory, cmd/mesh-controller (node show and status)]
-fixed-by:
+fixed-by: mesh-controller 7683ba8, corrected by 1d9c102
 amended-design:
 ---
 

04-ISSUES/087-the-controller-cannot-tell-a-host-is-too-old/01-resolution.md

@@ -0,0 +1,143 @@
+# 087 — resolved: the mesh knows which host runs a machine
+
+*2026-09-30.*
+
+## The field existed and was thrown away on arrival
+
+The machine has reported its host version since
+[ADR 0141](../../02-DECISIONS/0141-the-host-delivers-its-own-successor.md) — `Host` on the report, with
+a comment saying why it must be there: *"without it nothing can say a machine is behind."*
+
+**The controller's own copy of the report did not have the field.** Two structs describe one message,
+one on each side of the wire, and only the sending side had it — so it unmarshalled into nothing and the
+mesh could not answer a question the machine had been answering for a week. That is the whole of this
+issue's mechanism, and it is worth stating plainly because neither side was wrong on its own.
+
+## What it says now
+
+`node show` names it per machine:
+
+```
+  last heard from   here
+  host              2026-09-30-0214
+```
+
+`not reported — this machine has not said since the mesh began keeping it` where the mesh has not been
+told, because a machine that has not said is a different thing from a machine running nothing.
+
+`status` names the machines that are behind another:
+
+```
+1 machine(s) run an older host than another machine does:
+  ace          2026-09-29-0113
+
+  the newest any machine reports is 2026-09-30-0214. A host refuses a declaration carrying a
+  field it does not know, whole — so a new field reaches these machines last
+```
+
+## Disagreement, not staleness, and that is deliberate
+
+The open questions asked whether the controller should refuse to send a declaration a node cannot
+parse. It cannot yet, honestly: **nothing delivers a host version** (ADR 0141 is accepted and not
+built, which is [issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md)),
+so the mesh holds no canonical current version and "behind" has no fixed point to be behind.
+
+What it can say truthfully is that these machines do not all run the same host, and which is newest of
+the ones it has been told about. That is the fact that matters before a declaration gains a field: **the
+oldest host in the mesh is what the mesh may send.**
+
+Two deliberate refusals to guess:
+
+- **A machine that has reported nothing is not called behind.** It may be running anything. `node show`
+  says it has not said, per machine, which is the honest form.
+- **Versions compare as strings.** That suits the timestamps and commits this mesh uses and is wrong
+  for a scheme where `10` sorts before `9`. Said in the code at the place that would have to learn,
+  rather than left as a surprise.
+
+## What I shipped first was wrong, and the mesh said so within the hour
+
+The first version reported *"N machine(s) run an older host than another machine does"* and worked out
+which by comparing versions as strings. **A host reports its version as a commit, and commits have no
+order.**
+
+On the live mesh, with the adopted machine pushed for the first time:
+
+```
+3 machine(s) run an older host than another machine does:
+  g14          04a27ca
+  novox        04a27ca
+  shanks       04a27ca
+```
+
+Those three run the **newer** host — installed 09:18, against the adopted machine's 01:13. `ced54d4`
+sorts above `04a27ca` and that is all it means. An arbitrary lexicographic result, presented as a fact,
+about the one thing this record exists to make trustworthy.
+
+The code carried a caveat saying versions compare as strings and that this "is enough for the timestamps
+and commits this mesh uses". That was the error, written down and not noticed: it is enough for
+timestamps and it is **meaningless** for commits, and the mesh reports commits.
+
+It now reports the split and claims no ordering:
+
+```
+4 machine(s) do not all run the same host:
+  04a27ca      g14, novox, shanks
+  ced54d4      ace
+
+  a host refuses a declaration carrying a field it does not know, whole — so the mesh may
+  send only what every one of these understands. Which of them is newer is not readable
+  from a commit; that needs a version the host reports as ordered
+```
+
+More useful as well as more honest — the reader sees who is on which side of the split, which is what
+decides whether a field can be sent — and it leaves the ordering where it belongs: with the host, which
+would have to report something ordered for anybody to have it.
+
+**This is [issue 145](../145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md)
+arriving by my own door**, an hour after closing it: a report that confidently says the opposite of the
+truth is worse than one that says less, because it trains a reader to distrust the whole surface.
+
+## Measured on the mesh, and one limitation it exposed
+
+All four machines run the identical host binary — same digest, installed within eighteen seconds of each
+other — and at first only one reported its version. The other three said *not reported*, which read as a
+difference between machines where there was none.
+
+**A machine states its host version only when the mesh sends it a declaration.** The report is published
+after an apply; the periodic reconcile that runs every five minutes publishes nothing, because it is the
+machine keeping itself as declared rather than answering anything. So a machine that is current and idle
+never says, and the mesh cannot distinguish that from a machine running something ancient.
+
+Confirmed by pushing: before, `not reported`; after, `04a27ca` — the same version the machine that had
+been pushed already reported.
+
+```
+novox      host   04a27ca
+shanks     host   04a27ca
+g14        host   04a27ca
+ace        host   not reported — this machine has not said since the mesh began keeping it
+```
+
+`ace` has not been pushed since the field existed; it is adopted and parked.
+
+**This is enough for the purpose and not enough for the claim.** For deciding whether a new declaration
+field is safe it is sufficient, because pushing is what the mesh is about to do anyway and the answer
+arrives with the act. For *knowing what the mesh is running*, it is not: a long-idle machine's entry is
+as old as its last push, and the honest reading of `not reported` is "nobody has asked recently" rather
+than "this machine is silent". The words say the first, which is why they are those words.
+
+Making a heartbeat carry it would close the gap and is a change to what a heartbeat is — a bare word
+that the node is there, deliberately carrying nothing else. Left alone rather than widened in passing.
+
+## The open questions, answered as far as they can be
+
+- *Should a node report the version of its host?* It already did. The gap was the reading.
+- *Should the mesh refuse to send a field no node understands yet, or refuse per node and say so?*
+  Neither, yet — refusing needs the mesh to know which fields need which version, which is the third
+  question below and is not answered here. What it does is make the disagreement visible before
+  somebody adds a field.
+- *Is there a general shape — a declaration saying which version of the host it needs?* Still open, and
+  now cheaper to answer: the versions are recorded, so a minimum-version field on a declaration has
+  something to compare against. It belongs with
+  [issue 107](../107-a-declaration-carries-no-order/00-report.md), which wants to add a field and is the
+  first thing this makes safe.

04-ISSUES/088-the-forges-own-address-names-a-port-it-may-not-have/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-22
-located-in: []
+located-in: [mesh-catalog modules/gitea, mesh-controller internal/catalogue/declaration.go]
-fixed-by:
+fixed-by: mesh-controller 7352c84, merged in #46 — a module is told its port in a container's environment too, as a file already was
 amended-design:
 ---
 
@@ -39,3 +39,9 @@ the assignment happens to differ.
 - Should composition refuse an environment value that names a port the module does not fix, the
   way it refuses other claims a module cannot make?
 - Which other modules write their own address, with a port, into their environment?
+
+## Closed
+
+*2026-09-29, in a grooming pass rather than by whoever fixed it.* The forge's address follows a moved port the same way every other reader does. Found by
+reading what the code repositories' commits cite: the fix names this issue and is on `main`. It was
+not re-verified on a machine, and this record says so rather than implying a run that did not happen.

04-ISSUES/089-a-contributed-route-names-a-port-the-node-may-have-moved/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-22
-located-in: []
+located-in: [mesh-controller internal/catalogue/declaration.go, mesh-catalog (every routed module)]
-fixed-by:
+fixed-by: mesh-controller bdf965d (a route names the endpoint it serves) with `portOfEndpoint` and `AtPublishedPort` — the contribution carries the endpoint's declared port and the machine-side redirection is applied to it like any other
 amended-design:
 ---
 
@@ -45,3 +45,9 @@ precisely because the predecessor holds the usual one.
   keeping the mapping out of rendered configuration?
 - What should refuse a declaration whose contributed route names a port nothing on that node
   listens on?
+
+## Closed
+
+*2026-09-29, in a grooming pass rather than by whoever fixed it.* A route names an endpoint rather than a port, and the redirection that turns a declared port into the published one is applied to contributions too. Found by
+reading what the code repositories' commits cite: the fix names this issue and is on `main`. It was
+not re-verified on a machine, and this record says so rather than implying a run that did not happen.

04-ISSUES/107-a-declaration-carries-no-order/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-23
 located-in: [mesh-controller internal/link, mesh-host internal/link]
-fixed-by:
+fixed-by: mesh-host PR 59 (the host refuses an older sequence and drains by it), mesh-controller PR 160 (each send is numbered under the node's hold) — measured 2026-09-30, 02-resolution.md
 amended-design:
 ---
 
@@ -40,3 +40,16 @@ exists there and is thrown away at the wire.
   separate genesis-digest branch is needed on the host?
 - Is a sequence enough, or does a mode change deserve its own marker, so a replayed converged
   declaration is refused by mode as well as by order?
+
+## What has since made this safer to do (2026-09-30)
+
+Adding a `sequence` to a declaration is adding a field, and a host refuses a declaration carrying a
+field it does not know — whole. That was
+[issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md), and it is resolved: the
+mesh now records which host each machine reports and `status` names every machine running an older one
+than another does.
+
+So the flag day is visible before it is walked into, which it was not when this was filed. It does not
+make the field free: **the oldest host in the mesh is still what the mesh may send**, and one machine of
+four is behind today. A sequence that an old host refuses takes that machine out of the mesh's reach
+entirely — worse than the replay it prevents, which has never been observed.

04-ISSUES/107-a-declaration-carries-no-order/01-diagnosis.md

@@ -0,0 +1,76 @@
+# 107 — diagnosis: the fix is a flag day, and it should wait for delivery
+
+*2026-09-30. Read, measured, and not built — deliberately.*
+
+## The premise is confirmed
+
+A host parses a declaration with unknown fields refused, and the code says why rather than leaving it
+to be inferred:
+
+> `DisallowUnknownFields` is the whole point rather than strictness for its own sake: a field the host
+> does not know is a thing the control plane believes it asked for.
+
+So adding `sequence` and `supersedes` is not an additive change. **Any host that has not been upgraded
+refuses the whole declaration and applies nothing** — which is exactly the behaviour that keeps a
+half-understood declaration off a machine, and exactly what makes this expensive.
+
+## What has changed since this was filed
+
+[Issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md) is resolved: the mesh now
+records the host version each machine reports and `status` names every machine running an older host
+than another does. The flag day is visible before it is walked into, which it was not on 2026-09-23.
+
+That makes the cost measurable rather than hypothetical, and the measurement is the reason this is not
+being built today.
+
+## Why it waits
+
+**One machine of four runs an older host, and it cannot be upgraded.** `ace` is adopted, deliberately
+parked until the network and module-assignment work is settled, and **nothing delivers a host version at
+all** — [ADR 0141](../../02-DECISIONS/0141-the-host-delivers-its-own-successor.md) is accepted and not
+built, which is [issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md).
+Every machine takes a hand-placed binary.
+
+So shipping the field means, in order: place a host by hand on three machines, unpark the fourth, place
+it there too, and only then turn the controller half on. A machine missed in that sequence is a machine
+the mesh cannot send anything to at all — not degraded, unreachable.
+
+**And the fault it prevents has never been observed.** The record says so itself: *"Not observed;
+constructed from the code, and narrow."* It needs a backlog of more than sixteen declarations queued
+across a `converge`/`adopt` pair, or a broker slow enough to split one, and the host already applies the
+newest of a drained batch and refuses a declaration that is not the last by digest.
+
+**Trading a machine's reachability for a replay nobody has seen is the wrong way round.** The right
+order is [issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md) first —
+when the mesh can deliver a host, a declaration field costs a rollout instead of an expedition — and the
+work order already puts that in its last group, as the proof that the mesh can make another of itself.
+
+## What the open questions look like now
+
+- *A per-node `sequence` under the controller's node hold, and `supersedes` as the previous digest?*
+  Still the right shape. The controller already holds the lock and already records each send, so the
+  order exists and is thrown away at the wire — unchanged since this was filed.
+- *Genesis signing its bundle as sequence zero?* Yes, and it is the cheaper half: the bundle is written
+  by the host that will read it, so it has no flag day of its own.
+- *Is a sequence enough, or does a mode change deserve its own marker?* A sequence alone does not stop
+  a replayed *converged* declaration reaching a node that has since been returned to adopted, which is
+  the incident of issue 104 by another door and is what this record names as its real risk. It wants
+  both, and the second is the one worth having first.
+- **And one this record did not ask:** should a declaration say which host version it needs? 087 makes
+  that comparable for the first time, and it is the general form of the answer — a field that announces
+  its own requirement, rather than a flag day per field, for ever.
+
+## Status
+
+Left `located`. The owner is unchanged, the shape of the fix is agreed, and the gate is
+[issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md) rather than anything
+in this record. **This is a judgement about order, not a refusal** — it is cheap to overrule, and the
+code is a day's work once a host can be delivered.
+
+## The gate has opened (2026-09-30, evening)
+
+The mesh delivers the host now — built by its own toolchain, published to its own registry, delivered
+over the bus and started by the launcher, on all four machines
+([issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/01-progress.md)). A declaration
+field is a build and a push, not an expedition. The order this record asked for — hosts first, then
+the controller — is now two commands and a status line that says when the first has finished.

04-ISSUES/107-a-declaration-carries-no-order/02-resolution.md

@@ -0,0 +1,60 @@
+# 107 — resolved: a declaration carries its order
+
+*2026-09-30. Measured on the mesh.*
+
+## What was done
+
+**Hosts first, then the controller** — the order [issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md)
+says a new declaration field needs, and now a build and a push rather than an expedition
+([issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/01-progress.md)).
+
+The host understands a `sequence` on a declaration and tolerates its absence: absent reads as "no
+order claimed", not "first", so a controller that sends none is still understood and a host that kept
+a declaration before it understood the field compares nothing. It refuses a declaration with a lower
+sequence than the one it kept, whole, and says why; and the drain that picks one declaration from a
+batch keeps the highest sequence rather than the last to arrive — which is the case the report
+constructed, a backlog drained out of order.
+
+The controller numbers each send: the next number for that node, taken under the node's hold, before
+the body exists, so the number is inside what the mesh signs and a replayed older declaration cannot
+borrow a newer one's.
+
+## Measured
+
+```
+push shanks; push shanks
+sequence in kept declaration: 2
+node      sequence
+novox     2
+shanks    2
+ace       (none — not sent since numbering)
+g14       (none)
+status:   nobody "not running what the mesh would send them"
+```
+
+Both applies went through; neither was refused; the machine holding the earlier one accepted the later.
+
+## The subtlety, which would have read every machine as behind for ever
+
+The mesh decides a machine is behind by comparing the digest of what it **would** send against what it
+**did** send. A number changes the bytes. So the read-only comparison composes with the number the
+machine was *last* sent — not a fresh one — and is byte for byte what was sent when nothing else
+changed. Without that, numbering would have made `status` name all four machines as out of date on
+every reading, permanently.
+
+## The open questions
+
+- *A per-node `sequence` under the controller's node hold?* Yes, as described. **`supersedes` — the
+  previous digest — is not added.** A strictly-greater sequence gives the ordering; a chain of digests
+  would give continuity, which nothing here needs yet and which every re-composition would break.
+- *Genesis signing its bundle as sequence zero?* Zero is "no order claimed", which is what the bundle
+  carries by carrying nothing. Same rule, no genesis branch.
+- *A marker for a mode change?* Not needed for the incident it guards: a replayed converged declaration
+  reaching a node returned to adopted is already refused **by mode**, before this check runs.
+
+## How it is checked
+
+Host: an older sequence is refused, a newer or equal one is not, and no order claimed on either side
+compares nothing; the drain keeps the highest sequence, and falls back to arrival when none is claimed.
+Controller: a send carries its number inside the signed bytes, an unnumbered send is byte for byte what
+it was before, and each node's counter is one higher per send and readable for the comparison.

04-ISSUES/109-a-container-keeps-the-address-it-was-made-with/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-24
-located-in: [mesh-host internal/apply]
+located-in: [mesh-controller internal/catalogue/declaration.go (every container was given the roster at creation)]
-fixed-by:
+fixed-by: mesh-controller PR 161 — no container is given a mesh name; it resolves through its machine's resolver (ADR 0148, landed 2026-09-30 once issue 110 did)
 amended-design:
 ---
 
@@ -59,3 +59,27 @@ container is made with are the same kind of input, read once at creation, and ar
   and is the stronger statement; it is also what the mesh's own resolver exists for.
 - Either way: what tells an operator that a container is running with an address the node no longer
   has? Nothing did.
+
+## Answered at the cause (2026-09-30)
+
+This was the first of three arrivals of one fact: a container is given the mesh's names when it is
+created and never looks again, so a name that moves afterwards is wrong inside it for as long as it
+runs. It arrived again as [issue 135](../135-a-containers-mesh-names-are-not-compared/00-report.md),
+whose fix made the names comparable — and that fix made the roster part of every container's identity,
+which arrived as [issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md).
+
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) ends the
+copying: a container resolves through its machine's resolver at the moment it asks. The shape this
+record reports then has nowhere to occur. It is gated on
+[issue 110](../110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md), so
+until that lands the mesh still copies and still compares.
+
+## Resolved (2026-09-30)
+
+110 landed the same day ([its resolution](../110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/01-resolution.md)),
+and mesh-controller PR 161 then removed the copy: no container is given a mesh name or a mesh address,
+and a module's own declared entries are the only `host` lines it carries. Verified on the control-node
+after its containers were recreated once — the last time a name will do that: the forge's container
+carries no extra hosts and resolves another machine and a routed name through the machine's resolver,
+so the shape this record describes has nowhere to occur. Checked in the controller's tests: a
+container's declaration is byte-for-byte the same under a roster of one machine and a roster of three.

04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md

@@ -1,8 +1,9 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-24
-located-in: []
+located-in:
-fixed-by:
+  - mesh-catalog modules/dnsmasq (the runtime was never told; the resolver answered by interface)
+fixed-by: mesh-catalog PR 175 (the runtime is reloaded and keeps its containers over a restart) and PR 176 (the resolver answers by address, so a query from a bridge is admitted) — measured 2026-09-30, 01-resolution.md
 amended-design:
 ---
 
@@ -51,3 +52,22 @@ knows that is what the rule means.
   the runtime's default one? That is a stronger rule and would have prevented 109 as well.
 - What checks it? A converged bed with a container on the default network resolving a mesh name is
   the missing assertion; nothing in the resolver's own beds covers the filter.
+
+## What now depends on this (2026-09-30)
+
+This stopped being a container-DNS inconvenience.
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) decides
+that a container resolves the mesh's names rather than being given a copy of them, which is what stops
+one name moving from replacing every container in the mesh
+([issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)) and what makes a
+stale address impossible rather than merely noticed
+([issues 109](../109-a-container-keeps-the-address-it-was-made-with/00-report.md)
+and [135](../135-a-containers-mesh-names-are-not-compared/00-report.md)).
+
+**That decision cannot land until this one does**, and not partly: a container on the runtime's default
+network is the case with no DNS at all, and it is the case the mesh's own forge runs in. Two of four
+machines also bind the resolver to loopback only, so the runtime hands their containers a public
+resolver. Both halves are this issue.
+
+*Later the same day: the second half was wrong, and the first had a different cause than the one above.
+[01-resolution.md](01-resolution.md) has what was actually found.*

04-ISSUES/110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/01-resolution.md

@@ -0,0 +1,64 @@
+# 110 — resolved: a container on any network reaches the resolver, and is answered
+
+*2026-09-30. Measured on the three converged machines; the adopted one holds its resolver module until it
+is taken and is not covered.*
+
+## What was actually wrong
+
+Not what the report predicted. The report named the filter: a container on the runtime's default
+network asks from a bridge address, and the converged filter admitted queries by source address only.
+That was true when it was written and was fixed before this issue was ever tested — the filter admits
+by the link a packet arrives on ([ADR 0144](../../02-DECISIONS/0144-anything-on-a-machine-may-call-anything-on-it.md)),
+and a container's bridge is admitted whole. Tested on every machine: the query arrives, the filter
+passes it.
+
+Three other things were wrong, each hiding the next.
+
+**The runtime had never been told.** The resolver module writes the runtime's `dns` key into the
+runtime's own configuration file. The runtime reads that key when it starts and not on a reload, and on
+two machines the runtime predated the file — so every container they started got a public resolver, and
+`novox.internal` came back as not existing. Nothing reported this: the file was present and current,
+the resolver ran, and a name not existing is a valid answer. Fixed in mesh-catalog PR 175: the module
+also sets `live-restore` and reloads the runtime when its file changes, so the one restart the `dns` key
+needs no longer stops every container. The restart is then the operator's, once per machine; done on
+both today, with every running container kept.
+
+**The resolver dropped the query.** With the runtime corrected, a container's query reached the resolver
+— and got no answer, on every machine, including the one whose runtime had been right all along. The
+socket was bound to the private address; the filter admitted the packet; dnsmasq received it and
+discarded it without a line of log. Its configuration said `interface=mesh0`, and dnsmasq admits a
+query by the interface it arrives on when told an interface: a container's query is addressed to the
+private address but arrives on the runtime's bridge, and the bridge is not `mesh0`. Fixed in mesh-catalog
+PR 176: the resolver is told the address to answer on, not the interface that carries it, and a query to
+that address is admitted whatever bridge brings it. The bridges are the runtime's to name.
+
+**The report's second half was wrong.** "Two of four machines bind the resolver to loopback only" was
+an inference from the containers' behaviour, and the behaviour had the cause above. The resolver bound
+the private address on all four; nothing had asked it there.
+
+## What is verified
+
+From a container on the runtime's default network, started by hand and given nothing, on each of the
+three converged machines: `novox.internal` answers with the hub's private address, through the machine's
+own resolver. That is the fourth check of
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) — "on every
+network the runtime offers" — and its first step; the record's step 2 (the runtime told per machine, as
+a file) was already how the module works. Step 3 may now begin.
+
+## What checks it
+
+By hand, today. Nothing in the mesh asserts that a container can resolve a mesh name: the resolver's
+own tests cover what it answers, not who can ask. The check that would have caught all three faults is
+the one the report asked for and 0148 lists — a container on the default network resolving a mesh name
+— and it is not built. It belongs with the reachability check of
+[issue 145](../145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md),
+which is parked; until then this is a thing a person verifies after touching the resolver, the filter,
+or the runtime's configuration.
+
+## What this cost to find
+
+The three faults produced one symptom — a container that cannot resolve — and each fix revealed the
+next. The first was found by reading the runtime's own view of its configuration rather than the file;
+the second by capturing the query on the bridge and finding it arrive and go unanswered; the third only
+by admitting the first belief was wrong. A machine that had been believed to work all day had never
+worked either.

04-ISSUES/114-should-the-controller-be-a-container-or-a-process/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-24
 located-in: [mesh-controller module.json, mesh-host internal/apply]
-fixed-by:
+fixed-by: ADR 0142 — the mesh's own components are delivered as binaries on the machine, so the controller is a process; the delivery itself is issue 142
 amended-design:
 ---
 
@@ -84,3 +84,21 @@ restart and run-to-completion semantics — so this would not need host-side wor
 
 The two do not collapse into one. The controller is not a code-carrying sidecar, and `network: host`
 is what makes the asymmetry visible here and nowhere else.
+
+## Answered
+
+*2026-09-29, in a grooming pass.* This asked a question rather than reporting a defect, and the
+question was taken: [ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md)
+decides that the mesh's own components — the host, the controller, the catalogue, the builder, the
+vault — are **binaries on the machine**, delivered by the mechanism
+[ADR 0141](../../02-DECISIONS/0141-the-host-delivers-its-own-successor.md) built, and that
+third-party software (the store, the registry, the broker) stays a container because an image is the
+right way to carry somebody else's build.
+
+So the operating experience this record was written from — every mutating command reached through
+`docker exec mesh-controller` — is answered, and answered against the container.
+
+**The delivery is a separate matter and is not this record's.** Step 1 of it is built and no
+component travels yet; that is
+[issue 142](../142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md).
+

04-ISSUES/117-a-modules-own-code-is-a-container-and-a-process/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-25
 located-in: [hq, mesh-catalog modules/showcase, mesh-sdk src/tools/index.ts, mesh-tools]
-fixed-by:
+fixed-by: hq 83791f0 (PR 196) — ADR 0150: a module's own code runs as supervised processes under the module's one account; designs 18 and 20 now cite it, and ADR 0047 carries a dated note pointing at it
 amended-design:
 ---
 
@@ -99,3 +99,27 @@ what the sidecar is has nowhere in the design layer to look, which is how this w
   is mechanically checkable: the resource types a design doc names are a closed set, and every
   member of it either appears in a decision or does not. Whether that check is worth writing is
   part of this issue, not settled by it.
+
+## Answered (2026-09-30)
+
+[ADR 0150](../../02-DECISIONS/0150-a-modules-own-code-runs-as-supervised-processes-under-one-account.md)
+settles all three disagreements, and the design documents win two of them:
+
+1. **Container or unit — a supervised process.** ADR 0047's argument never required a container. It
+   argued for a runtime *per module*, because a node-wide one could not hold a per-module account and
+   per-module runtimes on one tool key would be handed calls for tools they do not have. A unit per
+   module satisfies that exactly, and a unit runs as an account. The container was the mechanism to
+   hand. [ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md) had
+   already gone the same way for the mesh's own components, and a module is not a container.
+2. **One process or several — several, under one account.** 0047's "one module, one process, one
+   account" carried its weight in the last clause; its stated worry was "not a second one to scope and
+   seal", which is about a second *identity*. Processes sharing the module's one account create none.
+   What a module may not have is two accounts.
+3. **Whether the record was consulted — fixed rather than answered.** Designs 18 and 20 now name 0150
+   in `decisions:`, and 0047 carries a dated note saying where its hosting form was settled, so neither
+   door leads to the wrong answer any more.
+
+**What this does not fix.** A module's code delivered as a binary is behind the same gap as the host's
+own ([ADR 0141](../../02-DECISIONS/0141-the-host-delivers-its-own-successor.md), accepted and not
+built): a container's code arrives by `docker pull` and this does not, so until delivery exists such a
+module is one somebody places by hand. 0150 records that as the cost of the decision, unpaid.

04-ISSUES/118-umamis-store-answers-the-dial-and-times-out-the-query/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-25
-located-in: [mesh-catalog modules/umami]
+located-in: [mesh-host internal/apply (the spec comparison, via issue 135) — not mesh-catalog modules/umami, which this record first named and which was never at fault]
-fixed-by:
+fixed-by: mesh-host e82789a (issue 135's fix) — recreating the container with a current roster ended it; verified live 2026-09-30
 amended-design:
 ---
 

04-ISSUES/118-umamis-store-answers-the-dial-and-times-out-the-query/01-resolution.md

@@ -0,0 +1,79 @@
+# 118 — resolved: it was issue 135, and it is over
+
+*Verified on the machine, 2026-09-30.*
+
+## It no longer happens
+
+```
+$ docker inspect -f '{{.RestartCount}}' umami
+0
+$ docker logs umami --tail 12
+26 migrations found in prisma/migrations
+No pending migrations to apply.
+✓ Database is up to date.
+▲ Next.js 16.3.4
+✓ Ready in 0ms
+$ curl -o /dev/null -w '%{http_code}' https://umami.novox.be/
+200
+```
+
+No restart loop, no `prisma.$queryRaw()` timeout, and the public name that had answered `502` since
+2026-09-25 answers `200`. The raw query that could not complete now runs twenty-six migrations and
+reports the store up to date.
+
+## What it was
+
+**The same fault as [issue 135](../135-a-containers-mesh-names-are-not-compared/00-report.md), which
+was diagnosed three days later without either record noticing the other.** 135's container is this
+one, named in its own evidence table:
+
+```
+umami          created 2026-09-23   novox.internal:10.42.0.1
+mesh-catalog   created today        novox.internal:10.10.0.1
+```
+
+The mesh's overlay range had moved. Umami had been created before the move and held the store's name
+at an address that no longer existed, while every container made after the move held the current one.
+That is why the dial appeared to succeed and the first real query timed out, and why the same query
+from the same network with the same credential answered in milliseconds — **what differed was the
+name, not the path, not the credential and not the store.**
+
+Today it holds `novox.internal:10.10.0.1`.
+
+## Why this record did not find it
+
+The report's own reasoning is worth keeping as a warning, because it is careful and it is wrong:
+
+> A dial that succeeds and a query that times out, from a container on one network to a store on
+> another, has the shape of a path-MTU or conntrack fault (large response packets dropped after the
+> small handshake ones pass), or of the store accepting the TCP connection while the backend it
+> proxies for is wedged.
+
+Both are good hypotheses about a network path. Neither is the answer, and the record also names the
+move that would have found it — *"what is known to differ for umami against every working consumer of
+the same store tonight is nothing yet — that comparison is the first move"* — and then did not make it.
+Comparing umami's hosts entries against any container created that week would have shown a five-day-old
+address in one field.
+
+**A stale name presents as a network fault.** That is the lesson, and it is the reason
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) stops
+copying names into containers at all: not because detecting staleness is hard, but because it disguises
+itself as something else for five days while every check reports success.
+
+## What actually ended it
+
+Issue 135's fix — `mesh-host` e82789a, *a container's mesh names are part of what it is* — put the
+roster into the spec digest the host compares, so a container whose names moved is recreated like one
+whose image moved. That recreated umami with a current roster and ended this.
+
+That fix has since been superseded in turn, by 0148, because making the roster part of every
+container's identity meant one name moving replaced every container in the mesh
+([issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)). So this record
+closes on a fix that is itself on the way out — which does not make it less closed, and is worth saying
+plainly rather than leaving a reader to find out.
+
+## Not carried forward
+
+The `502` had one other contributor worth recording as ruled out: a stale duplicate Traefik router for
+this name, hand-authored in the predecessor's era beside the mesh-written one, removed on 2026-09-25
+and — as the report says — changing nothing. It was not the cause and it is gone.

04-ISSUES/120-a-provisioner-remembers-what-it-did-not-what-is/00-report.md

@@ -1,8 +1,8 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-26
 located-in: [mesh-sdk src/provisioner, mesh-catalog modules/redis]
-fixed-by:
+fixed-by: mesh-catalog bbda88c, merged in #84 — every credential provider says whether it still holds a consumer
 amended-design:
 ---
 
@@ -60,3 +60,9 @@ checks it after the first pass.
   instance and leaves the gap for the others.
 - Where does the record of what was applied live, if not in memory? ADR 0114, still
   proposed, puts rotation state with the vault. The same place may answer this.
+
+## Closed
+
+*2026-09-29, in a grooming pass rather than by whoever fixed it.* A provisioner asks the backend what is there rather than trusting what it remembers doing. Found by
+reading what the code repositories' commits cite: the fix names this issue and is on `main`. It was
+not re-verified on a machine, and this record says so rather than implying a run that did not happen.

04-ISSUES/125-a-hold-is-not-a-line-in-the-apply-report/00-report.md

@@ -1,7 +1,9 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-26
-located-in: [mesh-host internal/apply, mesh-controller]
+located-in: [mesh-host internal/link (the apply report line), mesh-controller cmd/mesh-controller (status and its all-well condition)]
+fixed-by: mesh-host cbf5018, mesh-controller bfd983e
+amended-design:
 ---
 
 # 125 — a hold is not a line in the apply report, and an operator flew blind into an outage

04-ISSUES/125-a-hold-is-not-a-line-in-the-apply-report/01-resolution.md

@@ -0,0 +1,70 @@
+# 125 — resolved: a hold is a line in the report, and it stops the mesh reading as well
+
+*2026-09-30.*
+
+## What the four surfaces say now
+
+The report named four surfaces, none of which carried the one sentence that mattered. Two of them
+already did by the time this was picked up, and two did not.
+
+**1. The apply report says what it held** — this was missing, and is the line the operator was reading
+when the count did not add up:
+
+```
+applied 330 resource(s), 16 held until their module is taken (route-proxy: 13, mailu: 3)
+```
+
+Grouped by module and ordered by name, because `take` acts on a module and that is the sentence an
+operator needs. An apply that held nothing says nothing extra — a line reporting `0 held` on every
+converged apply is one that stops being read.
+
+**2. `status` counts holds, and a hold breaks "all well"** — this was missing. Status now says:
+
+```
+15 resource(s) are held as found, because their module was assigned and never taken — so it is
+running none of what it declares:
+  novox        route-proxy (13), mailu (2)
+
+  `take <node> <module>` compares what runs against what it declares, and runs it
+```
+
+**And the sentence that was the fault no longer prints.** "all doing what they were told, all heard
+from, running what the mesh would send them" was *true* for the whole outage, and acting on it stopped
+the predecessor's proxy. A hold now suppresses it; being adopted still does not, and the difference is
+deliberate — adopted is a mode somebody chose and can leave alone, a module assigned and never taken is
+a half-finished action with nothing left to finish it.
+
+**3. `node show <node>` shows the node's own held list** — already true, and recorded here as checked
+rather than assumed. It reads `held` from what the machine last reported, with an `as of` beside it, and
+names each hold's kind, target, id, module, whether something other than the mesh has changed it, and
+where an original was kept.
+
+**4. The push's count** is unchanged and now interpretable, which was the ask: `sent 346` against
+`applied 330, 16 held until their module is taken (…)` is a pair a reader can resolve without opening a
+file on the machine.
+
+## Where the data came from
+
+**The host already reported it.** `Held` has been on the wire since [ADR 0100](../../02-DECISIONS/0100-a-node-in-use-is-adopted-before-it-is-converged.md),
+and the controller already recorded it and showed it in `node show`. Nothing needed a new field, a new
+message or a migration — which is why this is additive, and why the report's framing (*"the semantics
+are consistent and right; the reporting is what let them be forgotten"*) was exactly right.
+
+What was missing was that two surfaces never asked. Status read the mesh's take-time listing, so a
+module assigned after that listing showed nothing at all; the apply line counted what it applied and
+said nothing about the difference.
+
+## One thing deliberately not done
+
+**Status does not call a hold a fault.** It is correct behaviour, and a reader trained to see red for
+something the mesh did right will stop reading. It is reported as work outstanding, with the command
+that finishes it — and it withholds the all-well sentence, which is the part that carries the weight.
+
+## How it is checked
+
+- The apply line names the count and the module, and is empty when nothing is held (mesh-host).
+- Status finds a hold from what the machine reported, end to end through the store.
+- **A held module makes the all-well condition false**, asserted against the production condition
+  rather than a copy of it — that condition is now one named function for this reason.
+- The JSON form carries a row per machine and module, and omits the field entirely when nothing is
+  held.

04-ISSUES/128-the-hosts-file-is-written-whole/00-report.md

@@ -1,5 +1,6 @@
 ---
-status: located
+status: resolved
+fixed-by: mesh-host 1cb8953 and fdc768c — the mesh writes into a marked block of a text file instead of over it
 opened: 2026-09-26
 located-in: [mesh-controller internal/catalogue/facts.go, mesh-host internal/apply]
 ---
@@ -71,3 +72,9 @@ private network loses that name too.
 - The host's file resource supports `into: "json"` only; anything else is a whole write.
 - `node show <node>` on the adopted workstation: `holds file /etc/hosts
   mesh-wireguard.fact-node-names`, original kept.
+
+## Closed
+
+*2026-09-29, in a grooming pass rather than by whoever fixed it.* A shared hosts file keeps every line that is not the mesh's. Found by
+reading what the code repositories' commits cite: the fix names this issue and is on `main`. It was
+not re-verified on a machine, and this record says so rather than implying a run that did not happen.

04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md

@@ -1,7 +1,8 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-26
-located-in: [mesh-controller, mesh-catalog step-ca]
+located-in: [mesh-catalog modules/ca-trust]
+fixed-by: the ca-trust module registered from the catalogue and assigned to a workstation; verified in both directions 2026-09-30 (02-resolution.md)
 ---
 
 # 129 — nothing makes a machine trust the mesh's own certificate authority

04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/01-diagnosis.md

@@ -0,0 +1,82 @@
+# 129 — diagnosis
+
+*2026-09-30, from the workstation the issue was opened on.*
+
+## Still live, and reproduced exactly
+
+The certificate is genuine, the authority is the mesh's, and nothing on the machine trusts it:
+
+```
+$ openssl s_client -connect keycloak.novox.internal:443 -servername keycloak.novox.internal
+subject=CN=keycloak.novox.internal
+issuer=O=Mesh Internal CA, CN=Mesh Internal CA Intermediate CA
+Verify return code: 20 (unable to get local issuer certificate)
+
+$ curl https://keycloak.novox.internal/
+curl: (60) SSL certificate OpenSSL verify result: unable to get local issuer certificate (20)
+```
+
+`trust list` holds no entry for the mesh. The anchors present are two `mkcert` development roots and
+the predecessor's lab root — the report's account of the trust store is unchanged.
+
+The public name on the same proxy verifies cleanly (`CN=keycloak.novox.be`, Let's Encrypt, return code
+0), which places the fault exactly where the report puts it: not in the proxy, not in the authority,
+and not in the certificate.
+
+**The name matters, and the report's "every HTTPS name the mesh serves internally" is too broad.** The
+served internal name is `<label>.<node>.internal`. The hosts file also carries
+`<label>.<public-domain>.internal`, which nothing serves and which fails differently — that is
+[issue 157](../157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md), found while
+reproducing this, and it cost the first several minutes of this diagnosis.
+
+## The authority serves what the module needs
+
+`step-ca` is up and healthy, and publishes `roots: /roots.pem` for both `acme-ca` and
+`internal-acme-ca`. That endpoint returns PEM:
+
+```
+$ curl -sk https://127.0.0.1:9000/roots.pem
+-----BEGIN CERTIFICATE-----
+MIIBvzCCAWWgAwIBAgIQYa2CkdJk16JyG/dVy2qoEzAKBggqhkjOPQQDAjA+…
+```
+
+So `${bound:internal-acme-ca:roots}` in the `ca-trust` module composes to a URL that returns a
+certificate, and the module's own check — refuse a body that is not one — is checking the right thing.
+
+Worth recording because it was nearly filed as a defect: step-ca *also* serves `/roots`, which returns
+`{"crts":["-----BEGIN CERTIFICATE-----\n…"]}`. That body contains the literal text the module greps
+for, so had the module used `/roots` it would have installed JSON into the anchors directory and
+reported success. It does not use it. The guard is sound only because the published path is the PEM
+one, which is worth knowing before anybody changes either.
+
+## What is actually in the way
+
+**The module is not registered.** The report and the work plan both say it exists and is merged, which
+it does — `mesh-catalog modules/ca-trust`, on `main`. But the mesh has never been told about it:
+
+```
+$ mesh-controller module list | grep -iE 'ca-trust|step-ca'
+step-ca            1        built 67f5f4cf          on novox
+```
+
+39 of the catalogue's 76 manifests are registered. `ca-trust` is one of the 37 that are not, so it
+cannot be assigned to anything — "assign it to one machine" has no module to name.
+
+A dry run confirms it registers cleanly and needs no artifact built: it declares a directory, a script,
+a unit and a service, and no image.
+
+```
+$ mesh-controller build <catalogue> --path modules/ca-trust --dry-run
+… the manifest, parsed and validated
+```
+
+## So the remaining work is three steps, not one
+
+1. **Register it** — build it from the catalogue, which pins nothing because it has no artifacts.
+2. **Assign it** to a machine. The workstation this was observed on is the honest first choice: it is
+   where a person meets the fault, and it is where the check can be made with a plain client.
+3. **Verify** `curl https://<label>.<node>.internal/` with no flags, and `trust list` naming the mesh.
+
+Then removal, which the module declares and nothing has exercised: unassigning must take the anchor
+away and refresh the bundles ([ADR 0147](../../02-DECISIONS/0147-a-module-anchors-the-meshs-authority.md)),
+and that is the half most likely to be wrong, because it is the half nobody reaches by accident.

04-ISSUES/129-nothing-makes-a-machine-trust-the-meshs-authority/02-resolution.md

@@ -0,0 +1,104 @@
+# 129 — resolved: the workstation trusts the mesh, and stops when told to
+
+*Done and measured on the machine, 2026-09-30.*
+
+## What was done
+
+Three steps, not the one the plan expected — the module was merged and had never been registered
+(see [the diagnosis](01-diagnosis.md)):
+
+1. **Registered** `ca-trust` from the catalogue. No artifact to build: it declares a directory, a
+   script, a unit and a service, and no image.
+2. **Assigned** it to the workstation the issue was opened on, and pushed.
+3. **Verified** with a plain client, then **unassigned and pushed again** to exercise removal, then
+   assigned and pushed once more.
+
+The host's own account of arriving:
+
+```
+created ca-trust.state  (/var/lib/ca-trust)
+created ca-trust.anchor (/var/lib/ca-trust/anchor)
+created ca-trust.unit   (/etc/systemd/system/mesh-ca-trust.service)
+updated ca-trust.trust  (mesh-ca-trust.service): boot disabled to enabled, stopped to running
+```
+
+## It works, by the check the report asked for
+
+The report's own reproduction, with no flags and nothing installed by hand:
+
+```
+$ curl -sS -o /dev/null -w '%{http_code}' https://git.novox.internal/
+200
+$ openssl s_client -connect git.novox.internal:443 -servername git.novox.internal
+issuer=O=Mesh Internal CA, CN=Mesh Internal CA Intermediate CA
+Verify return code: 0 (ok)
+$ trust list | grep -A2 Mesh
+    label: Mesh Internal CA Root CA
+    trust: anchor
+    category: authority
+```
+
+Four internal names, all verifying: `git` 200, `umami` 200, `keycloak` 302, `drive` 302. Before this,
+every one of them was `curl: (60) … unable to get local issuer certificate (20)`.
+
+**And the consequence the report named specifically**: git over HTTPS to the mesh's forge, which it said
+had forced the working clone URL to be ssh-only.
+
+```
+$ git ls-remote https://git.novox.internal/novox/hq.git HEAD
+76fbe323ea3401fcdadbf500c61bd3fa5a0a8603	HEAD
+```
+
+## Removal is symmetric, which nothing had ever shown
+
+[ADR 0147](../../02-DECISIONS/0147-a-module-anchors-the-meshs-authority.md) says the module anchors the
+authority *and takes it away again*. That half had never run. Unassigning and pushing:
+
+```
+removed ca-trust.trust (mesh-ca-trust.service)
+removed ca-trust.unit  (/etc/systemd/system/mesh-ca-trust.service)
+removed ca-trust.anchor (/var/lib/ca-trust/anchor)
+removed ca-trust.state  (/var/lib/ca-trust)
+```
+
+Then: the anchor file gone, `trust list` naming no authority of the mesh's, and the plain client back to
+`unable to get local issuer certificate (20)`. A machine that leaves the mesh stops trusting it, as the
+record claims.
+
+**The order is what makes it work, and is worth saying.** The service is removed *first*, so systemd
+runs the unit's `ExecStop` — which is what deletes the certificate and refreshes the bundles — while the
+script it calls still exists. Had the script or the state directory gone first, stopping the unit would
+have had nothing to run, and the anchor would have been left behind with nothing declaring it. Nothing
+in the module says this; it is the host's removal order that makes the module's symmetry real.
+
+## What this leaves
+
+- **Three machines of four** *(extended the same day, after the first was proven)*. `ca-trust` is now
+  assigned to every converged machine, and each verifies with a plain client:
+
+  ```
+  novox    mesh CA in trust store: 1    https://git.<node>.internal/ -> 200
+  g14      mesh CA in trust store: 1    https://git.<node>.internal/ -> 200
+  shanks   mesh CA in trust store: 1    https://git.<node>.internal/ -> 200
+  ```
+
+  Before, each of the two that had not been assigned it answered
+  `curl: (60) … unable to get local issuer certificate (20)` and held no entry for the mesh.
+
+  **`ace` is deliberately not among them.** It is the adopted machine, still carrying the
+  predecessor's resolver and filter, and it is not converged until the network and module-assignment
+  work is settled. Assigning a module to it would hold rather than run
+  ([ADR 0100](../../02-DECISIONS/0100-a-node-in-use-is-adopted-before-it-is-converged.md)), which is
+  correct and is not the same as trusting anything.
+
+- **It arrives per assignment, which is a shape worth questioning.** The report's reasoning — that
+  being on the private network is what makes a machine one that speaks to the mesh's names — argues
+  for a fact carried to every machine on the network, the way the roster and the registry trust are.
+  ADR 0147 chose a module assigned per machine, and this record does not reopen it; the cost is that
+  a machine joining the mesh trusts nothing until somebody remembers a second command.
+- **`service-manager` reports `degraded` on this machine** and the module ran anyway. Worth knowing that
+  the capability gate passes on a degraded service manager, since a module whose whole delivery is a
+  unit is the kind that would be worst served by one.
+- **The predecessor's authority is still in the trust store**, beside the mesh's now rather than instead
+  of it. The report notes that it cannot be retired while anything on the machine speaks TLS to a mesh
+  name; that is no longer true here, and retiring it is its own piece of work.

04-ISSUES/130-undeclaring-a-service-stops-it/00-report.md

@@ -1,5 +1,6 @@
 ---
-status: located
+status: resolved
+fixed-by: mesh-host 3112c88 — undeclaring gives a unit back the state it was found in, and removes only a process the mesh made
 opened: 2026-09-27
 located-in: [mesh-host internal/apply/apply.go (remove)]
 amended-design: 02-DECISIONS/0118-undeclaring-gives-a-unit-back-the-state-it-was-found-in.md
@@ -15,7 +16,7 @@ found that the host's `remove` path stops every `service` resource that is no lo
 delete". `store.Orphans` matches by id alone. So any of these stops the unit:
 
 - the module is unassigned — by mistake, or to switch it for another;
-- the node is sent a deliberately-empty declaration ([issue 127](../127-a-declaration-that-shrinks-to-empty-is-skipped-not-sent/00-report.md));
+- the node is sent a deliberately-empty declaration ([issue 149](../149-a-declaration-that-shrinks-to-empty-is-skipped-not-sent/00-report.md));
 - a later catalogue version renames the resource's `id`.
 
 That is right for a service the mesh brought into being. It is wrong for a unit the mesh
@@ -64,3 +65,9 @@ something to settle in passing.
 
 The unassign preview is partly answered — the host's plan names each unit it will stop — and the
 controller's side is left open.
+
+## Closed
+
+*2026-09-29, in a grooming pass rather than by whoever fixed it.* Undeclaring no longer stops a unit the mesh only reloaded or only kept running. Found by
+reading what the code repositories' commits cite: the fix names this issue and is on `main`. It was
+not re-verified on a machine, and this record says so rather than implying a run that did not happen.

04-ISSUES/135-a-containers-mesh-names-are-not-compared/00-report.md

@@ -68,3 +68,31 @@ container runtime's shape, not a choice; the answer is to recreate, which is wha
 that would rather re-read a roster from a file can already ask for one as a fact
 ([ADR 0120](../../02-DECISIONS/0120-a-roster-fact-carries-its-format-as-a-template.md)) and restart on
 it.
+
+## The container was umami, and it had its own record (2026-09-30)
+
+The container in the table above is umami, and its symptom had already been filed three days earlier as
+[issue 118](../118-umamis-store-answers-the-dial-and-times-out-the-query/00-report.md) — the store
+answering the dial and timing out the query, a restart loop, and a public name answering `502`. Neither
+record noticed the other: 118 was reasoning about path-MTU and conntrack on the network between the
+container and the store, which is what a stale name looks like from inside the container.
+
+118 is resolved by this record's fix and says so. Noted here because a symptom filed twice, diagnosed
+once, and closed in one place is how a repository comes to disagree with itself.
+
+## What replaced this fix (2026-09-30)
+
+The fix here — putting the mesh's names into the digest the host compares, so a container whose names
+moved is recreated like one whose image moved — worked, and cost more than it was worth. It made the
+roster part of every container's identity, so one name moving replaced every container in the mesh: a
+module assigned on one machine restarted the store, the registry, the edge and mail on another
+([issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)).
+
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) goes at
+the cause this record only described: **a copy taken at creation is stale the moment the roster moves,
+and detecting that is not as good as not copying.** A container resolves the mesh's names through its
+machine's resolver, at the moment it asks, so the fault this record reports cannot occur rather than
+being noticed a restart later.
+
+Recorded here because this is where somebody arrives to find out why the digest carries names, and the
+answer is that it did, for two days short of a month, and stopped.

04-ISSUES/139-an-internal-route-name-resolves-to-the-consumers-node/00-report.md

@@ -1,9 +1,9 @@
 ---
-status: open
+status: resolved
 opened: 2026-09-28
-located-in: [mesh-controller internal/catalogue]
+located-in: [mesh-controller internal/catalogue/declaration.go (composeName took the consumer's own name as the internal domain)]
-fixed-by:
+fixed-by: mesh-controller PR 163 — the internal name composes under the node whose proxy serves the route (ADR 0151, 2026-09-30)
-amended-design:
+amended-design: 03-DESIGN/01-to-be/08-connectivity.md
 ---
 
 # 139 — An internal route name resolves to the consumer's node, not the one that serves it
@@ -49,3 +49,15 @@ resolve whether or not anything answers.
   and if so, is `route` still one mesh-wide provision or a node-scoped seat with a mesh-wide fallback?
 - What certifies the name in either case? The certificate is obtained by whoever terminates TLS, and
   that is the question above in another form.
+
+## Answered (2026-09-30)
+
+[ADR 0151](../../02-DECISIONS/0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md):
+the internal name is composed under the node that serves the route — the machine the request arrives
+at — because `<x>.<node>.internal` means *goes to that node* and nothing else. The public name stays
+the consumer node's, which is where the operator put it. The first question is answered that way; the
+second, a per-node route holder, is a decision about seats and is left where it is; the third is
+unchanged, since the proxy that terminates the name is given it and certifies it.
+
+mesh-controller PR 163 carries it. On this mesh every route is served beside its module, so no name
+changed; the controller's tests hold the case where it would.

04-ISSUES/140-an-endpoints-reach-is-not-declared/00-report.md

@@ -1,5 +1,5 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-28
 located-in:
   - mesh-controller internal/catalogue/manifest.go
@@ -7,7 +7,7 @@ located-in:
   - mesh-controller internal/catalogue/declaration.go
   - mesh-controller examples/route-proxy
   - mesh-catalog (every routed module manifest)
-fixed-by:
+fixed-by: mesh-controller bdf965d (a module names its endpoints) and c68d3a7 (an assignment configures an endpoint as one thing) — the filter, the proxy's names and both authorities now read one statement
 amended-design: 03-DESIGN/01-to-be/08-connectivity.md
 ---
 

04-ISSUES/140-an-endpoints-reach-is-not-declared/01-resolution.md

@@ -0,0 +1,47 @@
+# Resolution
+
+*2026-09-29.*
+
+**Built, and this record did not say so.** The issue was written on 2026-09-28 and answered the same
+week by two commits in `mesh-controller`; nothing came back to close it, so the mesh's own account of
+itself said for a day that reach was declared nowhere while the code read it in three places.
+
+- `bdf965d` — *a module names its endpoints, and a route names the one it serves*. `listens[].name`
+  is the endpoint; a route contribution names the endpoint rather than repeating a port.
+- `c68d3a7` — *an assignment configures an endpoint as one thing*. The `endpoints` settings key, per
+  node, by endpoint name: `{"endpoints": {"ssh": {"port": 20134, "reach": "public"}}}` — port, label
+  and reach in one block, which is what [ADR 0138](../../02-DECISIONS/0138-an-assignment-binds-an-endpoint-and-says-how-far-it-reaches.md)
+  asked for and what [ADR 0046](../../02-DECISIONS/0046-a-module-configuration-is-its-assignments-not-its-manifest.md)
+  said configuration is.
+
+## The three readers, which is what the issue was about
+
+The complaint was that the per-node source override had exactly one caller. It now has three, and
+they are the three mechanisms reach was decided to settle at once:
+
+| reader | what it does with it |
+|---|---|
+| the filter | `Reaches` turns each endpoint's reach into the rule for its machine port |
+| the proxy's names | `composeName` composes the public name, the internal name, or both — and a name nobody asked for is not composed |
+| the authorities | the proxy certifies only names it was actually given, each from its own authority, through two host policies rather than one |
+
+**A routed endpoint keeps the manifest's port**, which is ADR 0138's own insight and older than it
+([ADR 0045](../../02-DECISIONS/0045-a-machine-firewall-is-the-sum-of-what-it-listens-on.md)): the
+proxy is how it is reached, so `public` there asks for a public *name*, not an open port.
+
+**An endpoint that is not routed is reached and never named.** Git over ssh is that case — the one
+the issue said the model could not express — and it is now the ordinary one.
+
+## How it is checked
+
+`internal/catalogue/endpoints_setting_test.go`: a block says port, label and reach; a block may say
+only a reach; a name the module does not declare is refused; a reach outside the four values is
+refused; and saying the same thing twice — once in the block, once through the older per-port keys —
+is refused rather than resolved by whichever is read last. The proxy's half is `policy_test.go` and
+`authority_test.go`: a name the mesh did not send is not certified, by either authority.
+
+## What is left, and it is not this
+
+The older keys (`ports`, `expose`, and reach keyed by port) still work beside the block. They are
+what the block replaces, and retiring them is its own small change — not a gap in what reach can
+say.

04-ISSUES/142-the-host-is-the-one-thing-the-mesh-does-not-deliver/00-report.md

@@ -74,3 +74,18 @@ every future change of this shape, and it was paid today.
   argues for the former.
 - Does the same gap apply to the launcher and the units beside the binary, which are also files no
   declaration names?
+
+## What now waits on this (2026-09-30)
+
+[Issue 107](../107-a-declaration-carries-no-order/00-report.md) — a declaration carries no order, so a
+host cannot tell an older one from a newer. Its fix adds a field to the declaration, and a host refuses a
+declaration carrying a field it does not know, **whole**. So it is a flag day: every host upgraded, then
+the controller.
+
+With nothing delivering a host, that means placing a binary by hand on every machine and unparking the
+adopted one, and a machine missed in the sequence is a machine the mesh cannot send anything to at all.
+107 is held for that reason rather than for anything in its own diagnosis.
+
+**This is what makes a declaration field cost a rollout instead of an expedition**, which is a use for
+this record beyond keeping machines current: it is the thing standing between the mesh and its own
+protocol evolving.

04-ISSUES/142-the-host-is-the-one-thing-the-mesh-does-not-deliver/01-progress.md

@@ -0,0 +1,185 @@
+# 142 — the mesh can now build and publish its own host
+
+*2026-09-30. Both of ADR 0141's named gaps are closed; the host is not yet placed by the mesh.*
+
+## What ADR 0141's insight named, and what each cost
+
+> The remaining work is a way to build the host and a way to name a version in a path, and until both
+> exist nothing delivers a version and every machine takes the fallback.
+
+**1. Nothing could compile it.** The toolchain list was a closed set of typescript and python, and its
+own warning — every language is another implementation of the contracts modules share, so adding one
+commits to keeping N implementations in step — does not attach to Go. Go is how the host, the control
+plane and the builder are written, and none of them is a module in that sense: the host is what
+*applies* modules.
+
+Closed by a Go toolchain naming a new base module, `mesh-tools-go`. Named and not pinned, so the mesh
+answers with the copy it holds and moving compiler is a build rather than an edit to the control
+plane's source ([ADR 0044](../../02-DECISIONS/0044-a-public-name-is-provisioned-like-any-capability.md),
+[0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md)).
+
+**2. A version could not reach the path.** An archive named a fixed path and nothing interpolated the
+build into it, so nothing could ask for `…/versions/<version>/`.
+
+Closed by `${version}` in any value of a resource that uses an archive or a bundle. **The version is
+the artifact's digest, short, and not the commit**: two builds of one commit are meant to be the same
+bytes — the toolchains are `-trimpath` for that — so a content-addressed version means an unchanged
+build resolves to the path it already had, where a commit-named path would move for an identical binary
+and recreate everything reading it.
+
+## Three more things were in the way, and none was in the record
+
+Found by doing it, in the order they appeared:
+
+- **`sourcesFor` turned every entrypoint into a `.ts` file.** One language's file extension, written
+  into the code that serves every language. The extension is the toolchain's now.
+- **The output directory was left to the compiler.** `tsc --outDir` makes one; `go build -o` writes into
+  a directory and does not create it, failing with a message about a path rather than about a build.
+  Made for every toolchain, because which compilers are forgiving is not something a reader should have
+  to know.
+- **A bundle was refused if it named what it is built from.** The reason — a bundle is the module's own
+  directory compiled whole — holds for an interpreted language and cannot hold for a compiled one: a Go
+  repository carries several commands, the host and its bootstrap among them, and "the module's own
+  directory" is then not a package at all. A compiled bundle may now say which package; the refusal
+  stands for every interpreted one.
+
+And a fourth, in the base module itself: its first Dockerfile ran `apt-get`, and the golang image the
+mesh holds is Alpine. That is
+[issue 136](../136-a-module-may-name-a-program-the-machine-does-not-have/00-report.md) in an image
+rather than on a machine, and the build refused rather than a module failing later — which is the
+behaviour that issue wants.
+
+## Measured: the mesh compiles its own host and publishes it to its own registry
+
+```
+$ mesh-controller build <the host's repository>
+  host-arch    bundle artifact-store://mesh-host/host-arch/blobs/sha256:ad62528c…
+
+mesh-host 1, built on novox from b5196e97
+```
+
+Fetched from the mesh's registry and opened:
+
+```
+mesh-host: ELF 64-bit LSB executable, x86-64, statically linked, stripped
+$ ./mesh-host help
+mesh-host — the node host
+```
+
+Statically linked matters: what a machine holds is a file rather than a container, so a binary needing
+a libc it did not bring is a delivery that works until a machine differs.
+
+**The host is a module now**, with one bundle for `arch` built from `cmd/mesh-host`. A system is
+required for a compiled artifact because a binary is pinned at link time so a host refuses to touch a
+machine it was not built for ([ADR 0005](../../02-DECISIONS/0005-the-node-host.md)); `arch` is what all
+four of this mesh's machines report themselves to be, and another system is another artifact and
+another build.
+
+## What is left
+
+**The host module declares no resources**, so nothing places the built bundle on a machine yet. That is
+the next piece and it is the one with the interesting question in it: the resource is an archive
+unpacked to `…/versions/${version}/`, applied by the host that is running, and the bootstrap is not
+circular because the two are different versions ([ADR 0141](../../02-DECISIONS/0141-the-host-delivers-its-own-successor.md)).
+The host half of that mechanism — versions side by side, the newest runs, the running one stands aside
+between reconciles, rollback picks a directory — is built and tested and has never had a version to
+work on.
+
+**One thing worth noticing while writing it.** The systems list exists because "the difference between
+two of them is a C library, not a kernel" — and a static Go binary has no C library. So one build would
+in fact run on all three. The pin is then a policy (a host refuses a machine it was not built for)
+rather than a necessity, which is a reasonable thing to keep and is worth knowing is a choice.
+
+**And a sharper one, asked as a question and worth its own record.** The artifact's system is validated
+and then read by nothing: it does not reach the compiler, no machine is matched against it, and nothing
+chooses between two artifacts by it. The host built here is x86-64 because the build machine is, not
+because anything in the declaration said so — correct for this mesh by coincidence. That is
+[issue 159](../159-an-artifacts-system-is-checked-and-then-ignored/00-report.md).
+
+## Delivered, started, and one fact short (2026-09-30, later)
+
+The loop closed. The launcher is delivered as a **file** resource rather than inside the archive, and
+that difference is the safety: a file is written atomically — temp file, then rename — so the running
+launcher keeps the inode it was started from, where an archive writes in place with truncate and would
+cut the script a running shell is reading. The manifest carries a second copy of the launcher and a
+test refuses any difference from `packaging/nox-mesh-host-launch`.
+
+On the workstation, in order: the version landed, the launcher was replaced, the running host saw a
+delivered version and stood aside, and after one restart of the unit the launcher started
+`/usr/lib/nox-mesh-host/versions/637f65559d16/nox-mesh-host`. **A host the mesh compiled, published,
+delivered and started.**
+
+It would then have refused the first declaration it was asked to apply. The host's Makefile links in
+two facts the mesh's toolchain does not, and one of them — the system it was built for — is read before
+anything is applied. That is
+[issue 161](../161-a-delivered-host-carries-none-of-its-link-time-facts/00-report.md), and the machine
+is back on its hand-placed binary until it is answered.
+
+**The fallback is what made that safe**, and it was not luck: the launcher runs the pinned version, or
+the newest delivered one, or the one placed by hand — so moving the delivered versions aside restored
+the machine in one step.
+
+## Self-update works (2026-09-30, end of the day)
+
+```
+running: /usr/lib/nox-mesh-host/versions/093231796eb0/nox-mesh-host
+mesh-controller node show shanks
+  host              093231796eb0
+```
+
+One machine runs a host the mesh compiled, published, delivered and started, applying declarations and
+reporting the version it was delivered as. The two facts a delivered binary was missing are
+[issue 161](../161-a-delivered-host-carries-none-of-its-link-time-facts/01-resolution.md) and resolved:
+the system comes from the artifact, the version from where the binary sits.
+
+**Three machines still run a hand-placed host**, and rolling each forward is one assignment and one
+push. The control node is worth last.
+
+One thing this found on the way out: an archive cannot be undeclared, and the attempt stops the machine
+applying anything at all — [issue 162](../162-an-archive-cannot-be-undeclared/00-report.md). It is how
+undoing the first delivery froze the workstation, and it is not specific to the host.
+
+## Every machine self-updates (2026-09-30, evening)
+
+```
+shanks  76f4566bef3d/nox-mesh-host  active
+g14     76f4566bef3d/nox-mesh-host  active
+novox   76f4566bef3d/nox-mesh-host  active
+ace     76f4566bef3d/nox-mesh-host  active
+
+mesh-controller status: (no host split)
+```
+
+The last delivery was unattended on all four: the fixed host was built, pushed, each machine stood
+aside exactly once for the genuinely newer version, and the delivered launcher started it — no
+restart by hand. A following push that delivered nothing new was applied and reported by every
+machine and stood nobody aside, which is the check
+[issue 163](../163-a-delivered-host-stood-aside-on-every-push-and-reported-nothing/00-report.md) asks
+for.
+
+**Two more faults on the way, both mine, both found by reading the machine rather than the success
+line.** A delivered host compared the newest delivered version against its link-time stamp rather
+than the version it was running, so it stood aside on every push and — because standing aside cancels
+the report — never reported again (163). And the adopted machine kept its found launcher as the
+adoption rule says, so the delivery there needed a `take` before the launcher moved.
+
+**The crossover needs one restart of the unit per machine, once.** The launcher process that was
+running on each machine was the old script, executing from its own inode; a new file beside it
+changes nothing until the unit restarts. Every subsequent delivery is unattended.
+
+**Timing, measured:** on a machine, hearing a declaration to reporting it applied is about three
+seconds. A push as the operator sees it takes 17–20 seconds, and the difference is the control plane
+composing the declaration before it sends. A `--wait` shorter than that reads as "did not report" for
+a machine that did; the three-minute default read as slowness for a machine that never would. Neither
+number is a defect being chased here, and both are worth knowing before reading a push's answer.
+
+## What this leaves
+
+- [Issue 162](../162-an-archive-cannot-be-undeclared/00-report.md): an archive cannot be undeclared, so
+  the host module — and any module with an archive — cannot be unassigned, and trying stops the machine
+  applying anything.
+- [Issue 107](../107-a-declaration-carries-no-order/00-report.md) is unblocked: a declaration field is
+  now a build and a push rather than an expedition.
+- Three stale version directories on the workstation from the first attempts, moved aside under
+  `/var/lib/mesh-host/versions-held-back/`, and a backup of the adopted machine's hand-placed binary
+  beside its state. Both are safe to delete and are not the mesh's to delete.

04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md

@@ -4,7 +4,7 @@ opened: 2026-09-29
 located-in:
   - mesh-controller internal/catalogue/filtering.go (fixed for this instance)
   - mesh-controller (what status reports, and what it does not ask)
-fixed-by:
+fixed-by: partly — mesh-controller 1da96e8 makes the report state its own scope; nothing dials a provision yet, which is ADR 0146 and is not built
 amended-design: 03-DESIGN/01-to-be/10-delivery.md
 ---
 
@@ -90,9 +90,17 @@ same machine, another machine's service exposed to the private network, and anot
 exposed publicly. The filter had the second and third and expressed the first as a list of addresses that
 no container could match.
 
-**The question this issue is actually about is still open.** Nothing in the mesh would have said a grant
+**And then the question this issue is actually about was answered on its own terms.**
-had stopped working, and nothing does now. That is not answered by a configuration fix, and it should not
+[ADR 0145](../../02-DECISIONS/0145-a-module-checks-what-the-mesh-claims-is-reachable.md): a module on
-be answered by a mechanism adopted as a remedy for one.
+every machine serves an endpoint of its own and dials every other machine's, from the position the
+callers are in. Its probe is its own endpoint declared reachable over the private network, so it is
+admitted by exactly the rule that governs every internally-exposed service and fails when that rule is
+wrong — where a probe on a service every machine has would have passed for all eleven hours, because the
+services every machine has are the ones never closed.
+
+Adopted on its merits rather than as the remedy for a configuration error, which is what 0143 was and
+why it went. The module is written and merged; it is not yet assigned, so every machine currently reads
+unchecked.
 
 ## Open questions
 

04-ISSUES/145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/01-resolution.md

@@ -0,0 +1,64 @@
+# 145 — partly resolved: the report says what it is not a claim about
+
+*2026-09-30.*
+
+## What was done
+
+The sentence that was true for eleven hours now states its own scope, immediately below itself:
+
+```
+4 machine(s), all doing what they were told, all heard from, running what the mesh would send
+them, and every module current with its source
+
+  That is the mesh and the machines agreeing. Nothing here dials a provision:
+  no grant the mesh composed has been tested, so a module unable to reach what it
+  requires would not appear above (04-ISSUES/145)
+```
+
+That is the whole of what this change does, and it is deliberately small. It does not check anything.
+It closes the distance between *the machines are as the mesh described them* and *it works* by naming
+it, and that distance is where the eleven hours went: the report was read as the second and only ever
+meant the first.
+
+Two other reports in this group now carry real information they did not
+([issue 125](../125-a-hold-is-not-a-line-in-the-apply-report/00-report.md): a hold is a line in the
+apply report and breaks the all-well sentence;
+[issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md): the mesh knows which
+host runs a machine). Neither would have caught this fault, and both were the same shape of blindness.
+
+`printStatus` is now separated from the asking, so these words can be read by a test with no store, bus
+or machine — they have been acted on and been misleading twice, which makes them worth holding still.
+
+## What is NOT done, and why this record stays open
+
+**Nothing dials a provision.** [ADR 0146](../../02-DECISIONS/0146-connectivity-is-checked-by-name-per-hosting-form.md)
+decides how it should be done — a module on every machine serving an endpoint of its own and dialling
+every other machine's, one name per hosting form, over TLS with the certificate verified. **Nothing is
+built**, the module that existed was deleted, and the work was deferred deliberately by the operator.
+It is not this record's to start.
+
+So the measured fault of this issue — that the mesh can only report on itself — is unchanged. What
+changed is that the report no longer implies otherwise.
+
+## The open questions, where they stand
+
+- *Should a grant be checked, and from where?* Answered by ADR 0146 and not built: from the position
+  the callers are in, by a module on every machine, per hosting form.
+- *What would it cost to be wrong in the other direction?* Unanswered and important. A check that
+  reports a provision broken while it works trains a reader to ignore the report, which is the failure
+  this whole issue is about arriving by the other door.
+- *What should `status` say about a machine whose modules cannot reach each other?* Answered in part:
+  until something checks, it says that it has not checked. What it says when a check exists is ADR
+  0146's to settle.
+- *Is there a cheaper signal than a probe?* Still open. The affected module logged the failure 6,154
+  times; the mesh reads no module's logs and arguably should not, but something a module could *say*
+  about its own provisions would have surfaced this in minutes.
+- *Does the same blindness apply to a provider that lost a consumer's grant?* Still open, and still
+  nothing checks it.
+
+## One thing worth carrying forward
+
+**The certificate half of ADR 0146 is now possible where it was not.** Its check requires an internal
+name fetched over TLS with the certificate verified, and until 2026-09-30 no machine trusted the mesh's
+authority at all ([issue 129](../129-nothing-makes-a-machine-trust-the-meshs-authority/02-resolution.md)).
+Three of four do now. Whoever builds 0146 no longer has to solve that first.

04-ISSUES/146-the-foundation-cannot-be-raised-on-the-bus-the-mesh-runs-on/00-report.md

@@ -0,0 +1,85 @@
+---
+status: located
+opened: 2026-09-29
+located-in: [mesh-host examples + internal/link, mesh-controller internal/broker]
+---
+
+# 146 — the foundation cannot be raised on the bus the mesh runs on
+
+## What was observed
+
+Raising a first node in the lab, to check a module against a real mesh, fails before any module is
+reached. Two separate faults, in the two bundles that exist:
+
+**The older bundle raises a control plane that cannot start.** It brings up the previous broker,
+and the control plane it then starts says, once every few seconds, for ever:
+
+```
+mesh-controller: this control plane has no MESH_BUS_NATS, so it cannot reach the mesh's bus
+```
+
+That is the control plane being right. The mesh moved to one bus
+([ADR 0131](../../02-DECISIONS/0131-everything-on-the-mesh-speaks-to-the-broker-seat.md)) and the
+bundle did not. Every bed that raises a foundation raises this one, so every bed is in this state.
+
+**The newer bundle, written for the new bus, stops one step earlier.** Its certificate step asks a
+container to make the broker's certificate:
+
+```
+docker run --rm --entrypoint sh -v <the broker's tls volume>:/tls <the bus image> \
+  -c "test -f /tls/tls.crt || (openssl req -x509 ... )"
+...
+failed bus-certificate: running the action: docker exited 127
+```
+
+127 is *command not found*. The bus's image has a shell and no `openssl`; the previous broker's
+image had both, which is why the step worked when it was written against that one. Substituting the
+store's image — the only other image the bundle carries — does not help: it has no `openssl`
+either. So the step as written cannot succeed with anything the bundle names, and the fault is not
+one image's: **the bundle asks for a certificate to be made by a tool it never says must be there.**
+
+Measured 2026-09-29 on a fresh lab machine, both bundles, from bare.
+
+## Why this is here and not a note in the knowledge base
+
+The mesh's own foundation is the one thing it cannot raise. Nothing reports that: the bundles are
+files in a repository, nothing applies them but a person raising a node, and the last thing that
+did was the hand-driven cut-over
+([ADR 0131](../../02-DECISIONS/0131-everything-on-the-mesh-speaks-to-the-broker-seat.md), whose
+work was done on the machines rather than from a bundle). So the state where the mesh cannot make
+another one of itself is reachable, and was reached, without anything saying so.
+
+It is also load-bearing for everything else: a lab bed proves a claim by raising a mesh, so while
+this holds, **no bed can run**, and every "checked in the lab" written from now on is a promise
+against a suite nobody can execute.
+
+## What would have prevented it
+
+- **Something raising the foundation on a schedule, from the bundle, as it is written** — the
+  bundle is the mesh's own installer and nothing installs from it. A bed that raises a first node
+  is exactly that check, and it is the bed that cannot run.
+- **A step naming what it needs.** The certificate step names an image and assumes a program inside
+  it. An action that said which tool it requires would have failed at the declaration rather than
+  at 127 on a machine.
+
+## Evidence to carry into diagnosis
+
+- `mesh-host examples/foundation-first-node.lock` — the previous broker, no `MESH_BUS_NATS`.
+- `mesh-host examples/foundation-first-node-nats.lock` — the new bus; `bus-certificate` and its
+  `verify` both run `openssl` in the bus's image.
+- The bus image the bundle pins has `sh` and no `openssl`; the store's image likewise.
+- The lab rewrites a bundle's registry-prefixed third-party references to upstream ones for a
+  machine with an uplink (`test/integration/harness.ts`); the new bundle's bus reference needed
+  that rule added, which is done and is not this issue.
+
+## What one of its fixes then did to a running mesh (2026-09-30)
+
+The change that stopped the doubling — putting the stream into a push consumer's delivery subject —
+is correct on a foundation being raised and fatal on a mesh that is already running: the server will
+not move that subject while a subscriber is bound, and a node is bound to its declaration consumer
+the whole time it is up. The control plane crash-looped on the first build that carried it.
+
+Recorded and fixed as [issue 156](../156-moving-a-consumers-delivery-subject-stops-the-control-plane/00-report.md).
+Noted here because this record is where somebody will arrive when reading why the subject carries the
+stream at all, and the answer is incomplete without it: **the raise path was the only one exercised,
+and it is the one path on which nothing is bound.**

04-ISSUES/146-the-foundation-cannot-be-raised-on-the-bus-the-mesh-runs-on/01-diagnosis.md

@@ -0,0 +1,184 @@
+# Diagnosis
+
+*2026-09-29, by raising a first node in the lab over and over and writing down each thing it hit.*
+
+Not one fault. **Four, stacked**, each hidden behind the one before it, and every one of them the
+same shape: a step that was right while the mesh ran on the previous broker and was never asked a
+question again after the bus changed. Nothing had raised a foundation since, so nothing said so.
+
+## 1 — the bundle's bus image is named for a registry that is gone *(fixed)*
+
+The newer bundle pins `<a lab registry>/nats@…`, which resolves nowhere outside the lab that
+raised that registry. The lab already rewrites the store's and the previous broker's references to
+upstream ones for a machine with an uplink; the bus had no such rule because no bed had ever tried
+to raise this bundle. Added (`mesh-lab test/integration/harness.ts`). The digest is the bundle's
+own — what the registry served was a copy, so the same digest resolves upstream, and this is a
+prefix being removed rather than a reference being replaced.
+
+## 2 — the bus's certificate was made by a tool the bus does not have *(fixed)*
+
+```
+failed bus-certificate … docker exited 127
+```
+
+The step ran `openssl` inside the broker's image. The previous broker's image carried it; the bus's
+does not — it is Alpine with a shell and no `openssl` — and neither does any other image the bundle
+names, so there was nothing to substitute. **The program that needs the certificate now makes it**:
+`mesh-controller broker certificate --into <dir>`, with `--check` as the step's verify. The
+controller is already on the machine at that point (the schema step ran it) and needs nothing from
+the image it writes into. Self-signed, as before and on purpose — a host pins this server's exact
+certificate ([ADR 0004](../../02-DECISIONS/0004-a-node-and-how-it-joins.md)) and at that moment
+there is no authority to ask. Idempotent, because a second certificate is one every host that
+pinned the first no longer believes. It runs `--user 0:0`: the volume is root's, and the control
+plane's image runs as nobody, which is right for the long-lived server and wrong for a one-shot
+writing into a fresh volume.
+
+## 3 — enrolment dialled TLS at a bus that speaks first *(fixed)*
+
+```
+mesh-host: cannot reach the broker at …:5671: tls: first record does not look like a TLS handshake
+```
+
+Enrolment opened a raw TLS connection to check the pinned certificate before saying anything. NATS
+speaks its own protocol and upgrades afterwards, so the handshake met a plaintext greeting. The pin
+was never the problem: the same pinned configuration is handed to the client that presents the
+token, and the verification runs inside *that* handshake — so what
+[ADR 0004](../../02-DECISIONS/0004-a-node-and-how-it-joins.md) requires still holds, and holds
+better, because the one-time secret is sent only after the certificate has been checked. The raw
+dial is gone from the enrolment path and kept only as what its tests always proved: that a wrong
+certificate is refused before a byte of application data is sent.
+
+**Then, immediately behind it:**
+
+```
+mesh-host: this token is for the "" bus, and the mesh's bus is nats
+```
+
+The enrolment left the transport empty and meant *whatever the mesh runs today*, which was true
+while two buses existed and became a refusal the moment one did. The host knows which bus the mesh
+runs; it says so now.
+
+## 4 — a first node cannot be let onto its own bus *(open, and this is the real one)*
+
+```
+mesh-host: cannot reach the bus at …:5671 as anchor: nats: Authorization Violation
+```
+
+The bus's user list is a file beside its configuration. The installer carries the first one — the
+controller's own account at a bootstrap password — and **the controller composes every user after
+that** (design 25 §6; the controller's own test asserts the carried list matches what it would
+derive). On the running mesh that composition reaches the bus because the bus is a *module*, with
+the list delivered to it the way anything is delivered to a module.
+
+At genesis there is no module. The foundation's bus is raised by the installer, the control plane is
+given no way to write beside its configuration — it mounts the certificate and nothing else — and so
+the account a joining node needs cannot come into existence. **The first node cannot join the mesh
+it just raised.**
+
+That is not a line to fix in a bundle. It is the open half of the mesh delivering its own components
+([ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md)) and of the
+bus becoming a module: either the installer's bus is raised as the module the mesh will go on
+managing, or genesis carries a user list that includes the first node's enrolment and the controller
+takes over from there. Both are decisions, not patches, and both belong to the genesis step that was
+deliberately left until last.
+
+## 5 — the composed user list has to be placed by hand at genesis *(fixed)*
+
+The account a token is the password of is **not recorded at all**: the composer names an enrolment
+user for every machine with a live token, nothing minted a credential for it, and the composition
+left it out as a user with no password. The comment above the issuing code already claimed
+otherwise — *"the account is created before the token is handed over"* — which is how it went
+unnoticed. Issuing a token now records that account, with the token's own secret as its password,
+because that is the string the machine will present.
+
+Placing it is the other half. The list reaches the machine running the bus in that machine's
+declaration, which a machine that has not enrolled does not get, so at genesis it cannot arrive
+that way. **The control plane composes and says what it composed** — `broker accounts`, to standard
+output — and whoever is raising the machine writes it beside the bus's configuration and makes the
+server re-read it. Twice, because two accounts come into existence at different moments: the
+enrolment when the token is issued, and the machine's own when it enrols. A control plane that
+wrote the file itself would have to know where the bus keeps its configuration and how to make it
+reload, which is the module's knowledge and is what the module takes over on the first push.
+
+With that, **a first node enrols against the bus it just raised** — measured, from bare, in the
+lab.
+
+## 6 — and is enrolled twice, keeping a credential the mesh has replaced *(fixed)*
+
+```
+mesh-controller: enrolled anchor
+mesh-controller: enrolled anchor      (the same second)
+```
+
+One `enrol` on the machine, two enrolments in the control plane. Each mints the node a fresh bus
+password and returns it; the machine keeps the answer to the first, and the mesh keeps the hash of
+the second. The machine then reconnects for ever as a user whose password the mesh rotated out from
+under it — *authentication error - User "node.anchor"* on the bus, `Authorization Violation` in the
+host's log, and a node that never reports.
+
+What is ruled out: the host asking twice — it asks again only when the mesh says *try again*, and
+a refused attempt is not logged as an enrolment. Redelivery by the consumer — there is one
+consumer, its acknowledgement window is thirty seconds, and the handler is quick.
+
+What is left: the client re-publishing when an acknowledgement is slow, which is what its defaults
+do. That was addressed by giving the publish a message id derived from its own bytes, so the stream
+discards the copy — **and the duplicate survived it**, so either the id is not reaching the stream
+or the second copy is not a copy. This is where the trail stops.
+
+Worth saying plainly: **the mint is the fragile part, not the delivery.** An enrolment answered
+twice is survivable if the answer is the same both times, and it cannot be — the mesh keeps only
+the hash, so a second answer is necessarily a different credential.
+
+**Found, and it is not about enrolment at all** *(fixed)*. The bus's own counters settled it: one
+message published, one held in the stream, one delivery, nothing redelivered — and the controller
+enrolled the machine twice. So the handler ran twice on one delivery.
+
+A push consumer delivers onto an ordinary subject, and **everything subscribed to that subject gets
+a copy**. The controller holds a consumer called `controller` on CONTROL and another called
+`controller` on EVENTS, and the delivery subject was derived from the consumer's name alone — so
+both were `_DELIVER.controller`, the one process held both subscriptions, and every message from
+either stream was acted on twice.
+
+Enrolment is where it drew blood, because enrolling twice mints twice and the second credential
+replaces the first. But it applied to **every report and every event the controller follows**, and
+it is the kind of fault that leaves no trace: nothing is redelivered, no counter is wrong, the work
+simply happens twice. The comment in the receiving code about a merge that ran the whole catalogue
+five times over on 2026-09-28 is the same shape seen from the other end.
+
+The stream is in the delivery subject now, because the pair is what identifies a consumer — the
+server scopes a durable's name to its stream, and this subject was the one place that scoping was
+dropped. A subscriber's permission gains the same shape, keeping the bare name so an existing
+consumer keeps working until the controller's next assertion moves it.
+
+*How it is checked:* the consumers the mesh asks for are asserted to deliver onto distinct subjects,
+in the controller's own suite. Against a server it would be invisible, which is the point.
+
+## Where it belongs
+
+`mesh-host` (the bundle and the enrolment path) and `mesh-controller` (the certificate command, and
+the composition that cannot reach the bus at genesis). Three of the four are fixed on branches; the
+fourth is the genesis work.
+
+## What made it slow, and what was changed so it is not
+
+Six faults behind one another, each found by raising a machine and reading what it said. What cost
+the most was not the faults:
+
+- **Every bed's own instructions named the bundle that cannot work**, so the first three attempts
+  ended in a control plane crash-looping on a missing bus. They name the working one now.
+- **A host binary built without its system** refuses everything it is given with *this host was
+  built for ""*, which reads like a broken bundle. The lab's README says so.
+- **`make image` in the control plane had been broken for as long as its base was pinned**: the
+  Dockerfile's fallback is a Go older than the module asks for, and the pipeline never saw it
+  because the pipeline passes the declared base in. It reads the base from the manifest now.
+- **Leaving the machine standing is what answers the question.** Every finding above came from
+  shelling in afterwards — the host's log, the bus's log, the file the bus was actually handed —
+  and none from the test's own output, which says only that nothing converged. The bed takes
+  `MESH_LAB_KEEP`, and the README says to reach for it first.
+
+## What it cost, for the next person
+
+Every lab bed still names `foundation-first-node.lock` in its own instructions, and that bundle
+raises the previous broker with a control plane that refuses to start without `MESH_BUS_NATS`. Until
+the fourth fault is answered and the two bundles become one, a bed runs with `MESH_LAB_BUNDLE`
+pointing at the NATS bundle by hand, and stops at the enrolment.

04-ISSUES/147-the-operators-tools-still-dial-the-bus-that-was-removed/00-report.md

@@ -0,0 +1,95 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in: [mesh-catalog modules/mesh-console, mesh-tools src/mesh.ts, mesh-controller internal/broker]
+fixed-by: ADR 0152; mesh-controller PR 164 (invokes); mesh-tools PR 20 (mesh serve, the tools verb); mesh-catalog PR 181 (mesh-console)
+amended-design: 03-DESIGN/01-to-be/34-the-console.md
+---
+
+# 147 — the operator's tools still dial the bus that was removed
+
+## What was observed
+
+Every tool call an operator makes against the mesh fails, on every machine, with the same answer:
+
+```
+AMQP not connected — cannot reach hal/mesh@novox
+AMQP not connected — cannot reach hal/mesh@shanks
+```
+
+The mesh moved to one bus and the previous transport was deleted
+([ADR 0131](../../02-DECISIONS/0131-everything-on-the-mesh-speaks-to-the-broker-seat.md), cut over
+2026-09-28). The surface an operator drives the mesh through — the tool bridge their assistant
+speaks to — still opens an AMQP connection, so it cannot reach anything. Not one node answers,
+including the machine the operator is sitting at.
+
+**What that leaves.** The mesh itself is healthy: nodes are current, modules run, the bus carries
+the mesh's own traffic. What is gone is the way a person asks it anything. Every question — what a
+node runs, what is assigned, what a module's settings are — has to be asked by opening a shell on
+the machine and running the control plane's binary inside its container, which is precisely the
+path the tool surface exists to remove, and which nothing checks, records or permits.
+
+It also silently changes how work gets done: an assistant told to use the mesh's tools finds them
+dead, falls back to `ssh` and `docker exec`, and the operator discovers the fallback rather than
+the fault.
+
+## Why this is here and not a note in the knowledge base
+
+The rule is that everything happens over the mesh's bus. A surface that cannot reach that bus is
+that rule enforced by nothing — and the mesh reports itself healthy throughout, because what broke
+is not a module, a node or a provision but the thing standing outside asking them questions.
+
+The same cut-over removed the assistant's long-term memory (`recall`, `memorize`) for the same
+reason, which is why lessons from the last two days were written into this repository by hand.
+
+## What would have prevented it
+
+- **The tool surface as a consumer of the bus like any other**, so moving the bus moves it — rather
+  than a separate bridge with its own connection settings that nothing resolves.
+- **A check that a tool call reaches a node**, run where the mesh's other checks run. Every check
+  the mesh makes today is about the relationship between the mesh and a machine; none asks whether
+  a person can ask it anything ([issue 145](../145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md),
+  the same shape one level out).
+
+## Diagnosed at once, because the answer was in the configuration
+
+**The surface is not the mesh's.** The tool server the operator's assistant speaks to is the
+predecessor's brain, installed on the workstation and started as a local process, with the
+predecessor's broker URL — `amqp://…@<the control node>` — written into the assistant's own
+configuration. Nothing about it is a module: it has no manifest, no assignment, no seat, no account
+on the mesh's bus, and the mesh has never known it exists.
+
+So nothing regressed. The mesh removed a transport that this program still dials, and the program
+was never part of the mesh to be moved. **The mesh has a tool model** — `tools` in a manifest, the
+calls a seat accepts, the subjects a module answers on, and a control-plane command that asks one —
+and **nothing publishes an operator-facing surface onto it.** What an operator uses is the thing
+that came before, kept alive by a URL in a file.
+
+That is the issue, and it is larger than a broken connection: the way a person drives this mesh is
+outside the mesh.
+
+## Evidence to carry into diagnosis
+
+- The failure text names `hal/mesh@<node>`, so the bridge is resolving a node and then dialling the
+  old transport.
+- It fails identically for the local machine, which rules out reachability and points at the
+  transport alone.
+- The mesh's own traffic over the new bus is unaffected: nodes report, declarations apply.
+
+## Answered, 2026-09-30
+
+The work order's question — does the mesh grow its own operator surface, or is the surface an
+ordinary module — is answered by [ADR 0152](../../02-DECISIONS/0152-the-operators-surface-is-a-module-the-console.md):
+**the console is a module.** `mesh-console` is assigned to the machine a person sits at, holds a
+credential the mesh minted, calls tools under a grant its manifest declares (`invokes`), and serves the
+mesh's tools on that machine's loopback to an agent over MCP and to a person through the same endpoint.
+Designed in [34 — The console](../../03-DESIGN/01-to-be/34-the-console.md).
+
+What that leaves, said plainly so nobody reads this record as closed on the whole of its first
+paragraph: the console reaches every tool a *module* serves. The mesh's own questions — what a node
+runs, what is assigned — are the `mesh-controller` seat's tools under ADR 0132 and are not on the bus
+yet; for those a shell is still the way, and design 33 is where that closes.
+
+The predecessor's program on the workstation is not replaced by the mesh; it is left where it is and
+the assistant is pointed at the console beside it. The `hal` entry in the assistant's configuration
+still names things that are not the mesh's.

04-ISSUES/148-a-manifest-outside-this-catalogue-has-no-check/00-report.md

@@ -0,0 +1,50 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in: [mesh-controller cmd/mesh-controller, mesh-controller internal/catalogue]
+fixed-by: mesh-controller PR 164 (module check)
+amended-design: 03-DESIGN/01-to-be/12-a-module-repository.md
+---
+
+# 148 — a manifest outside this catalogue has no check
+
+## What was observed
+
+A module's manifest is validated by a **test** — `internal/catalogue`'s suite parses every manifest
+in the catalogue checkout beside it and fails on one it cannot resolve. That works, and it is how
+several real faults were caught before a machine saw them.
+
+It is available to exactly one repository: this one. Somebody describing their own application in
+their own repository — the case
+[ADR 0037](../../02-DECISIONS/0037-where-a-module-lives.md) calls *the case that matters most* —
+has no check at all. They write a manifest, register it with a running mesh, and find out whether
+it is valid when the mesh refuses it, or later, when a machine applies something that resolved and
+should not have.
+
+The same record asks for the answer: **a `module check` command on the control plane's binary**, so
+a manifest is checked by the tool rather than by a test that imports the tool's internals.
+
+## What would have prevented it
+
+Nothing prevents this; it was noticed and left. ADR 0037 named it on 2026-09-01 and the record sat
+`proposed` until 2026-09-29, so the missing half was never anybody's task.
+
+## Evidence to carry into diagnosis
+
+- `mesh-controller/internal/catalogue` — `ParseManifest` and `CatalogueProblems` are the check, and
+  both are internal.
+- The catalogue-wide test is `TestEveryCatalogueManifestDeclaresWhatItMounts` and its siblings; they
+  take a path from `MESH_CATALOG`, so the mechanism is already path-driven and not repository-bound.
+- `mesh-controller module add` refuses a bad manifest at registration, which is the same check far
+  too late: by then it is in a running mesh's records.
+
+## Built, 2026-09-30
+
+`mesh-controller module check <manifest>…` runs what registration runs — the strict parse, the
+per-manifest problems, and the cross-manifest rules over every manifest given — with no store and no
+mesh, prints every problem in the manifest's words, and exits non-zero on any. What it cannot judge
+without a store it says: a claim on one of the mesh's own seats is judged fully only at registration
+([ADR 0122](../../02-DECISIONS/0122-a-seat-is-data-a-rename-is-a-database-update.md)), and a seat
+declared by a module whose manifest was not passed reads as unknown. Written into
+[12 — A module repository](../../03-DESIGN/01-to-be/12-a-module-repository.md) as the section *a
+manifest is checked where it is written*. The console's own manifest was the first checked with it.

04-ISSUES/149-a-declaration-that-shrinks-to-empty-is-skipped-not-sent/00-report.md

@@ -1,10 +1,15 @@
 ---
-status: located
+status: resolved
 opened: 2026-09-27
-located-in: [mesh-controller cmd/mesh-controller/push.go]
+located-in: [mesh-controller cmd/mesh-controller/push.go, mesh-controller cmd/mesh-controller/sendable.go]
+fixed-by: mesh-controller sendable.go and push.go — an empty declaration is sent carrying `owns_nothing`, and the host refuses an empty body that does not carry it
 ---
 
-# A declaration that shrinks to empty is skipped, so the node keeps what it should drop
+# 149 — a declaration that shrinks to empty is skipped, so the node keeps what it should drop
+
+*Opened as 127 and renumbered on 2026-09-29: two records were given that number on the same day.*
+*The other kept it, because three documents and three source files cite it by number and nothing
+cited this one but a decision and a sibling issue, both corrected with this move.*
 
 ## What was observed
 
@@ -37,3 +42,17 @@ mean "own nothing", which the host already applies correctly when it receives on
 
 On ace, one command drops it permanently (the corrected controller never re-composes it):
 `sudo ufw delete allow 5671`. At ace's converge it would clear on its own.
+
+## Closed
+
+*2026-09-29, in a grooming pass.* Both halves are on `main` and both name this issue.
+
+- The control plane **sends** it: a declaration that composes to no resources goes out with
+  `owns_nothing`, and `push` says *sent, not skipped*.
+- The host **refuses an empty body that does not carry it**, so a truncated or mis-composed
+  declaration can never be read as "own nothing" — which is the failure the fix had to avoid while
+  making the empty case expressible.
+
+Closed by reading the code rather than by watching a machine let go of a stray resource; the record
+says so rather than implying a run.
+

04-ISSUES/150-a-route-is-contributed-before-its-module-is-taken/00-report.md

@@ -0,0 +1,52 @@
+---
+status: open
+opened: 2026-09-29
+located-in:
+  - mesh-controller internal/catalogue/declaration.go (contributions are emitted for an assigned module, taken or not)
+fixed-by:
+---
+
+# 150 — A route is contributed before its module is taken
+
+## What was observed
+
+ace is adopted and runs `route-adapter` beside the predecessor's traefik (ADR 0104). The first web
+module migrated there was searxng. `assign ace searxng` + `push ace` — the step that is supposed to
+change nothing on an adopted machine (ADR 0100, hq 125: assign holds, take replaces) — took
+`searxng.zurag.be` down:
+
+```
+https://searxng.zurag.be/  →  502
+```
+
+for about five minutes, until the module was unassigned again.
+
+## Why
+
+Assign held everything it found on the machine — the predecessor's `searxng` container, its
+directories — exactly as designed. But the module's **route contribution** is not a resource on the
+machine, so nothing held it: it reached `route-adapter` at once, which wrote
+`mesh-searxng.zurag.be.yml` into traefik's file provider pointing at the mesh's assigned machine port
+(`http://ace.internal:20000`) — where nothing listened, because the container that would was held.
+traefik's file router for the name then won over the predecessor's docker-label router for the same
+name, and the name served a dead backend.
+
+On this occasion the window was lengthened by an unrelated failure (the machine's docker address
+pools were exhausted, so the module's network could not be created), but the fault does not depend on
+it: **between assign and take, every routed module's public name points at a backend the mesh has
+deliberately not started.** The runbook's §6 step 2 ("check the preparation — HAL's service still
+serves") is false for every routed module on a node running route-adapter.
+
+## What the operator did
+
+Rolled back per runbook (unassign before take), then re-ran as assign → push → wait for the node to
+report its holds → take → push, keeping the window to the ~10 s between the two pushes. A workaround,
+not a fix: a module whose data must move between assign and take (runbook §6 step 3) cannot shrink
+its window this way.
+
+## What would be right
+
+A contribution from a module that is assigned but **not taken** on an adopted node should be held like
+the module's resources are — withheld from the provider until take — or the provider should be told
+the contributor is held and leave the predecessor's route alone. Either keeps "assign changes nothing"
+true for routed modules.

04-ISSUES/151-a-new-name-recreates-every-container-in-the-mesh/00-report.md

@@ -0,0 +1,106 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in:
+  - mesh-host internal/apply/apply.go (containerSpecReading hashes every `host` entry)
+  - mesh-controller internal/catalogue/declaration.go (withMeshNames gives every container the mesh's names)
+fixed-by: mesh-controller PR 161 — the roster left every container's declaration and so its digest (ADR 0148 step 3, 2026-09-30)
+---
+
+# 151 — A new name recreates every container in the mesh
+
+## What was observed
+
+Migrating one small module on ace (searxng) took four routine controller actions: `node public-domain
+ace zurag.be`, `assign ace searxng` + push, `unassign ace searxng` + push (a rollback, see
+[issue 150](../150-a-route-is-contributed-before-its-module-is-taken/00-report.md)), and assign + take
++ push again. Each push to ace also pushed novox ("this push left g14, novox, shanks behind … sending it
+too"). novox's host then **replaced every container it runs, twice**:
+
+```
+22:46  … updated postgres.server (mesh-store): replaced; a container's configuration is fixed when it is created
+22:46  … updated distribution.store (mesh-registry): replaced; …
+22:46  … updated route-proxy.server (route-proxy): recreated …
+22:51  … updated postgres.server (mesh-store): replaced; …
+22:52  … updated gitea.server (gitea): replaced; …
+22:52–22:55  mesh-vault, builder, all of mailu, keycloak, minio, mongodb, invoicing, photos, umami, …
+```
+
+Replacements per minute on novox, 22:46–22:55: 12, 8, 13, 5, 6, 8, 8, 8, 8, 4. The control plane was
+unreachable twice while its own store came back through crash recovery
+(`FATAL: the database system is starting up`), and dependants (umami) crash-looped until it did. None
+of the replaced containers belonged to the module being migrated, or to ace.
+
+## Why (confirmed part)
+
+Every container the mesh runs is given the mesh's names as `--add-host` entries
+(`withMeshNames`), and the host puts every entry into the container's spec digest — deliberately,
+since [issue 135](../135-a-containers-mesh-names-are-not-compared/00-report.md): a container left alone when the roster moved kept a five-day-old
+address and restarted 2286 times. A running container cannot have its hosts changed, so a changed
+digest means a replace.
+
+The consequence is that **the roster is part of every container everywhere**: anything that adds,
+removes or moves one name — a node's public domain, a routed name, an unassign — replaces every
+container on every machine that carries the list. On the hub that includes the control plane's store,
+the registry, the edge and mail.
+
+## Not yet established
+
+Which name moved in each pass. gitea's hosts after the fact list the `.internal` names and novox's
+internally routed `*.novox.be` names, and **not** `searxng.zurag.be` — so it is not simply "a routed
+name was added". Two passes suggest the set changed at assign and changed back at unassign; the
+declarations before and after would say, and nothing on the machine records the previous one.
+
+## Why it matters now
+
+The node-by-node migration adds names one module at a time. On ace alone that is ~25 web modules; if
+each assignment moves the roster, each is a full restart of every hub service, and a rollback is
+another. The migration is paused on this.
+
+## What has since been ruled out as a cause (2026-09-30)
+
+The roster was also moving for a reason that was not a name changing at all: a node whose plan would
+not compose had its routed names silently dropped from the roster handed to every machine, so a
+briefly unreachable store withdrew and restored a name on alternating passes. That is
+[issue 152](../152-a-nodes-plan-failure-silently-drops-its-routed-names/00-report.md), and it is
+fixed — it is what made the churn on the control node repeat every six minutes for twenty-five
+minutes rather than once per operator action.
+
+**It does not close this record.** A roster that changes for a real reason — a machine joining, a
+module assigned, a public domain set — still replaces every container on every machine that carries
+the list. 152 removed the false reasons; the question below is still open.
+
+## What would be right (for diagnosis)
+
+Keep 135's guarantee (no container runs with a stale address) without making the roster part of every
+container's identity — e.g. resolve mesh names through a resolver the container asks at lookup time
+rather than baked entries, or scope each container's entries to the names it actually binds.
+
+## Answered (2026-09-30): the first of those two
+
+[ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md) takes
+the resolver, not the scoping. Scoping was the close call and was rejected for contradicting the
+standing intent that anything on the mesh can call anything on it: it would make a name reachable only
+where the mesh was told in advance it would be wanted, and it would leave the roster in the digest, so
+the churn returns whenever a widely-bound name moves.
+
+Resolving at lookup time makes staleness impossible rather than detected — 109 and 135 stop being bugs
+that were fixed and become a shape that does not exist — and makes a name's blast radius nothing.
+
+**This record stays open**, because the record answers it and the code does not. Nothing may stop
+copying names until a container can reach the resolver from any of the runtime's networks
+([issue 110](../110-a-container-on-the-runtimes-own-network-cannot-reach-the-resolver/00-report.md)),
+which it cannot on two of four machines today. Removing the copy first reintroduces 109 and 135
+silently, on a live mesh, which is how both were found. The order is in the record.
+
+## Resolved (2026-09-30)
+
+Step 3 landed the day 110 did. mesh-controller PR 161 stops writing the roster into any container, so
+a container's digest no longer carries a name that is not its own. The controller's tests hold the
+record's check — a container's declaration does not move when the mesh's roster does, and does move
+when the module's own declared entries do.
+
+The first push after the change recreated every container once, because every digest lost its host
+entries at the same moment. That was the last such event: from here a name added or moved on one
+machine changes no container anywhere, and the record's second check — add a routed name, watch every
+other machine's apply report say nothing changed — is what the next module assignment will show.

04-ISSUES/152-a-nodes-plan-failure-silently-drops-its-routed-names/00-report.md

@@ -0,0 +1,124 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in:
+  - mesh-controller cmd/mesh-controller/plan.go (routeNamesInTheMesh skips a node whose plan will not compose)
+fixed-by: mesh-controller 6c5dfd0 (PR 147)
+amended-design:
+---
+
+# 152 — A node whose plan will not compose silently removes its names from every machine
+
+## What was observed
+
+For twenty-five minutes, and for seventeen of them after the last operator action, the control-node's
+host applied all 327 of its resources every ~6.5 minutes without pause, replacing every container on
+the machine each time — the control plane's own store and registry, the edge proxy, the forge, the directory, mail,
+and the bus the mesh runs on. The forge's web surface answered `502` throughout; load on the machine
+sat near 8. Nothing was converging: each pass ended and the next began four seconds later.
+
+The two machines carrying no containers were not churning. They were only knocked off the bus each
+time the control node re-created it, reconnected, re-heard the same declaration and applied it again
+as a no-op.
+
+## Why: the roster alternates between two values, and it is part of every container
+
+Two consecutive declarations were compared by reading the `--add-host` entries of four containers
+the moment each pass created them:
+
+```
+23:02  ace.internal  drive.novox.be  g14.internal  keycloak.novox.be  novox.internal
+       office.novox.be  portainer.novox.be  shanks.internal  umami.novox.be          (9 names)
+23:07  … the same nine, and searxng.zurag.be                                         (10 names)
+```
+
+One routed name — belonging to a module on another machine entirely — leaves the roster and comes
+back. Because the roster is part of every container's spec digest ([issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)),
+each flip is a different identity for every container on the machine, and a running container cannot
+have its hosts changed. So every flip replaces all of them.
+
+**What makes it flip is a swallowed error.** `routeNamesInTheMesh` composes every node's plan to
+find the names it serves, and when one will not compose it moves on:
+
+```
+plan, settings, err := planFor(ctx, open, n.Name)
+if err != nil {
+        continue
+}
+```
+
+A node whose plan cannot be composed *right now* therefore contributes no names — not "the mesh does
+not know", but "the mesh states these names do not exist", to every machine at once.
+
+## Why it sustains itself
+
+The loop closes through the control plane's own database:
+
+1. An apply replaces `mesh-store` — the store the control plane reads — by removing the container,
+   so postgres comes back through crash recovery.
+2. While it recovers it refuses connections: `FATAL: the database system is not yet accepting
+   connections / Consistent recovery state has not been yet reached` (observed, 21:08:39 UTC, every
+   pass).
+3. `planFor` for the other machine fails against that store. `routeNamesInTheMesh` swallows it and
+   drops its routed name.
+4. The roster changed, so all 327 resources differ, so all are replaced — including `mesh-store`,
+   and including the bus, which is why the host also cannot report: `applied, and could not tell the
+   mesh: reporting: nats: connection closed`.
+5. Back to 1.
+
+Every pass destroys the evidence the next pass needs to decide it has nothing to do, and nothing
+outside the machine has to be wrong for it to continue.
+
+**It is metastable, not permanent.** It ran from 22:46 to 23:11 — five full replacements of every
+container on the machine — and then stopped on its own, when one pass happened to read the store
+during a window it was up, composed the same roster twice running, and found nothing to do. Load fell
+from 7.8 to 1.7 and the machine returned to its five-minute idle tick.
+
+That it ends by luck is the point, not a mitigation. The exit condition is a race the mesh does not
+control, the operator cannot see, and nothing reports; the same four actions on a slower machine, or
+a larger store, would not have found it. An outage that clears itself after twenty-five minutes and
+five restarts of the forge, the directory and mail is not a smaller fault than one that does not — it
+is the same fault, harder to catch.
+
+## What it is not
+
+- Not the operator's four actions on the other machine. Those explain the first passes
+  ([issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md)); they were
+  finished seventeen minutes and three full passes before these measurements.
+- Not a file that keeps changing. `/etc/hosts`, the bus's account list and the vault's export were
+  hashed across passes and are byte-identical, and the directory reported `mode 755 to 700` every
+  pass while already being `700`. Those resources are **misreported as changed** and are worth their
+  own question, but they are not what moves a container's identity.
+- Not the lost report alone. A report that cannot be delivered explains a re-apply; it does not
+  explain a re-apply that finds 327 differences.
+
+## Why it matters beyond this outage
+
+The same `continue` makes every routed name in the mesh conditional on every node's plan composing at
+the moment any machine is pushed to. One unreachable or half-migrated machine is enough to withdraw
+its names from everywhere — and the withdrawal is indistinguishable, on the receiving machine, from
+the operator having removed them.
+
+The codebase already states the rule this breaks, forty lines away, about the same kind of lookup:
+
+> A lookup failure is an error, never "not found": collapsing the two composed a declaration without
+> the trust whenever the inventory hiccuped, delivered by a push that reported success.
+
+## How it was fixed, and how the fix is checked
+
+`planFor` now marks the two failures that really are a statement about the node — its set not
+composing, and a setting that reaches nothing — and the three gatherers pass over those and only
+those. Every other failure is raised, naming the machine and the read.
+
+Five tests hold it: a set that cannot compose is marked as the node's own; a store that cannot be
+read is *not*; one incoherent node still does not cost the rest their names; a roster is never
+returned beside an error; and the raised failure names what could not be read.
+
+The two sibling gatherers were audited and fixed the same way — the grant composer, which would have
+withheld a consumer's credential, and the private-network membership, which would have taken a
+machine off the overlay. Three other `planFor` callers were audited and left alone: they refuse or
+report rather than silently withdraw, which is the safe direction.
+
+**This does not close [issue 151](../151-a-new-name-recreates-every-container-in-the-mesh/00-report.md).**
+A roster that changes for a real reason still replaces every container in the mesh. This removes the
+false reasons; whether the roster belongs in a container's identity at all is that record's question.

04-ISSUES/153-an-adopted-machines-data-cannot-be-placed-where-it-is/00-report.md

@@ -0,0 +1,57 @@
+---
+status: open
+opened: 2026-09-29
+located-in:
+  - mesh-controller internal/catalogue/dir_into.go (dirsFor: a stated path or <data root>/<module>/<id>, nothing else)
+  - mesh-controller (accesses: the path is the manifest's literal)
+fixed-by:
+amended-design:
+---
+
+# 153 — An adopted machine's data cannot be placed where it is
+
+## What was observed
+
+Preparing ace's media modules (plex, sonarr, radarr, lidarr, bazarr, nzbget, qbittorrent, bookshelf)
+for migration. ace is adopted; its data is where the predecessor put it and **must stay there**:
+
+- the library and download spool: `/storage/media/*`, `/storage/downloads` — a separate ZFS pool,
+  ~40 TB, the operator's shared data (ADR 0051);
+- plex's own state: `/mnt/plex/{config,data,temp}` — 133 GB on a second disk;
+- large configuration directories held in place: lidarr 46 GB, radarr 17 GB, sonarr 3.2 GB.
+
+The catalogue's manifests name `/services/media/*` (as `accesses`) and `/services/<m>/config` (as
+owned directories), which is novox's layout, not ace's, and not a value a definition may carry.
+
+## What was decided, and what exists
+
+[ADR 0112](../../02-DECISIONS/0112-a-module-definition-names-no-node-mesh-or-path.md) (accepted) says
+exactly what is needed:
+
+> *Where* it is on the machine is the assignment's. A node has a default layout, and an assignment may
+> place a directory elsewhere: on a second disk, or where an adopted machine's data already is.
+
+> [ADR 0051]: an access keeps its shape and its semantics; its path moves from the definition to the
+> assignment.
+
+What the control plane implements (`dirsFor`): a directory is either a path the manifest states, or
+`<data root>/<module>/<id>` under the node's one data root. There is **no per-assignment placement of
+one directory**, and an `access` path is the manifest's literal — no setting reaches either.
+
+## Consequence
+
+Every module whose data an adopted machine already holds somewhere other than the default layout can
+only be migrated by (a) writing the machine's path into the manifest — which 0112 forbids and which
+is wrong on the next machine — or (b) moving the data into the placed layout in a window. (b) is
+acceptable for a 40 MB configuration and impossible for a 40 TB library the operator has ruled must
+never be moved, copied or re-owned.
+
+The same gap covers ownership: the predecessor runs ace's media stack as `1001:2000`; a manifest's
+`owner` is one value for every machine.
+
+## What would be right
+
+The two assignment halves 0112 decided: a setting that places a declared directory (by id) at a given
+path on this node, and a setting that says where an access's data is — both validated like
+`endpoints` (unknown ids refused), and an access placed by the assignment still never created,
+chowned or removed.

04-ISSUES/154-a-machines-own-network-is-not-a-reach/00-report.md

@@ -0,0 +1,42 @@
+---
+status: open
+opened: 2026-09-29
+located-in:
+  - hq 02-DECISIONS/0138 (reach: internal | public | both)
+  - mesh-controller internal/catalogue/filtering.go (Reaches)
+fixed-by:
+amended-design:
+---
+
+# 154 — A machine's own network is not a reach
+
+## What was observed
+
+Preparing ace's modules. ace sits on a home network (192.168.1.0/24) behind a router, and several of
+its services are reached **from that network by devices that will never be mesh machines**:
+
+- mosquitto `1883` — an IoT light switch (`sonoff-office-light-switch`) and home-assistant;
+- unifi `8080`/`3478 udp`/`10001 udp` — the access points' inform, STUN and discovery;
+- plex `32400` — LAN streaming clients (three connected at survey time);
+- home-assistant `8123`, and the resolver on the LAN address.
+
+ADR 0138 gives an endpoint's reach as `internal` (the private overlay), `public` (anywhere) or `both`.
+None of them says *this machine's own network*. The predecessor could: its unifi manifest opened
+inform/STUN/discovery `from: 192.168.0.0/16, 10.0.0.0/8, 172.16.0.0/12`.
+
+## Consequence
+
+The only reach that includes a LAN device is `public`. While ace is adopted that is harmless — its
+own firewall stays and admits the LAN — and behind NAT "anywhere" happens to mean the LAN. But:
+
+- it states the wrong thing: an operator reading `reach: public` on an IoT broker believes it is on
+  the internet, and a router port-forward added later for something else makes it so;
+- at `converge ace`, the mesh's filter is the sum of what it listens on (ADR 0045). An endpoint left
+  `internal` cuts every LAN device off at the flip; one set `public` opens it to the internet on any
+  machine with a public address.
+
+## What would be right (for diagnosis)
+
+A reach — or a source — that means the networks the machine is directly attached to (its uplink's
+subnets, as the machine reports them), so a LAN-only service is declared as exactly that and the
+filter can admit it without admitting the internet.

04-ISSUES/155-two-records-may-share-a-number-and-nothing-says-so/00-report.md

@@ -0,0 +1,58 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in: [hq 00-META/checks/cycle.py]
+fixed-by: hq f89aef9 (PR 192)
+amended-design:
+---
+
+# 155 — Two records may share a number, and every check passes
+
+## What was observed
+
+On 2026-09-29 two machines opened issues against this repository within the same hour. Both read
+`main` correctly and both took "the next free number", and they collided twice:
+
+| | one machine opened | the other had already used |
+|---|---|---|
+| first | 147, 148 | 147, 148 on an unmerged branch |
+| second | 149, 150 | 149 on an unmerged branch, 150 from renumbering the first collision |
+
+The first collision was reconciled by hand before merging. The second was **merged into `main`**, and
+`records.py`, `cycle.py` and `index.py` all reported success over a tree holding
+`149-a-declaration-that-shrinks-to-empty` beside `149-an-adopted-machines-data-cannot-be-placed-where-it-is`,
+and two folders numbered 150.
+
+## Why
+
+The number is allocated as `max(main) + 1`, and `main` lags every open pull request — seven of them
+that evening. Two readers of the same `main` therefore compute the same next number, and neither is
+doing anything wrong. The existing reconciliation precedent (a second record numbered 127 became 149)
+assumed a single writer, which stopped being true when a second machine began filing its own findings.
+
+## Why it matters
+
+An issue number is how every other record cites this one — `fixed-by:`, `located-in:`, a decision
+record's consequence, a commit message. Two records answering to one number is a citation that
+resolves to whichever folder the reader happened to open, and the failure is silent on both sides:
+the citer is not wrong, and the cited record exists.
+
+It is also exactly the class this repository says it does not permit — a rule (`00-META/process/03-issues.md`:
+"take the next free number") enforced by nothing.
+
+## How it was fixed, and how the fix is checked
+
+`cycle.py` now refuses a tree in which two issue folders share a leading number, and names both.
+Proven by adding a duplicate and watching it fail, then removing it and watching it pass.
+
+The colliding records were renumbered 153 and 154, in the branch that landed last — renumbering a
+branch whose author is still pushing only moves the race.
+
+**The check catches the collision; it does not prevent it.** Allocating a number still needs the open
+pull requests read as well as `main`. That is a habit the check now backstops rather than one it
+replaces, and playbook [03](../../00-META/process/03-issues.md) now says so at the step where the
+number is taken.
+
+[ADR 0080](../../02-DECISIONS/0080-the-development-cycle-is-checked.md) enumerates what `cycle.py`
+enforces and named four things; it carries a progressive insight naming the fifth. The decision
+stands — this is one more thing frontmatter and file names can carry, found by its absence.

04-ISSUES/156-moving-a-consumers-delivery-subject-stops-the-control-plane/00-report.md

@@ -0,0 +1,98 @@
+---
+status: resolved
+opened: 2026-09-29
+located-in:
+  - mesh-controller internal/broker/jetstream.go (EnsureConsumer)
+fixed-by: mesh-controller e7da39d (PR 148)
+amended-design:
+---
+
+# 156 — Moving a consumer's delivery subject stops the control plane, and only on a mesh that is running
+
+## What was observed
+
+The control plane crash-looped, every restart ending the same way:
+
+```
+mesh-controller: asserting how novox hears its declaration:
+  bringing consumer novox on NODES to match: nats: consumer name already in use
+```
+
+It came up on the first build of the controller in eight hours. The machines kept running what they
+already held — this stops the mesh being *changed*, not the services it placed — and nothing could be
+pushed, no report was consumed and no enrolment answered, for as long as it lasted.
+
+## Why
+
+[Issue 146](../146-the-foundation-cannot-be-raised-on-the-bus-the-mesh-runs-on/00-report.md) put the
+stream into a push consumer's delivery subject, because one process holding two consumers of the same
+name on two streams was given one subject and acted on every message twice.
+
+**The server will not move a push consumer's delivery subject while a subscriber is bound to it.** It
+refuses with `consumer name already in use` — a message about the name, for a conflict about the
+subject, which is why the trail starts in the wrong place.
+
+A node is bound to its declaration consumer the whole time it is up. That *is* a node listening for
+what it should be. So every node consumer in a mesh that is running is one the assertion cannot bring
+to match — and the assertion happens before the controller serves, so it never serves.
+
+The controller's own two consumers moved without trouble, and are on the new subject in the live mesh.
+It asserts them before it subscribes, so nothing was bound.
+
+## Why nothing caught it
+
+The change was exercised on a mesh being raised, where every consumer is created rather than updated
+and nothing is bound to any of them. On that path the code is correct. The test that would have caught
+it needs a mesh that is already running: an existing consumer, a subscriber still attached, and then
+the assertion.
+
+Reproduced exactly that way before the fix — same server version, same stream shape, same consumer —
+and it fails with the same words as the machine did. An earlier version of the same test unsubscribed
+first and passed against the code that was crash-looping on the control node.
+
+## What it is not
+
+- Not the change it shipped beside. The merge that triggered this build carried
+  [issue 152](../152-a-nodes-plan-failure-silently-drops-its-routed-names/00-report.md)'s fix and three
+  other commits that had never been deployed; this one is 146's.
+- Not a version difference. The test server and the mesh's broker are both nats-server v2.10.29.
+
+## How it was fixed
+
+The consumer that works is kept, and the assertion says so instead of failing.
+
+**Not deleted and re-made.** Re-making moves the subject, and a holder may not yet be allowed to
+subscribe to the new one: the wider grant travels in the bus's user list, which the control plane
+composes and a machine applies minutes later. On the live mesh the nodes are granted `_DELIVER.<node>`
+and not `_DELIVER.<node>.>` — re-making would have silenced every machine in the mesh, which is worse
+than the collision it was fixing and far harder to undo. That was the first fix written here, and the
+permission is the reason it was not shipped.
+
+**Not fatal**, which is what 146's change intended and did not do: the bare subject still delivers, and
+collides only where one holder has two consumers of one name. A node has one.
+
+## How the fix is checked
+
+Two tests against a real server: a consumer with a subscriber bound keeps its subject, is reported,
+and still delivers to that subscriber; a consumer with nothing bound moves, so 146's fix still applies
+where the collision actually was.
+
+## What is left
+
+The node consumers stay on the bare subject, which is correct and not tidy. Nothing is wrong while
+they do not move: one consumer per name per stream cannot collide with itself.
+
+The controller reports each one it kept, and did, on the start that fixed this — four node consumers
+and the build machine's worker, which is bound the same way and was not anticipated here:
+
+```
+consumer novox on NODES still delivers to "_DELIVER.novox" and not "_DELIVER.novox.NODES":
+  nats: consumer name already in use. It keeps working; the subject moves on an assertion
+  made while nothing is bound to it
+```
+
+**The wider grant has since landed** (2026-09-30, measured on the mesh's own broker config): every
+node is now allowed `_DELIVER.<node>.>` as well as the bare subject. That was the thing missing when
+this was diagnosed, and it is why re-making the consumers then would have silenced every machine.
+What remains is only the second half — an assertion made while each node is detached from its
+consumer — and that is its own piece of work, not a side effect of a restart.

04-ISSUES/157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md

@@ -0,0 +1,70 @@
+---
+status: resolved
+opened: 2026-09-30
+located-in:
+  - mesh-controller internal/catalogue/roster.go (the roster's entries for a routed name)
+fixed-by: mesh-controller PR 163 — a routed name is published as itself, once, with no suffixed alias (ADR 0151, 2026-09-30)
+amended-design:
+---
+
+# 157 — A routed name is published with an `.internal` alias that nothing serves
+
+## What was observed
+
+Every machine's hosts file carries two entries for each routed name — the name, and the name with
+`.internal` appended:
+
+```
+10.10.0.1	keycloak.novox.be.internal	keycloak.novox.be
+10.10.0.1	drive.novox.be.internal	drive.novox.be
+10.10.0.1	umami.novox.be.internal	umami.novox.be
+```
+
+The suffixed one resolves and is served by nothing. The proxy refuses it during the handshake, and
+says so exactly:
+
+```
+http: TLS handshake error from 10.10.0.3:33480:
+  no public route for "keycloak.novox.be.internal" in this mesh, so no certificate is asked for
+```
+
+A client sees `curl: (35) TLS connect error ... tlsv1 alert internal error` and no peer certificate —
+a server-side refusal, with nothing in it to say the name was never real.
+
+The name the proxy does serve is `<label>.<node>.internal` — `keycloak.novox.internal` — which is what
+[issue 139](../139-an-internal-route-name-resolves-to-the-consumers-node/00-report.md) describes. So
+there are two internal shapes for one service, one of them composed by appending the suffix to a name
+that already has a domain.
+
+## Why it matters
+
+**It sends a reader to the wrong diagnosis.** Reproducing
+[issue 129](../129-nothing-makes-a-machine-trust-the-meshs-authority/00-report.md) on 2026-09-30, the
+first three names tried came from the hosts file, all failed with a TLS alert rather than the
+verification error 129 reports, and the evidence pointed at the proxy having lost its internal
+certificates — a regression that had not happened. The correct name reproduces 129 exactly. Several
+minutes went into a fault that did not exist, and the only thing that distinguished the two was reading
+the proxy's own log.
+
+It is also a name in every machine's hosts file, and in every container's, that cannot be reached: the
+shape the mesh is otherwise careful about — writing a name that resolves to something that does not
+answer is worse than not writing it, because a connection to an address that does not answer hangs
+where a name that does not resolve fails at once ([ADR 0007](../../02-DECISIONS/0007-connectivity.md),
+and the same reasoning in `namesInTheMesh`).
+
+## Where to look
+
+The roster template renders one entry per name as `{{.FQDN}} {{.Name}}`, and `FQDN` is composed by
+appending the mesh suffix to the bare name. For a node that is right — `novox` becomes
+`novox.internal`. For a routed name the bare name is already a fully qualified public name, so the
+composition produces `keycloak.novox.be.internal`, which is not a name anything was told to serve.
+
+The fix is a judgement about what a routed name's internal form is, and 139 is the record that asks it;
+this one is the evidence that the current answer publishes a third thing that is neither.
+
+## Resolved (2026-09-30)
+
+The judgement 139 asked for is [ADR 0151](../../02-DECISIONS/0151-a-routes-internal-name-is-composed-under-the-node-that-serves-it.md):
+a routed name has no mesh form. The roster now publishes it as itself, once, at the serving node's
+address; the `<domain>.internal` line is gone from every machine's hosts file, and a controller test
+refuses it coming back.

04-ISSUES/158-the-proxy-re-reads-and-re-logs-every-route-every-two-seconds/00-report.md

@@ -0,0 +1,49 @@
+---
+status: open
+opened: 2026-09-30
+located-in: []
+fixed-by:
+amended-design:
+---
+
+# 158 — The proxy re-reads and re-logs every route it serves, every two seconds
+
+## What was observed
+
+The route proxy on the control node logs the whole of what it serves about every two seconds —
+measured 2026-09-30, **31 times in sixty seconds**, each line naming all 52 routes:
+
+```
+23:27:19 serving 52 route(s): autoconfig.novox.be, autoconfig.novox.internal, … www.praktijkdespiegel.be
+23:27:21 serving 52 route(s): autoconfig.novox.be, autoconfig.novox.internal, … www.praktijkdespiegel.be
+23:27:23 serving 52 route(s): autoconfig.novox.be, autoconfig.novox.internal, … www.praktijkdespiegel.be
+```
+
+Nothing is changing. The route set is identical every time.
+
+## Why it matters
+
+**It buries the only line that matters.** Between two of those entries sits the one error that explained
+a failing name:
+
+```
+23:27:23 http: TLS handshake error from 10.10.0.3:33480:
+  no public route for "keycloak.novox.be.internal" in this mesh, so no certificate is asked for
+```
+
+One line of signal to roughly 5,000 characters of repetition, and the diagnosis it belonged to
+([issue 157](../157-a-routed-names-internal-alias-is-served-by-nothing/00-report.md)) was found by
+grepping past it. A log that says the same true thing every two seconds is a log nobody reads, which is
+the same failure as one that says nothing — and it is the mesh's own accuracy rule pointed the other
+way: a report that is loud about the unchanging is not reporting.
+
+Whether the *re-read* is also wasteful is secondary and unmeasured — it may be a cheap file stat. The
+logging is not in question.
+
+## Where to look
+
+Not localised. The proxy is `mesh-controller examples/route-proxy`; whether it re-reads on a timer or on
+a file watch, and whether it logs unconditionally or only on change, is the first thing to read. Saying
+what changed — or saying nothing — is the behaviour wanted, and the mesh already has the rule written
+down for its own reports: a log that is quiet on success and loud on failure reads as broken when it is
+working, and one that is loud always reads as nothing.

04-ISSUES/159-an-artifacts-system-is-checked-and-then-ignored/00-report.md

@@ -0,0 +1,100 @@
+---
+status: located
+opened: 2026-09-30
+located-in:
+  - mesh-controller internal/catalogue/build.go (System is validated and read by nothing else)
+  - mesh-controller internal/builder (the compile invocation names no target)
+fixed-by:
+amended-design:
+---
+
+# 159 — An artifact's system is checked, and then nothing uses it
+
+## What was observed
+
+Asked whether the host is built for more than one architecture, 2026-09-30, having just built it
+through the new Go toolchain.
+
+[ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md) says an
+artifact declares what it targets, and that one artifact per target is one build each. The manifest
+layer enforces the first half strictly: a bundle in a language that compiles to a binary **must** name
+a system, must name one of `alpine`, `android`, `arch`, and must not name one at all if its language is
+interpreted. A manifest that gets any of that wrong is refused with a reason.
+
+**The field is then read by nothing.** Every use of it in the control plane is in the function that
+validates it. It does not reach the compiler, no machine is matched against it, and nothing chooses
+between two artifacts by it.
+
+So the compile runs with no target named and produces a binary for whatever the build machine happens
+to be. The host, declared `system: arch` and built on this mesh's only build machine:
+
+```
+mesh-host: ELF 64-bit LSB executable, x86-64, statically linked, stripped
+```
+
+Correct for every machine in this mesh, which are all x86-64 Arch — and correct by coincidence rather
+than by anything the declaration did.
+
+## Why it matters
+
+**A module declaring two systems would get two identical binaries.** Both would be published, both
+pinned, both delivered, and the one sent to the machine it was not built for would fail at exec with a
+message about a format — which is the shape ADR 0005's link-time pin exists to prevent, arriving
+because the pin was never applied.
+
+`android` in the list is the sharp end: it is not an x86-64 platform, and an artifact declared for it
+today would be an x86-64 binary wearing the label. Nothing would say so until a machine tried to run
+it.
+
+**And the field reads as implemented.** It is required, validated against a closed list, and refused
+with a careful message — every signal a manifest author gets says the mesh is acting on it. A field
+that is checked and ignored is worse than one that does not exist, because the check is what persuades
+you it works.
+
+## Two things this is not
+
+- **Not the same axis as the distribution.** `alpine`, `android`, `arch` are what a machine reports
+  itself to be, and the comment on the list says why: "the difference between two of them is a C
+  library, not a kernel". The processor is a second dimension and the manifest has no word for it at
+  all — so even a correct implementation of the current field would not answer the question that
+  started this.
+- **Not urgent for this mesh.** Four machines, all x86-64 Arch, one build machine. Nothing is broken
+  today and nothing will be until a machine differs — which is exactly how long a field like this stays
+  invisible.
+
+## The machine already says what it is
+
+Found while filing [issue 160](../160-a-machine-says-little-about-itself-and-only-when-asked/00-report.md):
+every machine reports its architecture and kernel in the same profile that carries its capabilities, and
+the mesh keeps them.
+
+```
+ ace    | amd64 | linux
+ g14    | amd64 | linux
+ novox  | amd64 | linux
+ shanks | amd64 | linux
+```
+
+Nothing in the control plane reads either, and `node show` prints the capabilities beside them without
+printing them. **So a fix does not need a new fact from the machine** — matching an artifact's declared
+system against what a machine reported is possible today, and the missing piece is only the comparison
+and a compiler told what to target.
+
+## Where to look
+
+The compile invocation is assembled in `mesh-controller internal/builder`, and for Go it would need
+`GOOS`/`GOARCH` set from the artifact rather than inherited from the build machine. That needs the
+manifest to carry a processor as well as a system, or the systems list to mean both — which is a
+decision, not a fix, and belongs with whoever answers whether one static binary should serve several
+distributions at all.
+
+**That last question is live.** The Go toolchain builds statically, so a single binary has no C library
+to differ about and would in fact run on Alpine and Arch alike. The per-system pin is then a policy — a
+host refuses a machine it was not built for — rather than a technical necessity, and worth knowing is a
+choice.
+
+## How a fix is checked
+
+An artifact declared for a system the build machine is not produces a binary for that system, shown by
+reading the file rather than by the build reporting success; and two artifacts declared for two systems
+do not have the same digest.

04-ISSUES/160-a-machine-says-little-about-itself-and-only-when-asked/00-report.md

@@ -0,0 +1,92 @@
+---
+status: open
+opened: 2026-09-30
+located-in:
+  - mesh-host internal/profile (what a machine collects about itself)
+  - mesh-controller internal/inventory (what the mesh keeps of it)
+  - mesh-controller cmd/mesh-controller (node show, which displays a part of it)
+fixed-by:
+amended-design:
+---
+
+# 160 — A machine says little about itself, and only when the mesh asks it something
+
+*Filed as a to-do rather than a fault: nothing is broken by it today. `open` is the status this
+repository has for "written down, nobody has started" — there is no `todo`.*
+
+## What is wanted
+
+**When a machine joins, the mesh should collect as much as it reasonably can about it**, and refresh
+that daily or thereabouts. It already asks what the machine *can do*; what it is made of is the same
+question one level down, and the mesh has no habit of asking it.
+
+## What is already collected, which is more than it looks
+
+A machine reports, and the mesh keeps:
+
+| | |
+|---|---|
+| eight capabilities | container-runtime, firewall, graphical-session, overlay, package-manager, privileged, seat, service-manager — each with a version or a reason it is absent |
+| **its architecture and kernel** | in the same profile, beside the capabilities |
+| which links face outside it | read from its own routing table on every apply |
+| the version of the host running on it | added 2026-09-30 |
+| what it found and is holding | on an adopted machine |
+| what is reachable on it | every listening socket and published port, on an adopted machine |
+| the firewall it was found with, and the tunnel it carried | on an adopted machine |
+
+**The architecture and the kernel are already there and nothing reads them.** Measured on the mesh:
+
+```
+ ace    | amd64 | linux
+ g14    | amd64 | linux
+ novox  | amd64 | linux
+ shanks | amd64 | linux
+```
+
+`node show` prints the capabilities and not these two. No code in the control plane reads either.
+
+## What is missing
+
+**The facts.** Nothing is collected about memory, disk, the processor beyond its architecture, the
+distribution and its version, whether the machine is virtual or physical, its uptime, its timezone, how
+many cores it has. Several of those are what somebody actually wants when deciding where a module
+should go, and the comment on module placement already imagines them: *"`seat: card1-DP-1`, an
+architecture, an amount of memory"*.
+
+**The refresh.** A machine publishes what it knows about itself **after an apply**, and the five-minute
+reconcile publishes nothing. So the mesh's picture of a machine is as old as the last time it sent that
+machine something — found the same day while checking host versions, where three machines read as
+having said nothing until each was pushed
+([issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/01-resolution.md)). A daily refresh is
+the same missing mechanism: a machine saying something on its own schedule rather than only when spoken
+to.
+
+**Somewhere to read it.** Whatever is collected has to be visible, or it joins the architecture in
+being true and unread.
+
+## The pattern this is the third instance of
+
+Three times in one day the mesh was found to be collecting something and reading it nowhere:
+
+- what an adopted machine holds — reported since the adoption work, and no surface counted it until
+  [issue 125](../125-a-hold-is-not-a-line-in-the-apply-report/01-resolution.md);
+- the host version — reported for a week, and the control plane's own copy of the report did not have
+  the field ([issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md));
+- the architecture and kernel — collected, stored, read by nothing, found today by being asked whether
+  the host is built for more than one processor
+  ([issue 159](../159-an-artifacts-system-is-checked-and-then-ignored/00-report.md)).
+
+**Collecting is the easy half and it is the half that gets done.** Whatever this issue adds should say,
+in the same breath, which surface shows it — or it will be the fourth.
+
+## Why it is not urgent
+
+Every machine in this mesh is `amd64` and every one reports so. One architecture is enough for now, and
+that is a decision rather than an oversight: a second processor is a second build of every component,
+and nothing needs one.
+
+## How a fix is checked
+
+A machine's own account of itself is visible in one place; it names at least what is listed as missing
+above; the newest of it is no older than a day on a machine nobody has pushed to; and a machine that
+cannot determine one of them says so rather than reporting a zero.

04-ISSUES/161-a-delivered-host-carries-none-of-its-link-time-facts/00-report.md

@@ -0,0 +1,95 @@
+---
+status: resolved
+opened: 2026-09-30
+located-in:
+  - mesh-host cmd/mesh-host/main.go (version and builtFor, both set at link time)
+  - mesh-controller internal/builder (the toolchain, which deliberately takes nothing from the module)
+fixed-by: mesh-controller (the system stamp, and one linker flag rather than two), mesh-host (the version read from the path) — verified on a machine 2026-09-30, 01-resolution.md
+amended-design:
+---
+
+# 161 — A host the mesh built carries none of the facts its Makefile stamps in
+
+## What was observed
+
+*2026-09-30, on the workstation, having just made the host self-updating.*
+
+The mesh compiled the host, published it, delivered it and the launcher started it. It ran, read the
+machine correctly, and **would have refused the first declaration it was asked to apply.**
+
+The host's own Makefile links in two facts:
+
+```
+LDFLAGS := -s -w -X main.builtFor=$(SYSTEM) -X main.version=$(VERSION)
+```
+
+The mesh's Go toolchain links in neither, on purpose: a toolchain accepts nothing from the module,
+because anything a module could override there it would be writing a Dockerfile to override
+([ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md)). So a
+delivered host has `builtFor = ""` and `version = "development build"`.
+
+**`builtFor` empty is the one that bites.** Before applying anything, the host asks which system it
+was built for:
+
+```go
+sys, err := system.For(builtFor)
+```
+
+and that answers, for an empty name:
+
+```
+this host was built for "", which is not a system it knows. Built hosts are: …
+```
+
+It is called before any resource is applied, so the failure is in the safe direction — the machine is
+not half-configured. It is still a host that cannot do its job, and nothing about it looks wrong: the
+unit is active, the link to the bus is up, and the log says it is hearing what the node should be.
+
+Measured: after the crossover the machine logged nothing further, where the previous host had written
+a reconcile line every five minutes.
+
+## Why this was found rather than reported
+
+Nothing reports it. The host does not check its own stamps at start, the mesh does not ask, and the
+declaration that would fail is the same declaration that would deliver a fix — so **a machine in this
+state cannot be repaired by the mesh.** It was restored by moving the delivered versions aside and
+letting the launcher fall back to the hand-placed binary, which is the fallback working exactly as
+designed.
+
+## What the records already say about half of it
+
+[ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md) settles the
+version and its answer is not implemented:
+
+> A component's version comes from where it sits, not from its linker. It is unpacked into a directory
+> named for its version, so it can read its own version from its path. The stamp goes, and with it the
+> need for a build to know what it will be called.
+
+That is exactly right and would also fix what the mesh reports: a delivered host would say
+`637f65559d16` rather than `development build`, and
+[issue 087](../087-the-controller-cannot-tell-a-host-is-too-old/00-report.md)'s host comparison would
+mean something for delivered hosts.
+
+**The system pin has no answer yet**, and it needs one before any mesh-built host can apply anything.
+The tension is real: the target is a property of the artifact and 0142 says so, but a toolchain that
+passed it would be linking a value into a variable whose name belongs to the module — which is the
+coupling the toolchain exists to avoid. Candidates, none decided:
+
+- the path carries it as well as the version, so the host reads both from where it sits, as 0142 does
+  for the version;
+- the bundle carries a small file beside the binary saying what it was built for, written by the
+  builder from the artifact's declaration;
+- the host stops being pinned at link time and refuses on a fact it reads from the machine instead —
+  which changes what [ADR 0005](../../02-DECISIONS/0005-the-node-host.md) decided and is the biggest of
+  the three.
+
+## What is true in the meantime
+
+Self-update works end to end and is one fact short of usable: the mesh builds the host, publishes it,
+delivers it to a machine, the running host stands aside, and the launcher starts the delivered one. The
+machine is left on its hand-placed binary until this is answered, which is one command to undo.
+
+## How a fix is checked
+
+A host the mesh built and delivered applies a declaration on a machine, shown by the machine's own
+reconcile line; and it reports a version that names the build it came from rather than a placeholder.

04-ISSUES/161-a-delivered-host-carries-none-of-its-link-time-facts/01-resolution.md

@@ -0,0 +1,59 @@
+# 161 — resolved: a host the mesh built runs a machine
+
+*2026-09-30. Measured on the workstation.*
+
+```
+running: /usr/lib/nox-mesh-host/versions/093231796eb0/nox-mesh-host
+agent:   active
+reconciles in the last six minutes: 1
+
+mesh-controller node show shanks
+  host              093231796eb0
+```
+
+A binary the mesh compiled, published to its own registry, delivered over the bus, started by the
+launcher, applying declarations, and reporting a version that names the build it came from.
+
+## The two facts, and where each now comes from
+
+**The system it was built for comes from the artifact.** ADR 0142 already made the target a property
+of the artifact rather than of the recipe, so the toolchain names the variable it fills and the
+artifact supplies the value. It is the one thing a toolchain takes from a module, and it is stated
+rather than inferred.
+
+**The version comes from where the binary sits**, which is what
+[ADR 0142](../../02-DECISIONS/0142-the-mesh-delivers-its-own-components-as-binaries.md) decided and
+nothing had implemented: a delivered host reads the directory it was unpacked into. A host placed by
+hand keeps its link-time stamp, which is the honest answer for one the mesh did not deliver — and is
+every other machine today.
+
+## Two mistakes on the way, both found by reading the output
+
+**A repeated flag is not a merged one.** The stamp was appended as a second `-ldflags`, and the Go
+command takes the last and drops the first. The binary gained its system and lost `-s -w`: 12.2MB
+against 8.5MB, with its debug info. The comment I had written said the linker "accepts and merges"
+them. It does not. Linker flags are the toolchain's own list now, composed into one flag, and a test
+refuses a compile line that carries `-ldflags` itself.
+
+**The delivered binary was named after its package.** `cmd/mesh-host` builds `mesh-host`; every
+machine runs `nox-mesh-host`, which is what the launcher looks for inside a version. The first
+delivery landed, reported `created … 1 file(s)`, and was invisible. An artifact says what its
+executable is called now.
+
+Both were caught by listing the directory and reading the binary rather than believing the line that
+said it worked.
+
+## What this cost while it was wrong, and what saved it
+
+A delivered host that cannot apply is a machine the mesh cannot repair, because the declaration that
+would fix it is the declaration it cannot apply. The workstation was restored by moving the delivered
+versions aside so the launcher fell back to the hand-placed binary — **the fallback in the launcher,
+working exactly as designed**, and the reason this was an inconvenience rather than an expedition.
+
+It also loops if you are not careful: the working binary applies, delivers a version, stands aside,
+and the broken one starts. Stopping the unit while the fix was built was the way through.
+
+## What is left
+
+**Three machines still run a hand-placed host.** Rolling them forward is one assignment and one push
+each, and the control node is worth doing last and watching.

04-ISSUES/162-an-archive-cannot-be-undeclared/00-report.md

@@ -0,0 +1,56 @@
+---
+status: located
+opened: 2026-09-30
+located-in: [mesh-host internal/apply (no removal for an archive)]
+fixed-by:
+amended-design:
+---
+
+# 162 — An archive cannot be undeclared, and trying stops the machine applying anything
+
+## What was observed
+
+*2026-09-30, unassigning the host module from the workstation to undo a delivery.*
+
+```
+holding this machine: 0 applied, and map[apply:applying "mesh-host.next": no way to remove a "archive"
+0 resource(s) were applied and remain; everything was attempted, so what is not listed as failed was done.
+```
+
+**Nothing was applied at all** — not the archive, not the other forty resources that had nothing to do
+with it. The machine stopped reconciling and stayed that way until the module was assigned again.
+
+## Why it matters
+
+Every other resource kind can be taken away. A file is removed and what was found under it is put
+back; a container is stopped and removed; a unit is given back the state it was found in
+([ADR 0118](../../02-DECISIONS/0118-undeclaring-gives-a-unit-back-the-state-it-was-found-in.md)). An
+archive has no removal at all, so:
+
+- **a module with an archive can never be unassigned** — the attempt fails for ever;
+- **the failure takes the whole apply with it**, so the machine applies nothing else either, and one
+  unassignable resource is a machine frozen against every other change;
+- it is silent in the mesh's terms: the push reported sent, and only the machine's own journal says
+  what happened.
+
+The host module is the obvious case and not the only one. An archive is for what inlining cannot
+serve — a theme, an icon set, a tree of configuration — and any module using one is in the same
+position.
+
+## What the right answer probably is, and the question in it
+
+The other kinds answer this by remembering what they found. An archive unpacks many files into a
+directory the mesh did not necessarily create, so removal has a real question in it: **remove what the
+archive put there, or remove the directory?** The first needs the applier to have recorded the file
+list; the second would delete whatever else lives there — and for the host's own versions directory,
+that is every other delivered version.
+
+Recording what was unpacked is the answer that matches how the rest of the host behaves, and it is
+what [issue 126](../126-a-volume-path-is-not-in-the-spec-comparison/00-report.md) and ADR 0118 already
+argue for elsewhere: the mesh gives back what it found.
+
+## How a fix is checked
+
+A module with an archive is assigned, pushed, unassigned and pushed again; what the archive put on the
+machine is gone, anything that was in the directory beforehand is still there, and the apply that
+removed it applied everything else in the same declaration.

04-ISSUES/163-a-delivered-host-stood-aside-on-every-push-and-reported-nothing/00-report.md

@@ -0,0 +1,56 @@
+---
+status: resolved
+opened: 2026-09-30
+located-in: [mesh-host cmd/mesh-host/main.go (the successor check after an apply)]
+fixed-by: mesh-host PR 58 — the check asks with the running version, read from the binary's path, not the link-time stamp
+amended-design:
+---
+
+# 163 — A delivered host stood aside on every push, and reported nothing
+
+## What was observed
+
+*2026-09-30, rolling the mesh-built host onto the last two machines.*
+
+Every push to a machine running a delivered host produced, in order:
+
+```
+host 093231796eb0 is delivered; standing aside so the launcher runs it
+applied 333 resource(s)
+applied, and could not tell the mesh: reporting: context canceled
+nox-mesh-host-launch: the host exited cleanly; starting it again
+nox-mesh-host-launch: running /usr/lib/nox-mesh-host/versions/093231796eb0/nox-mesh-host
+```
+
+— for the version it was **already running**. It restarted itself on every push, for ever, and the mesh
+never received a single report from it: `node show` kept the version from before the crossover, and
+the operator's push waited its full three minutes for an answer that was never coming.
+
+Read as healthy throughout: unit active, bus link up, "hearing what this node should be".
+
+## Why
+
+After an apply the host asks whether a newer host has been delivered than the one running, and the
+question was asked with the **link-time version stamp**. Since
+[issue 161](../161-a-delivered-host-carries-none-of-its-link-time-facts/01-resolution.md) a delivered
+host's version comes from where it sits and its stamp is `development build` — so the comparison never
+matched the newest delivered version, and "a newer host is waiting" was always true.
+
+Standing aside cancels the context the report is published with, so the report was lost on every one
+of those applies. Two faults from one wrong argument.
+
+The change that moved the version to the path was applied to the report and to the known-good record,
+and not here. Half a change, and the half left behind was the one that decides whether to exit.
+
+## Why the three-minute wait made it invisible
+
+The push's `--wait` timing out read as *slow*. It was not slow: **the report was never going to arrive.**
+The operator put it exactly: *if you don't get a response in five seconds, something is wrong.* A wait
+long enough to absorb a machine's whole apply is a wait long enough to hide that the machine never
+answered.
+
+## How it is checked
+
+A machine running a delivered host is pushed a declaration that delivers nothing new; it applies,
+reports, and does not stand aside. A machine running a delivered host is pushed a genuinely newer
+version; it stands aside once, and the next push it does not.

04-ISSUES/171-a-modules-own-resolver-knows-no-mesh-name/00-report.md

@@ -0,0 +1,78 @@
+---
+status: resolved
+opened: 2026-09-30
+located-in:
+  - mesh-catalog modules/mailu (eight containers name Mailu's own resolver, and one of them binds a mesh name)
+  - mesh-catalog modules/dnsmasq (dropped the DNSSEC bit its upstreams set)
+fixed-by: mesh-catalog PR 178 (mailu-admin uses the machine's resolver) and PR 179 (the machine's resolver passes the DNSSEC bit down) — 2026-09-30, the same afternoon
+amended-design:
+---
+
+# 171 — A module that names its own resolver knows no mesh name
+
+## What was observed
+
+The afternoon [ADR 0148](../../02-DECISIONS/0148-the-meshs-names-are-resolved-not-copied-into-containers.md)
+landed — no container is given the mesh's names any more; it asks the machine's resolver — the mail
+system's admin container began logging, 523 times in three minutes:
+
+```
+psycopg2.OperationalError: could not translate host name "novox.internal" to address: Name does not resolve
+```
+
+Mail was accepted on every port and the web front answered; the admin and the spam filter beside it
+were unhealthy, and anything that needed the database — a mailbox change through the API, the spam
+filter's domain list — failed. Found by the operator asking whether mail was back, forty minutes in.
+
+Mailu ships its own resolver, an unbound in a container, and every other Mailu container is told to
+use it — the module carries `dns: [192.168.203.254]` on eight containers. That resolver recurses from the
+root and knows nothing under `.internal`. Until that afternoon the admin container had the database's
+name anyway, because the mesh wrote every name into every container at creation; the copy was the only
+reason a container behind its own resolver could reach anything by a mesh name, and nobody knew it was
+load-bearing.
+
+**Removing the override was not enough.** Given the machine's resolver instead, the admin refused to
+start: `Your DNS resolver at 127.0.0.11 isn't doing DNSSEC validation`. Mailu checks, at start, that
+its resolver returns the Authenticated Data bit for a signed name. The mesh's resolver forwards to two
+upstreams that validate and set the bit, and dropped it on the way down — dnsmasq does unless told
+otherwise. Mailu's own unbound has no hook to forward a zone elsewhere, so it could not be taught the
+mesh's names either.
+
+## Why it matters beyond this instance
+
+**A container with a resolver of its own has opted out of the machine's, and nothing says so.** 0148
+made the machine's resolver load-bearing for every container; a `dns` on a container is a quiet
+exception to that, and the exception used to be papered over by the copy the record removed. The
+manifest field reads like a preference and is a decision about whether mesh names exist inside the
+container.
+
+**A resolver that forwards to validating upstreams and hides the fact is less useful than it could
+be**, and the first program to check found out.
+
+**The mesh reported nothing.** Every container ran; the failing one accepted connections; the report
+was about bytes. It is [issue 145](../145-a-machine-reads-healthy-while-its-modules-cannot-reach-each-other/00-report.md)
+again, and the check that would have caught it is the same unbuilt one.
+
+## What was done
+
+- The one Mailu container that binds a mesh name — the admin, through the database it is granted —
+  no longer names Mailu's resolver and uses the machine's, like every container without a `dns` of
+  its own (mesh-catalog PR 178). The other seven keep unbound: the spam filter needs a validating
+  resolver for its blocklist lookups, and none of them asks for a mesh name.
+- The machine's resolver passes the DNSSEC bit down from its upstreams, `proxy-dnssec` (PR 179). It
+  does not validate itself; the trust is the upstream's and the path to it, as a forwarding resolver's
+  always was, and the configuration says so.
+
+## What checks it
+
+The admin container's own start-up check, which is what failed, and the mesh's status once it reads
+healthy. A container-level check that a mesh name resolves from inside every declared container is
+the one 110 and 145 both ask for and is not built.
+
+## Open questions
+
+- Should a container's `dns` be refused, or made to say what it gives up? A module that names its
+  own resolver and binds a mesh name is a contradiction the controller can see at composition — the
+  grant hands it a name its resolver will not answer.
+- Should the machine's resolver validate rather than proxy? It would cost a trust anchor on every
+  machine and make the resolver slower to start; proxying was enough for the one program that asked.

04-ISSUES/172-the-ssh-client-block-matches-one-spelling-of-a-machine/00-report.md

@@ -0,0 +1,55 @@
+---
+status: located
+opened: 2026-09-30
+located-in:
+  - the predecessor's terminal module (still generating the operator's ssh client blocks on every workstation)
+  - mesh-controller internal/catalogue (the ssh-client roster, tested and not yet a catalogue module)
+fixed-by:
+amended-design:
+---
+
+# 172 — The ssh client block for a machine matches one spelling of its name, and the other gets the wrong user
+
+## What was observed
+
+On a workstation, 2026-09-30, reported by the operator. `ssh home-server` logs in; `ssh
+home-server.internal` is refused with `Permission denied (publickey)`. The operator expected the
+opposite, if either: the full mesh name is the one the resolver serves.
+
+The name is not the fault. Both spellings resolve to the machine's private address — the mesh's roster
+region in the hosts file carries `<node>.internal <node>` on one line, and the resolver answers
+anything under the node's name. What differs is the login: the generated client configuration has a
+`Host home-server` block naming the account to log in as, and `home-server.internal` matches no block,
+so ssh falls back to the operator's local username, which has no account on that machine. Spelled
+`account@home-server.internal` it works.
+
+The file is the predecessor's. `~/.ssh/config.d/mesh` says in its own header that it is generated by
+the predecessor's terminal module, which only ever wrote the bare name. The mesh's own ssh-client
+roster — every other machine's Host block, written as a marked region of the operator's `~/.ssh/config`
+with the account the mesh knows for that machine (to-be 29) — already matches both spellings, and a
+controller test holds `Host marge marge.internal`. It is composed and tested in the controller and is
+not a module in the catalogue, so no machine receives it; every workstation still runs the
+predecessor's generator.
+
+## Why it matters beyond this instance
+
+**A name the mesh serves and a name a person can use are not the same set**, and the difference is
+silent. The resolver, the hosts file and the certificate authority all treat `<node>.internal` as the
+machine's name; the one file that decides who you log in as does not know it. A person who learns the
+mesh's name from `status` or from a certificate and types it is refused with an error that says
+nothing about a missing Host block.
+
+**It is the migration story for the operator's own tooling, arriving as a symptom.** The mesh has the
+right file and does not ship it. Until the ssh-client roster is a module and is assigned to the
+workstations, the predecessor's generator keeps writing a file the mesh has already superseded, and
+every such file is one the mesh cannot correct.
+
+## Open questions
+
+- Should the ssh-client roster become a catalogue module now, assigned to every workstation, and take
+  the predecessor's `config.d/mesh` out of the operator's `Include`? Its content is settled; what is
+  not is the takeover of a file in a person's home that another generator still writes.
+- Should the block match a third spelling — the machine's public name, where it has one — or is that
+  a different key and a different account?
+- What checks it? A controller test holds the two spellings; nothing checks that the file a workstation
+  actually has is the mesh's rather than the predecessor's.

README.md

@@ -87,12 +87,15 @@ rather than location** — that these documents would be indexed into the knowle
 symptom search returns them beside everything else. One source, many surfaces. Where the source
 is authored is then a separate question.
 
-**That indexing does not exist.** It was checked on 2026-08-23 and returns nothing; it appears
+**That indexing never existed, and the store it would have indexed into is gone.** It was checked
-never to have existed. Until it does, the objection stands unanswered and this repository is
+on 2026-08-23 and returned nothing; the knowledge base it named was the predecessor's, and since the
-the fourth knowledge system it was argued not to be. Recorded as
+mesh moved to its own bus on 2026-09-28 nothing can reach it at all. The claim is recorded as
-[`04-ISSUES/006`](04-ISSUES/006-hq-is-not-indexed-into-the-knowledge-base/00-report.md), and
+[`04-ISSUES/006`](04-ISSUES/006-hq-is-not-indexed-into-the-knowledge-base/00-report.md) and answered
-left standing here rather than quietly reworded, because a claim that held up a decision and
+by [ADR 0025](02-DECISIONS/0025-the-design-record-is-read-not-copied.md): these documents are
-was never checked is precisely the failure this repository exists to name.
+**read, not copied** — an agent reads this repository and a search consults it — and that agent is
+not built. So today this repository is reachable by whoever knows to open it and surfaces to nobody
+else. Said here rather than quietly reworded, because a claim that held up a decision and was never
+checked is precisely the failure this repository exists to name.
 
 Answered separately, a repository of its own is the better home: