Compare commits

..
Author SHA1 Message Date
jschoubben 80df1b6361 Pin mesh-sdk 6e4a301, the release's final sdk (issue 774)
The sdk's conformance events are unchanged at 6e4a301 and captured so. The whole integration suite passes at this head
against a local store and bus: 1706 tests, 8 skipped for a registry, a container, a refusing bus or a lab's run.
2026-10-30 00:20:10 +01:00
jschoubben 816cf96459 Pin mesh-sdk 007b6fd, the release's final sdk, so the controller reads the release's manifests and composes as the group does (issue 774)
The sdk now refuses an offer that says no credential form (6b8c694, ADR 0570). So:
- the catalogue copy in testdata/beside is captured again at the release branch's catalogue 8ae8e88d, whose offers each
  say one, and the sdk's conformance events at 007b6fd, as CAPTURED asks;
- the fixtures of identity_bound_test and issued_test say the form their rows meant: pair for a login of each
  consumer's own, none for a keyless provision. No behaviour of the controller changed.
2026-10-30 00:07:04 +01:00
jschoubben 3f05df5d1a Pin mesh-sdk 2e5cc4f, so the controller reads the release's new manifest words (issue 774)
The controller on dbcaa29 parses manifests strictly and refuses `bus` on an artifact and `max-age-days` on a log, both
in the re-land group's catalogue. The SDK's conformance events are captured again at the pinned commit, as
testdata/beside/CAPTURED asks; they are unchanged.
2026-10-29 22:19:39 +01:00
mesh-tools e04846e9fa Merge pull request 'Pin mesh-sdk dbcaa29, so the controller composes with the group fix/1960-no-hold-spread's SDK (issue 1960)' (#480) from fix/1960-no-hold-spread into main 2026-10-29 15:35:31 +00:00
jschoubben 020c6a99b0 Pin mesh-sdk dbcaa29 and capture its conformance events there, so the controller composes with the group fix/1960-no-hold-spread's SDK (issue 1960)
mesh/delivery merged: source.main moved to novox/mesh-controller@e04846e9fa49
mesh/delivery-group group fix/1960-no-hold-spread merged: every member is merged
mesh/check pass 020c6a99b035 on 5a9766d4a47a | unit 3318 run, 1951 kept, 38 skipped | integration 3994 run, 2125 kept, 38 skipped | over bound: unit, …
2026-10-29 16:18:13 +01:00
mesh-tools 5a9766d4a4 Merge pull request 'Pin mesh-sdk 38dbe1d, so the controller knows the seat mesh-placement (issue 1126, delivery 2 of 2)' (#479) from feat/1126-placement-seat into main 2026-10-28 18:33:05 +00:00
jschoubben f46a08772f Pin mesh-sdk 38dbe1d, the group's head, so the controller knows the seat mesh-placement its holder claims (ADR 0567, issue 1126)
mesh/delivery merged: source.main moved to novox/mesh-controller@5a9766d4a47a
mesh/delivery-group group feat/1126-placement-seat merged: every member is merged
mesh/check failed: integration: meshbuildgo/cmd/mesh-build-go.TestAComposeAskIsBuiltOnceAndAnsweredOnTheBus: the compose ask mvsg85b34d25b4 never ende…
2026-10-28 18:49:02 +01:00
jschoubben 1c2adfdca9 Build with the group's mesh-sdk, so the controller knows the seat mesh-placement its holder claims (ADR 0567, issue 1126) 2026-10-28 18:14:58 +01:00
mesh-tools 0fbb5ef359 Merge pull request 'The controller gives its lease back at the stop signal and says a takeover once, so a controller delivery raises no urgent alarm (issue 906, ADR 0558)' (#478) from fix/906-lease-release into main 2026-10-27 00:07:46 +00:00
jschoubben 8f84731896 Name the epoch given back in the lease's delete marker, so only a holder's own release counts as a handover (issue 906, ADR 0558 §3, review round 1)
mesh/delivery merged: source.main moved to novox/mesh-controller@0fbb5ef35991
mesh/check being carried to 8f847318963c on 0fbb5ef35991 | from 2dafbe60e939
A plain delete marker on the lease key is also written by a command's own
lease and by a take given back over the floor; counted as a handover, it
would record a crashed holder's epoch released and say nothing. Release now
writes the marker itself, compare-and-set on the key's revision, with the
header Mesh-Lease-Released naming its epoch, and the next holder ends only
that epoch as released. A takeover whose predecessor cannot be read is now
raised as an expiry, not only printed.
2026-10-27 00:55:30 +01:00
jschoubben fbaa463a38 Give the controller lease back at the stop signal and say a takeover once, so a controller delivery raises no urgent alarm (issue 906, ADR 0558)
The 2026-10-26 replacements showed the lease was given back, but last: the
shutdown took 10 s, the gate had closed at half the key's age, and the release
recorded the epoch released only through the gate, so the next holder found
it expired and S12 said it, urgent, for an hour.

Now the stop signal gives the key back first and records the epoch released
whenever the delete succeeded; a candidate reads the key's delete marker on
the bus to tell a handover from an expiry; and a real takeover (lost or
expired) is raised and cleared once by the controller that took it, instead
of an hour of S12 observation.

Research 072 piece moved: the lease's takeover read from its own key on the
bus and said as an event (03-domains, mesh-controller row; 04-findings 2,
the watchdogs' age reading).
Quick fix weighed: drop the gate check from the release record only; it
leaves the release last in the shutdown and the hour-long alarm for a real
takeover.
2026-10-27 00:49:13 +01:00
mesh-tools 2dafbe60e9 Merge pull request 'Broker tests read a consumer's pending count through pendingOf wherever they expect a count after a publish (issue 1126)' (#477) from fix/1126-tests-wait-for-consumer-counts into main 2026-10-26 22:57:59 +00:00
jschoubben 8d58afea6c Broker tests read a consumer's pending count through pendingOf, which waits up to 10 s, wherever they expect a count after a publish (issue 1126)
mesh/delivery merged: source.main moved to novox/mesh-controller@2dafbe60e939
mesh/check being carried to 8d58afea6cb8 on 2dafbe60e939 | from b611f7375a47
The server counts a consumer's pending after it acknowledges the publish, so a read at once failed the try of
main on a loaded ace (TestAResetOfTheEventsConsumerLeavesTheMergesAndTheirConsumerAlone, 2026-10-26 22:24 UTC).
Fixed for the kind, not the one test: all six reads of a positive count go through the package's own pendingOf,
whose deadline was 1 s.

Research 'The controller, NATS-native': none; this changes tests only, not a part of the controller.
2026-10-26 23:28:03 +01:00
mesh-tools b611f7375a Merge pull request 'Pin mesh-sdk main, whose seats promise own and settle, so the catalogue passes the module check again (issue 1126)' (#476) from fix/1126-controller-pins-sdk-main into main 2026-10-26 22:08:19 +00:00
jschoubben 253bbcea2d Pin mesh-sdk main f178323, whose delivery seat promises settle beside the vault's own, so the catalogue check passes against mesh-catalog main f0f7b673 (issue 1126)
mesh/delivery merged: source.main moved to novox/mesh-controller@b611f7375a47
mesh/check checking 253bbcea2da8 on b611f7375a47
Moves no piece of research 'The controller, NATS-native': a repin of the seat catalogue, no controller code changed.
2026-10-26 23:03:47 +01:00
mesh-tools 2c6382346f Merge pull request 'Pin the mesh-sdk whose vault seat promises own, so the catalogue passes the module check again (issue 1126)' (#475) from fix/1126-vault-seat-own into main 2026-10-26 22:00:46 +00:00
jschoubben b8cafccaf6 Pin mesh-sdk at e78e881, whose vault seat promises own, so the catalogue check passes against mesh-catalog main (issue 1126)
mesh/delivery merged: source.main moved to novox/mesh-controller@2c6382346f48
mesh/delivery-group group fix/1126-vault-seat-own merged: every member is merged
mesh/check failed: unit: github.com/novox/mesh-controller/cmd/mesh-controller.TestModuleCheckPassesTheCatalogue: the catalogue does not pass its own c…
Moves no piece of research 'The controller, NATS-native': a repin for a seat verb list, no controller code changed.
2026-10-26 22:40:34 +01:00
mesh-tools 82d212a2a1 Merge pull request 'The controller answers each node's placement from its own store, the one verb of its exit (issue 1126)' (#474) from feat/1126-nodes-compose into main 2026-10-26 21:11:21 +00:00
jschoubben df8ebf6a77 Pin mesh-sdk 9796444 (l2-next: feat/1126-nodes-compose merged with sdk main #101, the standing-test race fixes) (issue 1126)
mesh/delivery merged: source.main moved to novox/mesh-controller@82d212a2a1ff
mesh/delivery-group group feat/1126-nodes-compose merged: every member is merged
mesh/check checking df8ebf6a77c1 on 82d212a2a1ff | novox/mesh-controller#474 in group feat/1126-nodes-compose
2026-10-26 22:05:41 +01:00
jschoubben dc414fbfef Pin mesh-sdk 5f45175 (l2-next: feat/1126-nodes-compose merged with sdk main) after merging main (issue 1126) 2026-10-26 21:59:21 +01:00
jschoubben fd891688ac Pin mesh-sdk feat/1126-nodes-compose 03dce12, the group's head: the checker refuses a pin that is neither trunk nor the group's head (issue 1126)
mesh/delivery closed: a newer head of the same pull request
mesh/delivery-group group feat/1126-nodes-compose checking: 4 of 6 member(s) ready
mesh/check checking fd891688acdf on 2ecfbb3a29cb | novox/mesh-controller#474 in group feat/1126-nodes-compose
2026-10-26 20:35:49 +01:00
jschoubben 9d8a5a7ede Review of L2, round 2: the placement export is checked within an existing integration test over the same fresh mesh, so it adds no test of its own (the test budget; issue 1126)
mesh/delivery closed: a newer head of the same pull request
mesh/check checking 9d8a5a7ede90 on 2ecfbb3a29cb | novox/mesh-controller#474 in group feat/1126-nodes-compose
2026-10-26 20:23:51 +01:00
jschoubben 4e9e5c27d4 Review of L2: the export's test reads each printed placement back as the file the seed writes, at its version (S1); pin mesh-sdk 3968849, CAPTURED taken again (issue 1126)
mesh/delivery closed: a newer head of the same pull request
mesh/check failed: unit: ..test: ✔ a transient empty list for four passes retires nothing; five stable passes retire, data kept (290.408814ms) | 4e9e5…
2026-10-26 20:12:49 +01:00
jschoubben 7c8a2e03dd Pin mesh-sdk feat/1126-nodes-compose 9713c30: placement carries a settings-form module's values (issue 1126)
mesh/delivery closed: a newer head of the same pull request
mesh/check failed: ace: novox/mesh-catalog#505 does not merge into the trunk: files in conflict: .gitignore | 7c8a2e03dd8e on 2ecfbb3a29cb | unit 1245…
2026-10-26 19:40:43 +01:00
jschoubben 89f52a6504 Pin mesh-sdk 53c2def, the group's head, and take the SDK's conformance copy again at it (issue 1126)
mesh/delivery closed: a newer head of the same pull request
mesh/check failed: unit: git.novox.be/novox/mesh-catalog/checks/nobinaries.TestTheCatalogueHoldsNoCompiledProgram: modules/mesh-config/cmd/mesh-config…
2026-10-26 19:28:39 +01:00
jschoubben fb2f5a2962 The controller answers each node's placement from its own store, read-only, for the seed of the catalogue's placement files: the one verb of its exit (issue 1126, ADR 0555 §6; the operator 2026-10-26 19:06:40) 2026-10-26 19:16:09 +01:00
mesh-tools 2ecfbb3a29 Merge pull request 'The controller takes mesh-sdk a7a1fbd3, so its composer gives build modules the node's CPU share (issue 1806, ADR 0553/0554 change B)' (#473) from fix/controller-sdk-cpu-share into main 2026-10-26 15:41:03 +00:00
jschoubben 9ad8715871 Bump mesh-sdk to a7a1fbd3 so the build seats' runtimes are composed to yield (issue 1806)
mesh/delivery merged: source.main moved to novox/mesh-controller@2ecfbb3a29cb
mesh/check pass 9ad8715871ea on 2ecfbb3a29cb | carried from e1a55f099b4e | unit 1023 run, 171 kept, 2 skipped | integration 1701 run, 214 kept, 9 skip…
The controller is the only composer, and its pin (10b6e202) predates mesh-sdk #95: without the bump the
build runtimes never get yields {priority idle, cpu-share} or MESH_BUILD_CPU_SHARE (ADR 0553, ADR 0554,
change B). The operator said yes to this one change of the frozen controller on 2026-10-26 15:54.

A composition at both pins over every node's assignment differs only in mesh-build-go, mesh-build-node and
mesh-build-image (the yields and the share word); every other composition is byte-identical. The SDK's
conformance capture moves with the pin, as testdata/beside/CAPTURED asks.

Research 072 (the controller, NATS-native): no piece moved; a pin bump the operator allowed.
2026-10-26 16:34:55 +01:00
mesh-tools e1a55f099b Merge pull request 'Seat users pulled by kind may read their own seat's dead letters (issue 734)' (#472) from fix/734-lost-ask into main 2026-10-26 13:40:43 +00:00
jschoubben ccc7ba334c broker: DeadLetterReads grants only what a pulled ordered consumer needs, as mesh-sdk buscompose (issue 734, review round 1)
mesh/delivery merged: source.main moved to novox/mesh-controller@e1a55f099b4e
mesh/delivery-group group fix/734-lost-ask merged: every member is merged
mesh/check checking ccc7ba334c82 on 08a261ac1a6d | novox/mesh-controller#472 in group fix/734-lost-ask
NATS-native research piece: none moved; the copy stays in step with mesh-sdk until mesh-grants composes the list.
2026-10-26 14:17:00 +01:00
jschoubben 65d5ed3367 broker: a user of a seat pulled by kind reads its seat's dead letters, as mesh-sdk buscompose composes it (issue 734)
The checker hears from DEAD_LETTERS which of its asks a build seat gave up on, and ends the try or asks the check again;
the controller composes today's user list, so its copy of the composer carries the grant too. NATS-native research piece:
none moved; the composition itself moves to mesh-grants (ADR 0547), and this keeps the copy in step with mesh-sdk.
2026-10-26 13:51:35 +01:00
mesh-tools 08a261ac1a Merge pull request 'Declarations test: the indexer stand-in answers the trunk value at the trunk's commit alone (issue 1945)' (#471) from fix/declarations-test-stand-in into main 2026-10-26 12:30:18 +00:00
jschoubben 23226384e5 Answer the indexer's trunk value at the trunk's commit alone in the declarations test, so its take-in records mesh-host's build (issue 1945)
mesh/delivery merged: source.main moved to novox/mesh-controller@08a261ac1a6d
mesh/check being carried to 23226384e529 on 08a261ac1a6d | from 350f8ebdfe5f
The stub answered the trunk for every commit, so since registering on the
fingerprint (666d1b68, #465) mesh-host, registered at an older commit, looked
already at the trunk's fingerprint: the take-in only moved its commit, no build
was recorded, and the fingerprints never named the new node-engine. The real
indexer holds no value at a commit it never indexed.

(cherry picked from commit 779533764b)
2026-10-26 13:29:04 +01:00
mesh-tools 350f8ebdfe Merge pull request 'Broker: the bus node may follow the grants part's sealed user list and the grants part may write it, the operator's second one-grant exception (issue 1945)' (#470) from feat/1945-user-list-switch into main 2026-10-26 12:21:44 +00:00
jschoubben 09d2cc5056 Broker: the bus node may follow the grants part's sealed user list and the grants part may write it, the operator's second one-grant exception (issue 1945, ADR 0547 clauses 7-8)
mesh/delivery merged: source.main moved to novox/mesh-controller@350f8ebdfe5f
mesh/delivery-group group feat/1945-user-list-switch merged: every member is merged
mesh/check checking 09d2cc5056ef on 47e64f0284a9 | novox/mesh-controller#470 in group feat/1945-user-list-switch
2026-10-26 13:03:15 +01:00
mesh-tools 47e64f0284 Merge pull request 'Broker: the grants part may create a placed module's declared buckets, logs and streams, creation only (issue 1790; the operator's one-grant exception)' (#469) from fix/1790-grants-create-buckets into main 2026-10-26 08:17:19 +00:00
jschoubben 6927b02d4d Grant the grants part creating the streams manifests declare, so a new module state no longer waits for a controller start
mesh/delivery merged: source.main moved to novox/mesh-controller@47e64f0284a9
mesh/delivery-group group fix/1790-grants-create-buckets merged: every member is merged
mesh/check checking 6927b02d4d58 on fda4859a95b8 | novox/mesh-controller#469 in group fix/1790-grants-create-buckets
The operator's one more grant in the frozen controller's composer (2026-10-26 08:54), beside the
memberships grant of 22:31 (ADR 0547 clause 1 and 5): mesh-grants' principal may publish
$JS.API.STREAM.CREATE.<stream> and $JS.API.STREAM.INFO.<stream> for each bucket
(KV_<module>_<state>), log (LOG_<module>_<log>) and module stream a placed module's manifest
declares, each by its own name, and nothing wider. The writers table lets it create, never update
or delete. Byte-equal with mesh-sdk's buscompose (fix/1790-grants-create-buckets), whose expected
lists this code wrote; the same test file holds both copies. Nothing else changes, and the mesh-sdk
pin stays.

Research 072: moves the grants piece (the grants part split out of the controller) one step, the
creation of a module's declared bus objects; the controller's raise at start stays until it loses
its lease.

Issue: 1790
2026-10-26 09:02:44 +01:00
jschoubben fda4859a95 Merge pull request 'The bus composer grants mesh-grants publishing memberships, nothing more (ADR 0547, the operator's one grant of 2026-10-25 22:31)' (#468) from fix/1945-memberships into main
Reviewed-on: #468
2026-10-25 22:44:04 +00:00
jschoubben a81717e53c Seal the seeded account in the adapter-input test, the second fixture of its kind (issue 1945, PR #467)
mesh/delivery merged: source.main moved to novox/mesh-controller@fda4859a95b8
mesh/delivery-group group fix/1945-memberships merged: every member is merged
mesh/check checking a81717e53cf4 on 7cd4d18bd54e | novox/mesh-controller#468 in group fix/1945-memberships
The kind: a test that seeds a bus account without sealing it, which an account counted as issued only once sealed
(ADR 0543 §2) no longer counts. The whole integration level, run locally, found two: the addresses test (fixed
before) and TestTheAdaptersInputCarriesTheOwnSecretsTheBoxesTheLicencesAndTheBus. Test only.
2026-10-25 23:30:20 +01:00
jschoubben d01839c29c Grant the grants part reading every membership; seal the seeded account in the addresses test (ADR 0547 §5, review; issue 1945)
mesh/delivery-group group fix/1945-memberships rejected: composed together: the checker's verdict at the members' heads (2253ba8bbe9b): integration: github.com…
mesh/delivery closed: a newer head of the same pull request
mesh/check failed: integration: github.com/novox/mesh-controller/cmd/mesh-controller.TestTheAdaptersInputCarriesTheOwnSecretsTheBoxesTheLicencesAndThe…
The grants part reads each module's last membership before it publishes, so it is granted the direct get on every
membership subject beside the publish (review of mesh-grants, B1).

The test TestTheControlPlaneIsToldWhereTheNodePutTheStoreAndTheBroker failed on main since PR #467: it seeds
anchor.mesh-controller with no seal, and an account counts as issued only once sealed (ADR 0543 §2). The rule is right;
the fixture is fixed, sealing the seeded credential the controller already uses. Test only.
2026-10-25 22:51:25 +01:00
jschoubben f67d15536b Grant the grants part publishing the modules' memberships, and nothing more (ADR 0547 §5, issue 1945)
No membership is published since release 6 (issue 1945 notes 42, 47), so a seat-holding module placed now cannot
serve its seat or pull asks (mesh-build-go on ace, 2026-10-25). The operator decided at 22:31 that the grants part
publishes them from the first milestone: its principal, module mesh-grants, may publish mesh.assignment.*.* and no
stream definition. The SDK's copy of this composer gets the same line, so the two stay byte-equal.

Research 072 (the controller, NATS-native): this moves the memberships of its Bus row to the grants part.
2026-10-25 22:32:35 +01:00
94 changed files with 1746 additions and 523 deletions
+42 -6
View File
@@ -13,6 +13,7 @@ import (
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/lease"
"github.com/novox/mesh-controller/internal/link"
@@ -72,6 +73,14 @@ type actor struct {
// lease over in a second rather than in a renewal of five (novox/hq issue 591). A renewal is given
// the lease's own RenewEvery to answer however often it is made, so a slow bus does not lose it.
renewEvery, poll time.Duration
// releasing is held for the whole of a release, so a second caller — the stop signal's and serve's own
// deferred one — waits until the lease is given back and its epoch recorded, never past it to the store's
// close (novox/hq issue 906).
releasing sync.Mutex
// keeper and inv are where a takeover is said (sayTakeover): set once the serving controller's condition
// store opens; nil before, and in a command.
keeper *conditions.Keeper
inv *inventory.Inventory
}
// theLease is this process's standing to the lease.
@@ -174,8 +183,11 @@ func (a *actor) forACommand(ctx context.Context) (uint64, error) {
return holder.Epoch, nil
}
// release gives back what this process holds, at its end.
// release gives back what this process holds, at its end — or at the stop signal, before the rest of the
// process is taken down (serveUnderTheLease). A second call waits for the first to finish.
func (a *actor) release() {
a.releasing.Lock()
defer a.releasing.Unlock()
a.mu.Lock()
closing := a.close
a.close = nil
@@ -266,20 +278,40 @@ func (a *actor) serveUnderTheLease(ctx context.Context, inv *inventory.Inventory
}
a.mu.Lock()
a.close = func() {
if held, err := l.Epoch(); err == nil {
// Given back first, then recorded: an epoch is recorded released only when its key is gone, and
// whatever the gate says by then — a shutdown longer than half the key's age closed it, and the epoch
// was left open for the next holder to find "expired" (novox/hq issue 906, 2026-10-26).
if given := l.Release(context.Background()); given != 0 {
ending, cancel := context.WithTimeout(context.Background(), 5*time.Second)
if err := inv.EndEpoch(ending, held, inventory.EpochReleased); err != nil {
say("how epoch %d ended could not be recorded: %v", held, err)
if err := inv.EndEpoch(ending, given, inventory.EpochReleased); err != nil {
say("how epoch %d ended could not be recorded: %v", given, err)
}
cancel()
}
l.Release(context.Background())
js.Close()
}
a.mu.Unlock()
// **The stop signal gives the lease back first** (novox/hq issue 906): the moment ctx ends, before the rest
// of the process is taken down, the gate closes and the key goes, so the next controller takes the lease
// at once and the epoch is recorded released. Waiting for the process's own end left the key unrenewed for
// as long as the shutdown took — ten seconds on 2026-10-26, past the gate's half-age.
go func() {
<-ctx.Done()
if a.holds(l) {
say("stopping: the controller lease is given back first")
}
a.release()
}()
return gone, nil
}
// holds is whether this process holds l now.
func (a *actor) holds(l *lease.Lease) bool {
a.mu.Lock()
defer a.mu.Unlock()
return a.held == l
}
// took is the lease taken: recorded, earlier epochs nobody gave back ended as expired, kept.
func (a *actor) took(ctx context.Context, l *lease.Lease, inv *inventory.Inventory, epoch uint64, gone chan struct{}) {
a.mu.Lock()
@@ -288,15 +320,19 @@ func (a *actor) took(ctx context.Context, l *lease.Lease, inv *inventory.Invento
h := holderOf(instance)
recording, cancel := context.WithTimeout(ctx, 10*time.Second)
expired, err := inv.TookEpoch(recording, inventory.Epoch{Epoch: epoch, Instance: h.Instance, Host: h.Host,
Build: h.Build, Taken: time.Now()})
Build: h.Build, Taken: time.Now()}, l.GivenBack())
cancel()
if err != nil {
fmt.Printf("epoch %d could not be recorded as taken, so a stale refusal from it will not name it: %v\n", epoch, err)
}
if given := l.GivenBack(); given != 0 {
fmt.Printf("the lease was given back by the controller of epoch %d: taken at once, at epoch %d\n", given, epoch)
}
for _, e := range expired {
fmt.Printf("the controller of epoch %d (%s) stopped renewing the lease without giving it back: it is "+
"taken over at epoch %d\n", e.Epoch, e.Instance, epoch)
}
a.sayTakeover(ctx, epoch)
go l.Keep(ctx)
go func() {
<-l.Lost()
@@ -78,6 +78,8 @@ func TestTheRegistryTrustAndEveryImageFollowThePortTheNodeGaveTheStore(t *testin
func TestTheControlPlaneIsToldWhereTheNodePutTheStoreAndTheBroker(t *testing.T) {
open := aMesh(t)
// Issue 1126: the placement export over this fresh mesh, before anything below changes it.
placementExportIsEachNodesPlacementFromTheStore(t, open)
ctx := t.Context()
raw, err := os.ReadFile("../../module.json")
if err != nil {
@@ -111,6 +113,11 @@ func TestTheControlPlaneIsToldWhereTheNodePutTheStoreAndTheBroker(t *testing.T)
Kind: inventory.BusController, Node: "anchor", Module: "mesh-controller"}, "bootstrap"); err != nil {
t.Fatal(err)
}
// Seeded with the credential the controller already uses, so it is sealed: an account counts as issued only once
// sealed (ADR 0543 §2, PR #467), and genesis's seed is the credential in use.
if err := open.inventory.MarkBusAccountSealed(ctx, "anchor.mesh-controller"); err != nil {
t.Fatal(err)
}
if _, err := assign(ctx, open, "anchor", "mesh-controller"); err != nil {
t.Fatal(err)
}
@@ -217,6 +217,11 @@ func TestTheAdaptersInputCarriesTheOwnSecretsTheBoxesTheLicencesAndTheBus(t *tes
Module: "objects"}, "a-password"); err != nil {
t.Fatal(err)
}
// Minted and sealed to its node: an account counts as issued, and is named among the minted, only once sealed
// (ADR 0543 §2, PR #467).
if err := inv.MarkBusAccountSealed(ctx, "anchor.objects"); err != nil {
t.Fatal(err)
}
if err := inv.PutBusMembership(ctx, "laptop", "sealed-membership"); err != nil {
t.Fatal(err)
}
@@ -13,6 +13,7 @@ import (
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/identity"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
@@ -76,13 +77,18 @@ func aComposingMesh(t *testing.T) *composingMesh {
return m
}
// trunkKeys are a done key and the indexer's trunk value naming it.
// trunkKeys are a done key and the indexer's trunk value naming it. The indexer holds that value at the trunk's commit
// alone, as the real one does: answering it for any commit made a module registered at an older commit look already
// registered at the trunk's fingerprint, so the take-in only moved its commit and recorded no build (issue 1945).
type trunkKeys struct {
keyBuckets
trunk sdkdelivery.Module
}
func (k trunkKeys) ModuleAt(context.Context, string, string) (sdkdelivery.Module, error) {
func (k trunkKeys) ModuleAt(_ context.Context, _ string, commit string) (sdkdelivery.Module, error) {
if commit != "" && !sameCommitID(commit, k.trunk.Commit) {
return sdkdelivery.Module{}, builderkeys.ErrNoValue
}
return k.trunk, nil
}
-9
View File
@@ -88,8 +88,6 @@ func composeModulesByNode(ctx context.Context, open *stores, nodes []string) (ma
}
return u.Policy(), u.From, u.Why
}
// The memberships of each node's modules are issued by its composition (memberships.go): read once for the pass.
issuing, issuingErr := membershipsFor(ctx, open, link.OverNATS{Conn: bus.Conn(), JS: bus.Context()})
sort.Strings(nodes)
out := map[string][]string{}
errsOf := map[string][]error{}
@@ -117,13 +115,6 @@ func composeModulesByNode(ctx context.Context, open *stores, nodes []string) (ma
errsOf[node] = append(errsOf[node], fmt.Errorf("%s: %w", node, err))
return
}
// Its modules' memberships, before its compositions (issue 249): a failure is the node's, and its compositions
// are still written; the take-in that asked tries the node again.
if issuingErr != nil {
errsOf[node] = append(errsOf[node], fmt.Errorf("%s: %w", node, issuingErr))
} else if err := issuing.issue(ctx, node, declared); err != nil {
errsOf[node] = append(errsOf[node], fmt.Errorf("%s: %w", node, err))
}
// The provisioner's plans of the modules the node's composition hands to it, for the assignments mirror.
if err := open.inventory.RecordComposedPlans(ctx, node, plansKept(declared)); err != nil {
errsOf[node] = append(errsOf[node], fmt.Errorf("%s: its provisioner's plans: %w", node, err))
+2 -2
View File
@@ -23,10 +23,10 @@ func TestModuleCheckRefusesAnIdentityThatOverflowsWhatItRequires(t *testing.T) {
return p
}
objects := write("objects", `{"module":"objects","version":"1",
"provides":[{"name":"s3-bucket","scope":"mesh","identity":{"max":20,"in":"an S3 access key"}}],
"provides":[{"name":"s3-bucket","scope":"mesh","identity":{"max":20,"in":"an S3 access key"},"credential":{"form":"pair"}}],
"receives":{"s3-bucket":"/var/lib/mesh/objects/mesh.json"}}`)
resolver := write("resolver", `{"module":"resolver","version":"1",
"provides":[{"name":"wildcard-resolution","scope":"mesh","identity":false}]}`)
"provides":[{"name":"wildcard-resolution","scope":"mesh","identity":false,"credential":{"form":"none"}}]}`)
album := write("photoalbum", `{"module":"photoalbum","version":"1","requires":["s3-bucket"]}`)
nm := write("networkmanager", `{"module":"networkmanager","version":"1","requires":["wildcard-resolution"]}`)
+7 -2
View File
@@ -182,6 +182,9 @@ func TestANeedIsIssuedToItsHolderOnlyWhereTheProviderIsMovedAndItsSoftwareIssues
offer := catalogue.Offer{Name: "gitea-read", Scope: catalogue.ScopeMesh}
if issued {
offer.Credential = &catalogue.OfferCredential{IssuedBy: catalogue.IssuedBySoftware, Tool: "gitea_agent_token"}
} else {
// A login of each consumer's own, said: an offer naming no form has no credential (mesh-sdk 6b8c694).
offer.Credential = &catalogue.OfferCredential{Form: catalogue.CredentialFormPair}
}
return catalogue.Manifest{Module: "gitea", Version: "1", Moved: moved, Provides: []catalogue.Offer{offer}}
}
@@ -216,8 +219,10 @@ func TestANeedIsIssuedToItsHolderOnlyWhereTheProviderIsMovedAndItsSoftwareIssues
// marked moved; another module is not.
func TestTheModuleHoldingTheVaultSeatCountsAsMovedForItsConsumers(t *testing.T) {
vault := catalogue.Manifest{Module: "mesh-vault", Version: "1", Claims: []catalogue.Claim{{Name: "mesh-vault", Scope: catalogue.ScopeMesh}},
Provides: []catalogue.Offer{{Name: "secret", Scope: catalogue.ScopeMesh}}}
other := catalogue.Manifest{Module: "mesh-vault", Version: "1", Provides: []catalogue.Offer{{Name: "secret", Scope: catalogue.ScopeMesh}}}
Provides: []catalogue.Offer{{Name: "secret", Scope: catalogue.ScopeMesh,
Credential: &catalogue.OfferCredential{Form: catalogue.CredentialFormPair}}}}
other := catalogue.Manifest{Module: "mesh-vault", Version: "1", Provides: []catalogue.Offer{{Name: "secret", Scope: catalogue.ScopeMesh,
Credential: &catalogue.OfferCredential{Form: catalogue.CredentialFormPair}}}}
if got := needOf(t, map[string]catalogue.Manifest{"mesh-vault": vault}, "mesh-vault", "gitea", "secret"); !got.IssuedToHolder || !got.ProviderMoved {
t.Errorf("the vault's consumer is not answered by a moved provider: %+v", got)
}
+15 -2
View File
@@ -89,15 +89,28 @@ func TestTwoControllersOneActs(t *testing.T) {
t.Fatalf("B, waiting, may act: %v", err)
}
// A hands over, as a controller being replaced does: B takes the lease at once, at a higher epoch.
// A is stopped, as its service manager stops a controller being replaced (novox/hq issue 906): the stop
// alone gives the lease back, before anything else of A is taken down, and B takes it within a second, at
// a higher epoch — no wait for the key's age, and no condition said.
stopA()
a.release()
stopped := time.Now()
var lostB <-chan struct{}
select {
case lostB = <-tookB:
case <-time.After(10 * time.Second):
t.Fatal("B did not take the lease A gave back")
}
if waited := time.Since(stopped); waited > time.Second {
t.Fatalf("B took the lease %s after A was stopped: the stop did not give it back first", waited)
}
a.release() // what serve's own end does after: waits for the release under way, gives nothing twice
conditionsOfB := conditions.NewInMemory()
keeperOfB := conditions.NewKeeper(ctx, conditions.Options{Store: conditionsOfB, History: conditionsOfB})
b.watchTakeovers(ctx, inv, keeperOfB)
keeperOfB.Close(ctx)
if said, err := conditionsOfB.Since(ctx, stopped.Add(-time.Minute)); err != nil || len(said) != 0 {
t.Fatalf("a lease given back is said as a condition: %+v, %v", said, err)
}
select {
case <-lostA:
default:
-186
View File
@@ -1,186 +0,0 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"reflect"
"slices"
"sort"
"strings"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/nats-io/nats.go"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/link"
)
// **A node's composition issues its modules' memberships** (novox/hq ADR 0160, ADR 0201; issue 249 met again under
// issue 1945).
//
// A membership is what an assignment serves and reaches: its subjects, its seats, and the state, logs, streams and waits
// its manifest declares, published last-per-subject on the AssignmentsStream, where the tool runner and every bus process
// read and follow their own. A push issued them before it sent a declaration; release 6 removed the push and nothing
// issued them after, so a module taken in with a new state, read or stream was granted it on the bus and refused it by its
// runtime, which keeps a module to the membership last issued (2026-10-25: mesh-delivery-gate's `conditions`, on novox
// from 20:33). Composing a node is now where its memberships are issued: from the same records the bus's grants are
// composed from (so what a runtime serves and what its account may are one composition), with the private network's
// addresses and what each module receives from the node's composition, before its compositions are written (the code
// arriving first was refused its own state, issue 249).
//
// Only a membership that differs from the one last issued is published, and only over that one, so a pass that changes
// nothing re-serves nothing and a newer one issued meanwhile is not overwritten (issues 930, 934).
// membershipPass is what one composer pass issues every node's memberships from, read once for the pass.
type membershipPass struct {
records broker.Records
where broker.Placements
bus link.OverNATS
}
// membershipsFor reads the records, asserts the memberships' stream and every declared bucket and log (a membership
// names them, and a module whose state does not exist fails its first read), and judges which machines are root-free
// (ADR 0259 §8: a channel's verified sender is composed only from one) when a seat promises one, as a push did.
func membershipsFor(ctx context.Context, open *stores, bus link.OverNATS) (*membershipPass, error) {
records, err := open.inventory.BusRecords(ctx)
if err != nil {
return nil, fmt.Errorf("the memberships are not issued: %w", err)
}
raiser := broker.OnConn(bus.Conn)
for _, s := range broker.MeshStreams() {
if s.Name != broker.AssignmentsStream {
continue
}
if err := raiser.EnsureStream(s); err != nil {
return nil, fmt.Errorf("the memberships are not issued: their stream: %w", err)
}
}
buckets, err := open.inventory.DeclaredBuckets(ctx)
if err != nil {
return nil, fmt.Errorf("the memberships are not issued: the modules' state: %w", err)
}
if _, err := broker.RaiseBuckets(raiser, buckets); err != nil {
return nil, fmt.Errorf("the memberships are not issued: the modules' state could not be asserted: %w", err)
}
logs, err := open.inventory.DeclaredLogs(ctx)
if err != nil {
return nil, fmt.Errorf("the memberships are not issued: the modules' logs: %w", err)
}
if _, err := broker.RaiseLogs(raiser, logs); err != nil {
return nil, fmt.Errorf("the memberships are not issued: the modules' logs could not be asserted: %w", err)
}
var whys map[string]string
if promisesVerifiedSender(records) {
records.RootFree, whys = rootVerdictsOf(judgeRootFree(ctx, newRootReader(ctx, open.inventory, bus.Conn),
records.Nodes, time.Now()))
}
where := broker.PlacementsOf(records, records.Interchangeable)
for _, line := range droppedVerifiedSenders(records, where, whys) {
fmt.Println(line)
}
return &membershipPass{records: records, where: where, bus: bus}, nil
}
// promisesVerifiedSender says a seat some module holds promises a verified sender: only then is the root-free judgement,
// a discovery on the bus, worth its seconds. Pure.
func promisesVerifiedSender(records broker.Records) bool {
for _, declared := range records.Assigned {
for _, d := range declared {
for _, s := range d.Holds {
if slices.Contains(s.Capabilities, "verified-sender") {
return true
}
}
}
}
return false
}
// issue issues node's memberships on the bus, from its composition's addresses and received contributions.
func (p *membershipPass) issue(ctx context.Context, node string, declared sendable) error {
issued, err := issueComposedMemberships(node, p.records, p.where, declared.Mesh, declared.Received,
func(node, module string) ([]byte, uint64, error) {
m, err := p.bus.JS.GetLastMsg(broker.AssignmentsStream, broker.MembershipSubject(node, module), nats.Context(ctx))
if err != nil {
return nil, 0, err
}
return m.Data, m.Sequence, nil
},
func(node, module string, body []byte, seq uint64) error {
return p.bus.PublishMembershipAfter(ctx, node, module, body, seq)
})
if len(issued) > 0 {
fmt.Printf("memberships: %s: issued %s\n", node, strings.Join(issued, ", "))
}
return err
}
// issueComposedMemberships composes the membership of every module assigned to node and publishes each one that differs
// from the one last issued, over that one (seq 0 when none was: only if none was issued meanwhile). It answers the
// modules it issued; every failure is tried past and joined in the error.
func issueComposedMemberships(node string, records broker.Records, where broker.Placements, mesh []string,
received map[string]map[string][]catalogue.Contribution,
last func(node, module string) ([]byte, uint64, error),
publish func(node, module string, body []byte, seq uint64) error) ([]string, error) {
var issued []string
var errs []error
for _, d := range records.Assigned[node] {
body, err := composedMembership(node, d, where, mesh, received[d.Module])
if err != nil {
errs = append(errs, fmt.Errorf("%s on %s: its membership: %w", d.Module, node, err))
continue
}
was, seq, err := last(node, d.Module)
switch {
case errors.Is(err, nats.ErrMsgNotFound):
was, seq = nil, 0
case err != nil:
errs = append(errs, fmt.Errorf("%s on %s: its membership as last issued could not be read: %w", d.Module, node, err))
continue
}
if sameMembership(was, body) {
continue
}
if err := publish(node, d.Module, body, seq); err != nil {
errs = append(errs, fmt.Errorf("%s on %s: its membership could not be issued: %w", d.Module, node, err))
continue
}
issued = append(issued, d.Module)
}
sort.Strings(issued)
return issued, errors.Join(errs...)
}
// composedMembership is one assignment's membership as a node's composition issues it: MembershipFor, with the private
// network's addresses and what the module receives, by requirement (ADR 0167).
func composedMembership(node string, d broker.Declared, where broker.Placements, mesh []string,
received map[string][]catalogue.Contribution) ([]byte, error) {
m := broker.MembershipFor(node, d, where)
m.Mesh = mesh
for requirement, given := range received {
raw, err := json.Marshal(given)
if err != nil {
return nil, err
}
if m.Receives == nil {
m.Receives = map[string]json.RawMessage{}
}
m.Receives[requirement] = raw
}
return json.Marshal(m)
}
// sameMembership says two memberships say the same, whatever the order of their fields. Pure.
func sameMembership(was, now []byte) bool {
if was == nil {
return false
}
var a, b any
if json.Unmarshal(was, &a) != nil || json.Unmarshal(now, &b) != nil {
return false
}
return reflect.DeepEqual(a, b)
}
+6 -1
View File
@@ -133,9 +133,14 @@ func nodeCommand(ctx context.Context, args []string) error {
// refuse (novox/hq issue 358). Read-only on the node and here, so a verb may run it too.
return nodeJudgeDirectories(ctx, open, args[1:])
case "placement-export":
// Each node's placement as the catalogue's placement file says it (novox/hq ADR 0555 §6): read-only, so a verb
// may run it.
return placementExport(ctx, open, args[1:])
default:
return fmt.Errorf("node has no %q; it has add, list, show, public-domain, account, agent-account, hand-over, "+
"setuid-search and judge-directories", args[0])
"setuid-search, judge-directories and placement-export", args[0])
}
}
+50
View File
@@ -0,0 +1,50 @@
package main
import (
"context"
"errors"
"sort"
"git.novox.be/novox/mesh-sdk/go/compose/frombus"
"git.novox.be/novox/mesh-sdk/go/delivery"
)
// placementExport prints each node's placement as the catalogue's placement file says it (novox/hq ADR 0555 §6, issue
// 1126): what a person decided of the node and where its modules run, read by this controller's own code from its own
// store, as JSON keyed by node. The one verb the controller gains for its exit (the operator, 2026-10-26 19:06:40, ADR
// 0547 §1's one exception besides §5's grant): read once by the seed of the placement files, whose pull request merges
// only when its composition check composes every node equal to what it runs. It reads the same Input every composition
// of this controller reads (readMeshAndNodes, readSettings) and writes nothing.
//
// It moves a piece of the research "The controller, NATS-native": the controller's last node data (node rows, placement,
// ports, pins, bindings, config placement) leaves it for the catalogue, before its exit at ADR 0547's third milestone.
func placementExport(ctx context.Context, open *stores, args []string) error {
if len(args) > 1 {
return errors.New("node placement-export [<name>]")
}
in, _, err := readMeshAndNodes(ctx, open)
if err != nil {
return err
}
if _, err := readSettings(ctx, open, &in); err != nil {
return err
}
var names []string
if len(args) == 1 {
names = []string{args[0]}
} else {
for _, n := range in.Nodes {
names = append(names, n.Name)
}
sort.Strings(names)
}
out := make(map[string]delivery.Placement, len(names))
for _, name := range names {
p, err := frombus.Export(in, name)
if err != nil {
return err
}
out[name] = p
}
return printJSON(out)
}
@@ -0,0 +1,71 @@
//go:build integration
package main
import (
"encoding/json"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/compose/frombus"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"git.novox.be/novox/mesh-sdk/go/overlay"
)
// Issue 1126 (novox/hq ADR 0555 §6, the operator's one export verb of 2026-10-26 19:06:40): `node placement-export`
// answers, from the real store, each node's placement as the catalogue's placement file says it: what a person decided
// of the node (its endpoint, site, hub and address on the private network) and its modules in the order the composition
// reads them, exactly frombus.Export of the input this controller composes from; one node when named; a node the store
// lacks is refused. Checked within TestTheControlPlaneIsToldWhereTheNodePutTheStoreAndTheBroker, over the same fresh
// mesh, so the export adds no integration test of its own (the test budget).
func placementExportIsEachNodesPlacementFromTheStore(t *testing.T, open *stores) {
t.Helper()
ctx := t.Context()
in, _, err := readMeshAndNodes(ctx, open)
if err != nil {
t.Fatal(err)
}
if _, err := readSettings(ctx, open, &in); err != nil {
t.Fatal(err)
}
var all map[string]sdkdelivery.Placement
if err := json.Unmarshal([]byte(stdoutOf(t, func() error { return placementExport(ctx, open, nil) })), &all); err != nil {
t.Fatal(err)
}
if len(all) != 2 {
t.Fatalf("exported %d nodes, want anchor and laptop", len(all))
}
for i, name := range []string{"anchor", "laptop"} {
got := all[name]
d := got.Decided
if d.Endpoint != name+".example:51820" || d.Site != "here" || d.Hub != (name == "anchor") || d.Address != []string{"10.77.0.1", "10.77.0.2"}[i] {
t.Errorf("%s's decided facts are %+v", name, d)
}
if a := got.Assigned(); len(a) == 0 || !strings.Contains(strings.Join(a, ","), overlay.Name) {
t.Errorf("%s's modules are %v, want the private network among them", name, a)
}
// What it prints is the file the seed writes: at the placement file's version, read back by its reader.
if raw, _ := json.Marshal(got); got.V != sdkdelivery.Version {
t.Errorf("%s is exported at version %d: %s", name, got.V, raw)
} else if _, err := sdkdelivery.DecodePlacement(raw); err != nil {
t.Errorf("%s's export does not read as a placement file: %v", name, err)
}
want, err := frombus.Export(in, name)
if err != nil {
t.Fatal(err)
}
if !sdkdelivery.SamePlacement(got, want) {
t.Errorf("%s's export differs from frombus.Export of the controller's own input:\n%+v\n%+v", name, got, want)
}
}
var one map[string]sdkdelivery.Placement
if err := json.Unmarshal([]byte(stdoutOf(t, func() error { return placementExport(ctx, open, []string{"laptop"}) })), &one); err != nil {
t.Fatal(err)
}
if _, ok := one["laptop"]; !ok || len(one) != 1 {
t.Errorf("one node named exported %v", one)
}
if err := placementExport(ctx, open, []string{"stranger"}); err == nil {
t.Error("a node the store lacks was exported")
}
}
+3
View File
@@ -180,6 +180,9 @@ func serve(ctx context.Context) (err error) {
if keeper != nil {
defer closeConditions(keeper)
defer stopRunning() // on an error return, the goroutines end before the keeper closes
// The takeover of the lease said as it happened, raised and cleared (novox/hq issue 906).
theLease.watchTakeovers(ctx, inv, keeper)
defer theLease.watchTakeovers(ctx, nil, nil) // the keeper closes after this: nothing said into it then
}
// The bus's own objects, asserted on every start. **Not created once at genesis**: a stream
@@ -108,83 +108,14 @@ func TestADroppedVerifiedSenderIsSaidWithItsReason(t *testing.T) {
}
}
// **A node's composition issues every membership that changed** (issue 249 met again under issue 1945): release 6 removed
// the push that issued memberships, and mesh-delivery-gate taken in with a new state `conditions` was granted its bucket and
// refused it by the tool runner, which reads the membership last issued (2026-10-25, novox). The self-check's reissue
// leaves one never issued alone: the composition issues it.
func TestAComposedNodeIssuesEveryMembershipThatChanged(t *testing.T) {
gate := func(states ...string) broker.Declared {
d := broker.Declared{Module: "mesh-delivery-gate"}
for _, s := range states {
d.State = append(d.State, broker.Bucket{Module: "mesh-delivery-gate", Name: s})
}
return d
}
records := func(d broker.Declared) broker.Records {
return broker.Records{Nodes: []string{"novox"}, Assigned: map[string][]broker.Declared{"novox": {d}}}
}
issuedAs := func(d broker.Declared, mesh []string) []byte {
body, err := composedMembership("novox", d, broker.PlacementsOf(records(d), nil), mesh, nil)
if err != nil {
t.Fatal(err)
}
return body
}
newer := []byte(`{"node":"novox","module":"mesh-delivery-gate","newer":true}`)
for _, c := range []struct {
name string
before []byte // the membership last issued, nil for none
between bool // another issued meanwhile, after the read
issued []string
states []string // what the bus holds after, by state name; nil when it must hold newer
err string
}{
{name: "never issued is issued", issued: []string{"mesh-delivery-gate"}, states: []string{"asks", "conditions"}},
{name: "a new state is issued", before: issuedAs(gate("asks"), []string{"10.0.0.1"}),
issued: []string{"mesh-delivery-gate"}, states: []string{"asks", "conditions"}},
{name: "unchanged is not issued again", before: issuedAs(gate("asks", "conditions"), []string{"10.0.0.1"}),
states: []string{"asks", "conditions"}},
{name: "a newer one meanwhile is not overwritten", before: issuedAs(gate("asks"), []string{"10.0.0.1"}),
between: true, err: "wrong last sequence"},
} {
t.Run(c.name, func(t *testing.T) {
r := records(gate("asks", "conditions"))
where := broker.PlacementsOf(r, nil)
bus := &fakeBus{last: map[string][]byte{}, published: map[string][]byte{}, seq: map[string]uint64{}}
if c.before != nil {
bus.last["novox/mesh-delivery-gate"], bus.seq["novox/mesh-delivery-gate"] = c.before, 1
}
if c.between {
bus.between = func() { bus.last["novox/mesh-delivery-gate"], bus.seq["novox/mesh-delivery-gate"] = newer, 2 }
}
if n := reissueVerifiedSenders(r, where, bus.get, bus.put, func(string, ...any) {}); n != 0 {
t.Fatalf("the self-check's reissue issued %d", n)
}
issued, err := issueComposedMemberships("novox", r, where, []string{"10.0.0.1"}, nil, bus.get, bus.put)
if c.err == "" && err != nil || c.err != "" && (err == nil || !strings.Contains(err.Error(), c.err)) {
t.Fatalf("error %v, want %q", err, c.err)
}
if strings.Join(issued, ",") != strings.Join(c.issued, ",") {
t.Fatalf("issued %v, want %v", issued, c.issued)
}
if c.states == nil {
if string(bus.last["novox/mesh-delivery-gate"]) != string(newer) {
t.Fatalf("the newer membership was overwritten: %s", bus.last["novox/mesh-delivery-gate"])
}
return
}
var m broker.Membership
if err := json.Unmarshal(bus.last["novox/mesh-delivery-gate"], &m); err != nil {
t.Fatal(err)
}
var names []string
for _, s := range m.State {
names = append(names, s.Name)
}
if strings.Join(names, ",") != strings.Join(c.states, ",") || strings.Join(m.Mesh, ",") != "10.0.0.1" {
t.Fatalf("the bus holds states %v and mesh %v, want %v", names, m.Mesh, c.states)
}
})
func TestAMembershipNeverIssuedIsLeftToItsOwnSend(t *testing.T) {
records, _ := verifiedSenderRecords(map[string]bool{"novox": true})
bus := &fakeBus{last: map[string][]byte{}, published: map[string][]byte{}, seq: map[string]uint64{}}
var said []string
n := reissueVerifiedSenders(records, broker.PlacementsOf(records, nil), bus.get, bus.put,
func(f string, a ...any) { said = append(said, f) })
if n != 0 || len(said) != 0 {
t.Fatalf("issued %d, said %v", n, said)
}
}
+8 -1
View File
@@ -349,6 +349,11 @@ func (a *verbArguments) commandLine() ([]string, error) {
return nil, err
}
return []string{"node", "judge-directories", str("node")}, nil
case "placement-export":
if node := str("node"); node != "" {
return []string{"node", "placement-export", node}, nil
}
return []string{"node", "placement-export"}, nil
case "modules":
return []string{"module", "list", "--json"}, nil
case "seats":
@@ -1400,7 +1405,9 @@ func seatAnnouncement(handlers map[string]link.ToolHandler) micro.Info {
// nodeReads are the `node` subcommands a verb may run: the ones that only read.
var nodeReads = map[string]bool{"list": true, "show": true,
// Read-only on the node too: the engine judges its directories and changes nothing (novox/hq issue 358).
"judge-directories": true}
"judge-directories": true,
// Each node's placement as the catalogue's placement file says it, read from the store (novox/hq ADR 0555 §6).
"placement-export": true}
// flagsOnly says a command line's rest names no subcommand: empty, or beginning with a flag. For a command
// with no subcommands every word is a flag, its value or a name it reads.
+113 -15
View File
@@ -1,6 +1,7 @@
package main
import (
"context"
"fmt"
"sort"
"strings"
@@ -147,9 +148,10 @@ var signalsTable = []signalRow{
return newestOf(f.machines, func(m machineFacts) time.Time { return m.toolsHeard })
}},
{Row: "S12", Signal: "the controller lease renewed", Emitter: "controller", Trigger: "every 5 s",
Bound: "15 s (the key's age); a holder that lost the lease, or stopped renewing and was taken over, and a " +
"lease bucket found raised again from nothing, are said for an hour after; a controller serving without " +
"the lease, for as long as it does",
Bound: "15 s (the key's age); a holder that lost the lease or stopped renewing it is said once by the " +
"controller that took it over, raised and cleared as it takes it, and for an hour after while none " +
"holds it; a lease given back is never said; a lease bucket found raised again from nothing is said " +
"for an hour after; a controller serving without the lease, for as long as it does",
Kind: "lease-lost", Severity: conditions.Urgent, Phase: 2,
needs: func(f *signalFacts) error { return f.leaseErr }, watch: watchLease,
newest: func(f *signalFacts) time.Time { return f.lease.renewed }},
@@ -644,10 +646,29 @@ func watchLease(f *signalFacts) []conditions.Observation {
"does not know it lost it",
Said: l.resetSaid})
}
// An epoch below the one this controller holds was taken over by it: said once by the takeover itself,
// raised and cleared (sayTakeover), not observed for an hour after (novox/hq issue 906). One that ended
// while this controller does not hold the lease — serving without it — is still said for the hour.
var ended []inventory.Epoch
for _, e := range l.ended {
if l.held && e.Epoch < l.epoch {
continue
}
ended = append(ended, e)
}
if o, ok := lostObservation(ended, l.epoch, f.now); ok {
out = append(out, o)
}
return out
}
// lostObservation is the lease-lost condition of the epochs that lost the lease or let it expire within the
// hour, named by the newest; false when none did. An epoch given back is no loss.
func lostObservation(ended []inventory.Epoch, acting uint64, now time.Time) (conditions.Observation, bool) {
var lost []string
newest := inventory.Epoch{}
for _, e := range l.ended {
if e.Ended == nil || e.How == inventory.EpochReleased || f.now.Sub(*e.Ended) > advisoryQuiet {
for _, e := range ended {
if e.Ended == nil || e.How == inventory.EpochReleased || now.Sub(*e.Ended) > advisoryQuiet {
continue
}
lost = append(lost, fmt.Sprintf("epoch %d (%s) %s at %s", e.Epoch, e.Instance, e.How,
@@ -656,18 +677,95 @@ func watchLease(f *signalFacts) []conditions.Observation {
newest = e
}
}
if len(lost) > 0 {
how := "lost it: another holder wrote its key, or no renewal was made within the key's age"
if newest.How == inventory.EpochExpired {
how = "stopped renewing it without giving it back, and was taken over"
if len(lost) == 0 {
return conditions.Observation{}, false
}
how := "lost it: another holder wrote its key, or no renewal was made within the key's age"
if newest.How == inventory.EpochExpired {
how = "stopped renewing it without giving it back, and was taken over"
}
return conditions.Observation{Scope: conditions.ScopeCore, ID: "controller.lease", Token: "lost",
Kind: "lease-lost", Machine: newest.Host, Severity: conditions.Urgent, Source: "S12",
Summary: fmt.Sprintf("the controller of epoch %d (%s) %s; the controller of epoch %d acts now", newest.Epoch,
newest.Instance, how, acting),
Said: strings.Join(lost, "; ")}, true
}
// takeover is the condition a takeover says: the epoch just before the one taken, when it lost the lease or
// let it expire; false when it gave it back, or none is recorded.
func takeover(ended []inventory.Epoch, taken uint64, now time.Time) (conditions.Observation, bool) {
var before *inventory.Epoch
for i, e := range ended {
if e.Epoch < taken && (before == nil || e.Epoch > before.Epoch) {
before = &ended[i]
}
out = append(out, conditions.Observation{Scope: conditions.ScopeCore, ID: "controller.lease", Token: "lost",
Kind: "lease-lost", Machine: newest.Host, Severity: conditions.Urgent,
Summary: fmt.Sprintf("the controller of epoch %d (%s) %s; the controller of epoch %d acts now", newest.Epoch,
newest.Instance, how, l.epoch),
Said: strings.Join(lost, "; ")})
}
if before == nil {
return conditions.Observation{}, false
}
return lostObservation([]inventory.Epoch{*before}, taken, now)
}
// sayTakeover is the takeover said as an event (novox/hq issue 906): when the controller before this one lost
// the lease or let it expire, the lease-lost condition is raised and cleared at once by this controller, which
// holds the lease — the operator reads that it happened and that it is over, not an urgent alarm for an hour.
// A lease given back says nothing. Said once the condition store is open; a take before that is said then.
func (a *actor) sayTakeover(ctx context.Context, epoch uint64) {
a.mu.Lock()
keeper, inv := a.keeper, a.inv
a.mu.Unlock()
if keeper == nil || inv == nil || epoch == 0 {
return
}
now := time.Now()
reading, cancel := context.WithTimeout(ctx, 10*time.Second)
defer cancel()
ended, err := inv.EpochsSince(reading, now.Add(-advisoryQuiet))
if err != nil {
// Failure is loud: S12 no longer observes an epoch below the one held, so an unread takeover is said as
// one that was not a handover — raised, and left for S12 to clear (ADR 0558 §3).
fmt.Printf("how the epoch before %d ended cannot be read, so its takeover is said as an expiry: %v\n", epoch, err)
o := conditions.Observation{Scope: conditions.ScopeCore, ID: "controller.lease", Token: "lost",
Kind: "lease-lost", Severity: conditions.Urgent, Source: "S12",
Summary: fmt.Sprintf("the controller of epoch %d took the lease, and how the controller before it ended "+
"cannot be read: it is said as one that stopped renewing without giving it back", epoch),
Said: err.Error()}
if _, err := keeper.Observe(reading, o); err != nil {
fmt.Printf("the takeover at epoch %d could not be said: %v\n", epoch, err)
}
return
}
sayTakeoverOf(reading, keeper, ended, epoch, now)
}
// sayTakeoverOf raises and clears the takeover's condition over the epochs read.
func sayTakeoverOf(ctx context.Context, keeper *conditions.Keeper, ended []inventory.Epoch, epoch uint64, now time.Time) {
o, ok := takeover(ended, epoch, now)
if !ok {
return
}
if _, err := keeper.Observe(ctx, o); err != nil {
fmt.Printf("the takeover at epoch %d could not be said: %v\n", epoch, err)
return
}
if _, err := keeper.Clear(ctx, o.Key(), fmt.Sprintf("taken over: the controller of epoch %d holds the lease", epoch)); err != nil {
fmt.Printf("the takeover at epoch %d was said and could not be cleared: %v\n", epoch, err)
}
}
// watchTakeovers gives the serving controller's condition store to the lease, and says the takeover of the
// epoch it holds, if it took one over before the store opened.
func (a *actor) watchTakeovers(ctx context.Context, inv *inventory.Inventory, keeper *conditions.Keeper) {
a.mu.Lock()
a.keeper, a.inv = keeper, inv
held := a.held
a.mu.Unlock()
if held == nil {
return
}
if epoch, err := held.Epoch(); err == nil {
a.sayTakeover(ctx, epoch)
}
return out
}
// handActsWithin is how far back a repeated cause counts (S15).
+32 -1
View File
@@ -491,13 +491,41 @@ func TestAStaleWriterIsNamedByItsEpoch(t *testing.T) {
}
}
// **The lease lost is said for an hour, and serving without it for as long as it lasts** (S12).
// **The lease lost is said for an hour while nobody took it over, a takeover is said once as it happens —
// raised and cleared by the controller that took it — a lease given back is never said, and serving without
// it for as long as it lasts** (S12, novox/hq issue 906).
func TestALeaseLostOrMissingIsSaid(t *testing.T) {
now := time.Now()
f := calm(now)
expired := now.Add(-59 * time.Minute)
f.lease.ended = []inventory.Epoch{{Epoch: 41, Instance: "controller@anchor pid 7", Host: "anchor",
Ended: &expired, How: inventory.EpochExpired}}
// Held at epoch 57, above it: the takeover said it, and no hour of observation follows.
if got := watchLease(f); len(got) != 0 {
t.Fatalf("an epoch this controller took over is observed after the takeover: %+v", got)
}
store := conditions.NewInMemory()
k := conditions.NewKeeper(t.Context(), conditions.Options{Store: store, History: store, Now: func() time.Time { return now }})
sayTakeoverOf(t.Context(), k, f.lease.ended, 57, now)
if open, err := k.Open(t.Context()); err != nil || len(open) != 0 {
t.Fatalf("the takeover is left open: %+v, %v", open, err)
}
k.Close(t.Context()) // its transitions written to the history
history, err := store.Since(t.Context(), now.Add(-time.Minute))
if err != nil {
t.Fatal(err)
}
var changes []string
for _, e := range history {
if e.Key == "core.controller.lease.lost" {
changes = append(changes, e.Change)
}
}
if !slices.Equal(changes, []string{"raised", "cleared"}) {
t.Fatalf("the takeover of an expired lease is not raised and cleared: %v", changes)
}
// Not holding the lease, the loss is said for the hour.
f.lease.held = false
got := watchLease(f)
if len(got) != 1 || got[0].Key() != "core.controller.lease.lost" || !strings.Contains(got[0].Summary, "epoch 41") ||
got[0].Severity != conditions.Urgent {
@@ -512,6 +540,9 @@ func TestALeaseLostOrMissingIsSaid(t *testing.T) {
if got := watchLease(f); len(got) != 0 {
t.Fatalf("a lease given back is said as lost: %+v", got)
}
if _, said := takeover(f.lease.ended, 57, now); said {
t.Fatal("the takeover of a lease given back is said")
}
f.lease = leaseFacts{held: true, epoch: 501, renewed: now, reset: now.Add(-time.Minute), resetSaid: "moved past 500"}
if got := watchLease(f); len(got) != 1 || got[0].Key() != "core.controller.lease.reset" {
t.Fatalf("a lease bucket raised again from nothing was not said: %+v", got)
+1 -1
View File
@@ -3,7 +3,7 @@ module github.com/novox/mesh-controller
go 1.26.0
require (
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261024234636-10b6e202320f
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261029231453-6e4a3014cbde
github.com/jackc/pgx/v5 v5.10.0
github.com/nats-io/nats-server/v2 v2.11.17
github.com/nats-io/nats.go v1.54.0
+2 -2
View File
@@ -1,7 +1,7 @@
git.novox.be/novox/mesh-host v0.0.0-20261018171808-9d9e3a46c4f5 h1:E5stfIeCN6uYgA9TCv+4K8JgaURg3zQppvUgJLaDMP4=
git.novox.be/novox/mesh-host v0.0.0-20261018171808-9d9e3a46c4f5/go.mod h1:4wyoCr6IIIIarEUuROThWDj7DAvJvIidqMjWcpSnOwE=
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261024234636-10b6e202320f h1:lUEyNPPF3w4Q0LIgcUUD6ofkKVhCF0n7dVjMoixQmXo=
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261024234636-10b6e202320f/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261029231453-6e4a3014cbde h1:ddf/jKI9c87sQKDmZLoG6PMnvXwsNH16ByVveyuFME4=
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261029231453-6e4a3014cbde/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op h1:Z/MZK75wC/NSrkgqeNIa7jexam9uWzhLmFTSCPI/kn0=
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op/go.mod h1:FQyySiasQQM8735Ddel3MRojmy4dA1IqCeyJ5jmPMbI=
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+9 -7
View File
@@ -46,9 +46,8 @@ func TestAConsumerMadeFromNowNeverReplaysTheStreamsHistory(t *testing.T) {
t.Fatalf("a consumer made from now holds %d of the stream's past (policy %v)", info.NumPending, info.Config.DeliverPolicy)
}
_, _ = js.js.Publish("history.merged", []byte("new"))
time.Sleep(100 * time.Millisecond)
if info, _ = js.js.ConsumerInfo(stream, "fresh"); info.NumPending != 1 {
t.Fatalf("a new announcement is not pending: %d", info.NumPending)
if got := pendingOf(js, stream, "fresh", 1); got != 1 {
t.Fatalf("a new announcement is not pending: %d", got)
}
// Asserted again, it keeps where it is.
if err := js.EnsureConsumer(fresh); err != nil {
@@ -61,8 +60,8 @@ func TestAConsumerMadeFromNowNeverReplaysTheStreamsHistory(t *testing.T) {
if err := js.EnsureConsumer(old); err != nil {
t.Fatal(err)
}
if info, _ = js.js.ConsumerInfo(stream, "old"); info.NumPending != 6 {
t.Fatalf("the default should replay all six: %d", info.NumPending)
if got := pendingOf(js, stream, "old", 6); got != 6 {
t.Fatalf("the default should replay all six: %d", got)
}
before, after, err := js.ResetConsumer(stream, "old")
if err != nil {
@@ -89,10 +88,13 @@ func TestAConsumerMadeFromNowNeverReplaysTheStreamsHistory(t *testing.T) {
}
}
// pendingOf polls a consumer's pending count until it is want or a second has passed, and returns what it saw last.
// pendingOf polls a consumer's pending count until it is want or ten seconds have passed, and returns what it saw
// last. The server counts a consumer's pending after it acknowledges the publish, not before, so a test that reads
// the count at once fails on a loaded build node (a try of main on ace, 2026-10-26): every test that expects a
// count after a publish reads it through here.
func pendingOf(js *JetStream, stream, name string, want uint64) uint64 {
var got uint64
for deadline := time.Now().Add(time.Second); ; time.Sleep(20 * time.Millisecond) {
for deadline := time.Now().Add(10 * time.Second); ; time.Sleep(20 * time.Millisecond) {
if info, err := js.js.ConsumerInfo(stream, name); err == nil {
got = info.NumPending
}
@@ -73,15 +73,15 @@ func TestAResetOfTheEventsConsumerLeavesTheMergesAndTheirConsumerAlone(t *testin
if _, err := js.Context().Publish(ControllerFollows[0], []byte(`{}`)); err != nil {
t.Fatal(err)
}
merges, err := js.Context().ConsumerInfo(EventsStream, MergesConsumerName)
if err != nil || merges.NumPending != 1 {
t.Fatalf("the merges consumer holds %v (%v); want the merge", merges, err)
if got := pendingOf(js, EventsStream, MergesConsumerName, 1); got != 1 {
t.Fatalf("the merges consumer holds %d pending; want the merge", got)
}
if _, _, err := js.ResetConsumer(EventsStream, ControllerName); err != nil {
t.Fatal(err)
}
merges, err = js.Context().ConsumerInfo(EventsStream, MergesConsumerName)
if err != nil || merges.NumPending != 1 || merges.Delivered.Stream != 0 {
pending := pendingOf(js, EventsStream, MergesConsumerName, 1)
merges, err := js.Context().ConsumerInfo(EventsStream, MergesConsumerName)
if err != nil || pending != 1 || merges.Delivered.Stream != 0 {
t.Fatalf("after the reset the merges consumer holds %+v (%v); want its merge left alone", merges, err)
}
}
@@ -107,9 +107,8 @@ func TestAMergeTheOldConsumerHadNotAcknowledgedIsTheMergesConsumers(t *testing.T
if err := AssertMeshConsumers(js); err != nil {
t.Fatal(err)
}
merges, err := js.Context().ConsumerInfo(EventsStream, MergesConsumerName)
if err != nil || merges.NumPending != 1 {
t.Fatalf("the merges consumer holds %v (%v); want the merge the old consumer never acknowledged", merges, err)
if got := pendingOf(js, EventsStream, MergesConsumerName, 1); got != 1 {
t.Fatalf("the merges consumer holds %d pending; want the merge the old consumer never acknowledged", got)
}
}
+106
View File
@@ -0,0 +1,106 @@
package broker
import "testing"
// Defends issue 1790 and novox/hq ADR 0547 §5 (the operator, 2026-10-26 08:54): the grants part may create the stream
// of each bucket, log and stream a placed module's manifest declares, by its own name, and read it back; nothing
// wider, nothing updated or deleted, and no other principal so. Held byte-equal in the controller's copy. And ADR 0547
// clause 7 (issue 1945, the operator's second exception): the grants part writes its own object bucket of sealed user
// lists, and the node-engine of the node holding mesh-broker alone reads its one key and an object of it by name.
func TestTheGrantsPartCreatesOnlyDeclaredStreams(t *testing.T) {
records := Records{
Nodes: []string{"novox", "ace"},
Assigned: map[string][]Declared{"ace": {{Module: "plex"}}, "novox": {
{Module: "nats", SnapshotsTheBus: true},
{Module: "mesh-delivery-builder",
State: []Bucket{{Module: "mesh-delivery-builder", Name: "steps", History: 5}},
Logs: []Log{{Module: "mesh-delivery-builder", Name: "builds"}},
Streams: []ModuleStream{{Module: "mesh-delivery-builder", Name: "BUILDASKS", Events: []string{"ask"}}}},
{Module: GrantsModule, State: []Bucket{{Module: GrantsModule, Name: "assertions"}}},
}},
}
users, err := Users(records)
if err != nil {
t.Fatal(err)
}
perms := map[string]Permissions{}
for _, u := range users {
if u.Kind == KindNode {
p, err := PermissionsFor(u)
if err != nil {
t.Fatalf("%s: %v", u.Username(), err)
}
perms[u.Username()] = p
continue
}
if u.Kind != KindModule {
continue
}
p, err := PermissionsFor(u)
if err != nil {
t.Fatalf("%s: %v", u.Username(), err)
}
perms[u.Module] = p
}
for _, row := range []struct {
who, subject string
may bool
}{
{GrantsModule, "$JS.API.STREAM.CREATE.KV_mesh-delivery-builder_steps", true},
{GrantsModule, "$JS.API.STREAM.INFO.KV_mesh-delivery-builder_steps", true},
{GrantsModule, "$JS.API.STREAM.CREATE.LOG_mesh-delivery-builder_builds", true},
{GrantsModule, "$JS.API.STREAM.CREATE.BUILDASKS", true},
{GrantsModule, "$JS.API.STREAM.CREATE.KV_mesh-grants_assertions", true},
{GrantsModule, "$JS.API.STREAM.UPDATE.KV_mesh-delivery-builder_steps", false},
{GrantsModule, "$JS.API.STREAM.DELETE.KV_mesh-delivery-builder_steps", false},
{GrantsModule, "$JS.API.STREAM.PURGE.KV_mesh-delivery-builder_steps", false},
{GrantsModule, "$JS.API.STREAM.CREATE.KV_mesh-delivery-builder_undeclared", false},
{GrantsModule, "$JS.API.STREAM.CREATE.EVENTS", false},
{GrantsModule, "$JS.API.STREAM.CREATE.ASSIGNMENTS", false},
{GrantsModule, "$JS.API.STREAM.INFO.EVENTS", false},
{"mesh-delivery-builder", "$JS.API.STREAM.CREATE.KV_mesh-delivery-builder_steps", false},
// The sealed user list (ADR 0547 clause 7): the grants part writes its own bucket and no other.
{GrantsModule, "$JS.API.STREAM.CREATE.OBJ_mesh-grants-user-list", true},
{GrantsModule, "$O.mesh-grants-user-list.C.abc", true},
{GrantsModule, "$JS.API.STREAM.PURGE.OBJ_mesh-grants-user-list", true},
{GrantsModule, "$O.declaration-files-novox.C.abc", false},
{GrantsModule, "$JS.API.STREAM.DELETE.OBJ_mesh-grants-user-list", false},
{"mesh-delivery-builder", "$O.mesh-grants-user-list.C.abc", false},
// The bus node's engine reads the one key and an object by name; nothing of the bucket's list, no write.
{"node.novox", "$JS.API.DIRECT.GET.KV_mesh-grants_user-list.$KV.mesh-grants_user-list.sealed", true},
{"node.novox", "$JS.API.CONSUMER.CREATE.KV_mesh-grants_user-list.abc.$KV.mesh-grants_user-list.sealed", true},
{"node.novox", "$JS.API.DIRECT.GET.OBJ_mesh-grants-user-list.$O.mesh-grants-user-list.M.abc", true},
{"node.novox", "$JS.API.CONSUMER.CREATE.OBJ_mesh-grants-user-list.abc.$O.mesh-grants-user-list.C.abc", true},
{"node.novox", "$JS.API.DIRECT.GET.KV_mesh-grants_user-list.$KV.mesh-grants_user-list.list", false},
{"node.novox", "$JS.API.STREAM.INFO.OBJ_mesh-grants-user-list", false},
{"node.novox", "$JS.API.STREAM.MSG.GET.OBJ_mesh-grants-user-list", false},
{"node.novox", "$O.mesh-grants-user-list.C.abc", false},
{"node.ace", "$JS.API.DIRECT.GET.KV_mesh-grants_user-list.$KV.mesh-grants_user-list.sealed", false},
{"node.ace", "$JS.API.DIRECT.GET.OBJ_mesh-grants-user-list.$O.mesh-grants-user-list.M.abc", false},
} {
if got := MayPublish(perms[row.who], row.subject); got != row.may {
t.Errorf("%s may publish %s: %v, want %v", row.who, row.subject, got, row.may)
}
}
// A name that is no plain token is never granted, and a module other than the grants part is never granted one.
if got := CreateStreamGrants([]string{"*", ">", "KV_a.b", "KV_a_b"}); len(got) != 2 {
t.Errorf("grants for wildcard names: %v", got)
}
other, err := PermissionsFor(Principal{Kind: KindModule, Node: "novox", Module: "shop", CreatesStreams: []string{"KV_shop_x"}})
if err != nil {
t.Fatal(err)
}
if MayPublish(other, "$JS.API.STREAM.CREATE.KV_shop_x") {
t.Error("a module other than the grants part was granted a create")
}
// The writers table refuses the grants part an update, and a create of a stream it was not given.
g := Principal{Kind: KindModule, Node: "novox", Module: GrantsModule, CreatesStreams: []string{"KV_a_b"}}
if err := CheckWriters(g, []string{"$JS.API.STREAM.CREATE.KV_a_b"}); err != nil {
t.Errorf("the declared create refused: %v", err)
}
for _, refused := range []string{"$JS.API.STREAM.UPDATE.KV_a_b", "$JS.API.STREAM.CREATE.KV_a_c", "$JS.API.STREAM.CREATE.*"} {
if CheckWriters(g, []string{refused}) == nil {
t.Errorf("%s was not refused", refused)
}
}
}
+4 -1
View File
@@ -213,8 +213,11 @@ func TestNoPathDeletesALogOrABucket(t *testing.T) {
// Only the writers table names it, as a subject no one but the controller may publish.
// And the operator's release 6 hand path, whose object store purges a replaced object's chunks in a node's
// object bucket (novox/hq ADR 0539 clause 10): a grant, never a call of this repository.
// And the grants part's own object bucket of sealed user lists, whose put purges a replaced object's chunks
// (novox/hq ADR 0547 clause 7): a grant, never a call of this repository.
if !strings.HasSuffix(filepath.ToSlash(path), "internal/broker/writers.go") &&
!strings.HasSuffix(filepath.ToSlash(path), "internal/broker/composed.go") {
!strings.HasSuffix(filepath.ToSlash(path), "internal/broker/composed.go") &&
!strings.HasSuffix(filepath.ToSlash(path), "internal/broker/userlist_switch.go") {
found = append(found, path+": "+lit)
}
}
+11
View File
@@ -72,6 +72,17 @@ func TestAnAccountMayReadItsOwnMembershipAndNoOthers(t *testing.T) {
has(t, perms.Subscribe, "mesh.assignment.anchor.postgres")
has(t, perms.Publish, "$JS.API.DIRECT.GET.ASSIGNMENTS.mesh.assignment.anchor.postgres")
hasNot(t, perms.Subscribe, "mesh.assignment.>")
hasNot(t, perms.Publish, "mesh.assignment.*.*")
// The grants part alone publishes memberships (ADR 0547 §5), and defines no stream.
grants, err := PermissionsFor(Principal{Kind: KindModule, Node: "anchor", Module: GrantsModule, PasswordHash: "x"})
if err != nil {
t.Fatal(err)
}
has(t, grants.Publish, "mesh.assignment.*.*")
has(t, grants.Publish, "$JS.API.DIRECT.GET.ASSIGNMENTS.mesh.assignment.*.*")
hasNot(t, perms.Publish, "$JS.API.DIRECT.GET.ASSIGNMENTS.mesh.assignment.*.*")
hasNot(t, grants.Publish, "$JS.API.STREAM.CREATE.>")
hasNot(t, grants.Publish, "mesh.assignment.>")
}
// The runtime arriving on a machine changes nothing about what each module is issued (to-be 38 WP2):
+2 -3
View File
@@ -52,9 +52,8 @@ func TestAModuleStreamOnARealBus(t *testing.T) {
t.Fatalf("publishing %s's mail: %v", account, err)
}
}
time.Sleep(50 * time.Millisecond)
if info, _ := js.js.ConsumerInfo("MAIL", c.Name); info.NumPending != 1 {
t.Fatalf("the reader granted `mesh` has %d pending, want 1", info.NumPending)
if got := pendingOf(js, "MAIL", c.Name, 1); got != 1 {
t.Fatalf("the reader granted `mesh` has %d pending, want 1", got)
}
narrowed := c
narrowed.Filters = []string{"mesh.stream.mesh-mail.event.mail.mesh.message.changed"}
+42
View File
@@ -214,9 +214,17 @@ type Principal struct {
// StatesItsKeyToTheVault is a machine principal whose node-engine may ask the vault's `node-key` in its own name
// (issue 786, novox/hq ADR 0372): set while the vault's holder serves it. That one verb, and nothing else new.
StatesItsKeyToTheVault bool
// FollowsGrantsUserList is the machine principal of the node holding mesh-broker: its node-engine reads the grants
// part's sealed user list, one key and an object by name (novox/hq ADR 0547 clause 7, GrantsUserListFollows).
FollowsGrantsUserList bool
// BoundBuckets are the providers' buckets of entries, `<module>_bound`, a provisioner client reads its own
// machine's keys of (mesh-sdk go/provider, issue 679): every bucket named bound of a module the mesh places.
BoundBuckets []string
// CreatesStreams are the streams the grants part's principal alone may create (novox/hq ADR 0547 §5; the operator,
// 2026-10-26 08:54): the stream of every bucket, log and stream a manifest of a module placed anywhere declares,
// by name. Created only, never updated or deleted: a new module state is on the bus when the grants part hears its
// manifest, not at the controller's next start (issue 1790).
CreatesStreams []string
// PasswordHash is the bcrypt hash the mesh minted. The plaintext is sealed to the principal
// and never appears here: this file is written to a node's disk and read by a server, and a
@@ -908,6 +916,11 @@ func PermissionsFor(p Principal) (Permissions, error) {
if p.StatesItsKeyToTheVault {
pub = append(pub, "mesh.seat."+VaultNodeKey.Seat+"."+CallKind+"."+VaultNodeKey.Verb+"."+CallerToken(p.Username()))
}
// And, on the node holding mesh-broker, the grants part's sealed user list it follows from take-over (novox/hq ADR
// 0547 clause 7): one key, and an object by its name.
if p.FollowsGrantsUserList {
pub = append(pub, GrantsUserListFollows()...)
}
case KindModule:
// 0. The watchman alone hears who connects and disconnects, and asks who is connected (ADR 0431 §2).
@@ -915,6 +928,17 @@ func PermissionsFor(p Principal) (Permissions, error) {
sub = append(sub, BusConnectSubject, BusDisconnectSubject)
pub = append(pub, BusConnzSubject)
}
// The grants part alone publishes the modules' memberships into the assignments stream (novox/hq ADR 0547 §5;
// the operator, 2026-10-25 22:31): one per subject, `mesh.assignment.<node>.<module>`, reading the last one of each
// first so it publishes only what changed, and no stream it defines.
if p.Module == GrantsModule {
pub = append(pub, "mesh.assignment.*.*", "$JS.API.DIRECT.GET."+AssignmentsStream+".mesh.assignment.*.*")
// And creating each stream a manifest declares, by its own name (the operator, 2026-10-26 08:54; issue
// 1790): the create and the read of that one stream, never an update, a delete or a stream not declared.
pub = append(pub, CreateStreamGrants(p.CreatesStreams)...)
// And its own object bucket, where it writes the user list sealed to the bus node (ADR 0547 clause 7).
pub = append(pub, GrantsUserListWrites()...)
}
// The view alone reads the build seats' workers for the backlog of asks (issue 1887 A8a).
if p.Module == ViewModule {
pub = append(pub, BuildBacklogReads()...)
@@ -1581,6 +1605,24 @@ func ComposeAccounts(principals []Principal) (string, error) {
// from them. No other principal is granted the subjects below, and nothing is a wildcard over them.
const WatchmanModule = "mesh-delivery-watchman"
// CreateStreamGrants is what creating each named stream needs, and nothing wider (novox/hq ADR 0547 §5; the operator,
// 2026-10-26 08:54): its create, and its information to read back what it holds. Each by its own name: a name that
// is not one plain subject token is left out, since a wildcard there would be every stream.
func CreateStreamGrants(streams []string) []string {
var out []string
for _, s := range streams {
if !safeSubject.MatchString(s) {
continue
}
out = append(out, "$JS.API.STREAM.CREATE."+s, "$JS.API.STREAM.INFO."+s)
}
return out
}
// GrantsModule is the one module whose principal publishes the modules' memberships (novox/hq ADR 0547 §5): the grants
// part, which composes them from the manifests and placement. It defines no stream and is granted no other subject here.
const GrantsModule = "mesh-grants"
// The bus's own events, imported into MESH from the default system account `$SYS` under subjects of MESH's own
// (novox/hq ADR 0431 §1, issue 1147): the server says an account's connections opening and closing only to the system
// account, and the list of an account's connections only to a request made there. `$SYS` holds no user, so nobody
+20 -2
View File
@@ -24,8 +24,9 @@ import (
// subjects cover it, so what travels there — a code typed by the operator — is never persisted. A kinded
// holder asks with its own kind; the modules that watch the seat answer.
//
// And one read: **a holder's records**, a bucket the seat names, read by each user under its own name only
// (`$KV.<bucket>.<module>.>`), so an asker reads the state of its own asks and no other asker's.
// And two reads: **a holder's records**, a bucket the seat names, read by each user under its own name only
// (`$KV.<bucket>.<module>.>`), so an asker reads the state of its own asks and no other asker's; and on a seat pulled by
// kind, **what its workers gave up on**, the copies DEAD_LETTERS keeps of the seat's stream (DeadLetterReads, issue 734).
// BuildAskMaxDeliver bounds the deliveries of a build seat's ask (issue 1907): mesh-sdk delivery.AskMaxDeliver, written
// here as the broker mirrors the catalogue rather than importing it.
@@ -129,6 +130,9 @@ func SeatTrafficOf(module string, holds, uses, watches []Seat) SeatTraffic {
}
t.Records = append(t.Records, "$JS.API.DIRECT.GET.KV_"+b+".$KV."+b+"."+module+".>")
}
if s.PulledByKind && len(s.Accepts) > 0 && safeSubject.MatchString(s.Name) {
t.Records = append(t.Records, DeadLetterReads(seatStreamName(s.Name))...)
}
}
for _, w := range watches {
if w.Kinded {
@@ -151,6 +155,20 @@ func SeatTrafficOf(module string, holds, uses, watches []Seat) SeatTraffic {
return t
}
// DeadLetterReads is what a user of a seat pulled by kind is granted to hear which of the asks it put the seat's workers
// gave up on (novox/hq issue 734): an ordered consumer on DEAD_LETTERS filtered to the copies the controller keeps of that
// seat's stream (DeadLetterSubject(stream, *)), pulled, and its deletion. A copy carries the ask's own message id
// (Mesh-Dead-Msg-Id), so the asker ends the step it waits for instead of waiting for ever; it reads the seat's dead
// letters and no other stream's, and changes none.
func DeadLetterReads(stream string) []string {
return []string{
"$JS.API.STREAM.INFO." + DeadLettersStream,
"$JS.API.CONSUMER.CREATE." + DeadLettersStream + ".*." + DeadLetterSubject(stream, ">"),
"$JS.API.CONSUMER.MSG.NEXT." + DeadLettersStream + ".*",
"$JS.API.CONSUMER.DELETE." + DeadLettersStream + ".*",
}
}
// seatTrafficGrants is the bus permissions seat traffic needs: the subjects themselves, and the JetStream API a
// worker is pulled and acknowledged through and a record is read through.
func seatTrafficGrants(t SeatTraffic) (pub, sub []string) {
+13
View File
@@ -403,6 +403,19 @@ func TestTheBuilderMayAskABuildSeatAndASessionMayNot(t *testing.T) {
t.Errorf("the builder may not publish %s", s)
}
}
// Issue 734: the asker hears which of its asks the seat's workers gave up on, from DEAD_LETTERS, and no other stream's.
for subject, want := range map[string]bool{
"$JS.API.CONSUMER.CREATE.DEAD_LETTERS.x1.mesh.events.dead.SEAT_NODE_BUILD_GO.>": true,
"$JS.API.CONSUMER.DELETE.DEAD_LETTERS.x1": true,
"$JS.API.CONSUMER.MSG.NEXT.DEAD_LETTERS.x1": true,
"$JS.API.CONSUMER.CREATE.DEAD_LETTERS.x1.mesh.events.dead.EVENTS.>": false,
"$JS.API.STREAM.PURGE.DEAD_LETTERS": false,
"$JS.FC.DEAD_LETTERS.x1": false,
} {
if got := allowed(builder.Publish, subject); got != want {
t.Errorf("the builder may publish %s: %v, want %v", subject, got, want)
}
}
session := perms(t, Principal{Kind: KindPerson, Module: "jochen", Invokes: []string{"mesh-controller.status"}})
if allowed(session.Publish, "mesh.seat.node-build-go.accept.delivery") {
t.Error("a session's account may put an ask")
+55
View File
@@ -0,0 +1,55 @@
package broker
// **The grants part's user list, sealed to the bus node** (novox/hq ADR 0547 clause 7, issue 1945): the operator's
// second exception to the frozen controller (the first, memberships, 2026-10-25 22:31). The controller's copy of the bus
// composer grants what mesh-sdk's copy grants, byte for byte, and nothing else: the node-engine of the node holding
// mesh-broker reads the grants part's one key `user-list.sealed`, watched, and an object of its bucket by name; the
// grants part writes that bucket. Without it the bus node could never read the first list the grants part writes, since
// until take-over the list the bus loads is this one (mesh-sdk go/buscompose/userlist_switch.go keeps the other copy).
// The grants part's user list on the bus (mesh-sdk go/delivery userlist.go names them).
const (
GrantsUserListBucket = "mesh-grants_user-list"
GrantsUserListKey = "sealed"
GrantsUserListObjects = "mesh-grants-user-list"
GrantsUserListObjectStream = "OBJ_" + GrantsUserListObjects
)
// GrantsUserListFollows is what the node-engine of the node holding mesh-broker may publish to follow the grants part's
// list: the one key read and watched, and an object of the bucket by its name (as ADR 0486 §2 reads a declaration's
// files: never the object stream's info, which lists every name, never a message by sequence).
func GrantsUserListFollows() []string {
kv := "KV_" + GrantsUserListBucket
key := "$KV." + GrantsUserListBucket + "." + GrantsUserListKey
obj, s := GrantsUserListObjects, GrantsUserListObjectStream
return []string{
"$JS.API.DIRECT.GET." + kv + "." + key,
"$JS.API.CONSUMER.CREATE." + kv + ".*." + key,
"$JS.API.CONSUMER.DELETE." + kv + ".*",
"$JS.API.STREAM.INFO." + kv,
"$JS.FC." + kv + ".>",
"$JS.API.DIRECT.GET." + s + ".$O." + obj + ".M.*",
"$JS.API.CONSUMER.CREATE." + s + ".*.$O." + obj + ".C.*",
"$JS.API.CONSUMER.DELETE." + s + ".*",
"$JS.FC." + s + ".>",
}
}
// GrantsUserListWrites is what the grants part may publish to keep its object bucket.
func GrantsUserListWrites() []string {
obj, s := GrantsUserListObjects, GrantsUserListObjectStream
return []string{
"$JS.API.STREAM.CREATE." + s,
"$JS.API.STREAM.INFO." + s,
"$JS.API.STREAM.PURGE." + s,
"$JS.API.DIRECT.GET." + s + ".>",
"$JS.API.STREAM.MSG.GET." + s,
"$O." + obj + ".>",
}
}
// grantsOwnObjects is the grants part writing its own object bucket, and nothing of another.
func grantsOwnObjects(p Principal, pattern string) bool {
return p.Kind == KindModule && p.Module == GrantsModule &&
(pattern == "$O."+GrantsUserListObjects+".>" || pattern == "$JS.API.STREAM.CREATE."+GrantsUserListObjectStream)
}
+37 -1
View File
@@ -147,14 +147,19 @@ func Users(r Records) ([]Principal, error) {
for _, node := range sortedCopy(r.Nodes) {
witness := false
var checks []string
bus := false
for _, d := range r.Assigned[node] {
if d.Module == controllerModule {
witness = true
}
if d.SnapshotsTheBus {
bus = true
}
checks = append(checks, d.Checks...)
}
out = append(out, Principal{Kind: KindNode, Node: node, WitnessesController: witness, Checks: checks,
MovesIntoTheVault: servesMoveIn(r), StatesItsKeyToTheVault: servesVerb(r, VaultNodeKey)})
MovesIntoTheVault: servesMoveIn(r), StatesItsKeyToTheVault: servesVerb(r, VaultNodeKey),
FollowsGrantsUserList: bus})
// **And the machine's provisioner client, on an account of its own** (novox/hq ADR 0349 clause 11): one
// per machine that runs it, delivered to the module running it.
for _, d := range r.Assigned[node] {
@@ -207,6 +212,11 @@ func Users(r Records) ([]Principal, error) {
PerMachine: perMachineNames(d.State), KeyedReads: d.KeyedReads,
Logs: logNames(d.Logs), Streams: d.Streams, Hears: ComposedHears(d), Queues: d.Queues,
})
// **The grants part creates what a manifest declares** (novox/hq ADR 0547 §5; the operator, 2026-10-26
// 08:54): every placed module's buckets, logs and streams, each by its stream's name.
if d.Module == GrantsModule {
out[len(out)-1].CreatesStreams = declaredStreamsOf(r)
}
// **The vault's holder asks the desks and the issuing providers itself** (issue 723), under the
// account its own runtime runs on (ADR 0259 §8's runs-as): the one module granted it.
if holdsSeat(d, VaultSeat) {
@@ -418,3 +428,29 @@ func boundBucketsOf(r Records) []string {
func runnerServes(d Declared) bool {
return d.Module == RuntimeModule || d.NoAccount || d.Loads || len(d.Serves) > 0 || len(d.Holds) > 0 || len(d.Waits) > 0 || len(d.Checks) > 0
}
// declaredStreamsOf is the stream of every bucket, log and stream a module placed anywhere declares, once each, in name
// order (novox/hq ADR 0547 §5, issue 1790): `KV_<module>_<state>`, `LOG_<module>_<log>` and each module stream's own
// name. What the grants part may create, and nothing beside.
func declaredStreamsOf(r Records) []string {
seen := map[string]bool{}
for _, mods := range r.Assigned {
for _, d := range mods {
for _, b := range d.State {
seen["KV_"+b.Bucket()] = true
}
for _, l := range d.Logs {
seen[l.Stream()] = true
}
for _, s := range d.Streams {
seen[s.Name] = true
}
}
}
out := make([]string, 0, len(seen))
for s := range seen {
out = append(out, s)
}
sort.Strings(out)
return out
}
+20 -2
View File
@@ -37,6 +37,20 @@ type WriterRow struct {
// isController, and the other writers' tests, are who a row's writer is as a principal.
func isController(p Principal, _ string) bool { return p.Kind == KindController }
// createsDeclared is the controller, or the grants part creating one stream a manifest declares, by its own name: the
// create alone, never an update or a delete (novox/hq ADR 0547 §5; the operator, 2026-10-26 08:54).
func createsDeclared(p Principal, pattern string) bool {
if isController(p, pattern) {
return true
}
name, ok := strings.CutPrefix(pattern, "$JS.API.STREAM.CREATE.")
if ok && p.Kind == KindModule && p.Module == GrantsModule && name == GrantsUserListObjectStream {
return true
}
return ok && p.Kind == KindModule && p.Module == GrantsModule && slices.Contains(p.CreatesStreams, name) &&
safeSubject.MatchString(name)
}
// ownMachine is a node writing a subject whose third token is its own name, and no wildcard there.
func ownMachine(p Principal, pattern string) bool {
tokens := strings.Split(pattern, ".")
@@ -170,7 +184,8 @@ var WritersTable = []WriterRow{
// bucket's stream (ADR 0201), which defines nothing the mesh keeps.
Subjects: []string{"$JS.API.STREAM.CREATE.>", "$JS.API.STREAM.UPDATE.>", "$JS.API.STREAM.DELETE.>",
"$JS.API.CONSUMER.DURABLE.CREATE.>"},
Writes: isController},
Writes: createsDeclared,
Shared: "the grants part creates a stream a manifest declares, by its name, and never updates or deletes one (novox/hq ADR 0547 §5; the operator, 2026-10-26 08:54)"},
{State: "builds and their outcomes", Writer: "the build seat's holder", KeptIn: "its own state",
Others: "the controller asks",
Subjects: []string{"mesh.seat.node-build-agent.event.built", "mesh.seat.mesh-build-machine.event.built"},
@@ -237,7 +252,10 @@ var WritersTable = []WriterRow{
KeptIn: "the object bucket " + DeclarationFilesBucket + "-<node> of each node, an object removed a day after the " +
"node's latest declaration no longer names it",
Others: "a node-engine reads an object of its own bucket by its name and may neither list the bucket nor write it",
Subjects: []string{"$O.*.>"}, Writes: isControllerOrTheHandPath},
Subjects: []string{"$O.*.>"}, Writes: func(p Principal, pattern string) bool {
return isControllerOrTheHandPath(p, pattern) || grantsOwnObjects(p, pattern)
},
Shared: "the grants part writes its own object bucket " + GrantsUserListObjects + ", the user list sealed to the bus node, and no other (novox/hq ADR 0547 clause 7)"},
{State: "the declared-config mirror", Writer: "controller (lease holder), from its store, only on a change",
KeptIn: "key-value " + DeclaredConfigBucket, Others: "read through `reads` as mesh-controller." + MirrorDeclaredConfig,
Subjects: kvOf(DeclaredConfigBucket), Writes: isController},
+12 -3
View File
@@ -67,14 +67,23 @@ func (i *Inventory) HighestEpoch(ctx context.Context) (uint64, error) {
}
// TookEpoch records an epoch taken, and ends every earlier one still open as found expired: the lease
// was free to take, so whoever held it last neither holds it nor said it let go. Answers the epochs it
// ended that way, newest first.
func (i *Inventory) TookEpoch(ctx context.Context, e Epoch) ([]Epoch, error) {
// was free to take, so whoever held it last neither holds it nor said it let go. givenBack is the epoch the
// bus says gave the key back (the lease's own delete marker names it, novox/hq issue 906, ADR 0558 §3): that
// epoch, if still open, is ended as released — its holder gave the key back and had not yet recorded it — and
// every other as expired; zero is none. Answers the epochs it ended as expired, newest first.
func (i *Inventory) TookEpoch(ctx context.Context, e Epoch, givenBack uint64) ([]Epoch, error) {
tx, err := i.store.Pool().Begin(ctx)
if err != nil {
return nil, err
}
defer func() { _ = tx.Rollback(ctx) }()
if givenBack != 0 && givenBack < e.Epoch {
if _, err := tx.Exec(ctx,
`update controller_epoch set ended = now(), how = $2 where epoch = $1 and ended is null`,
int64(givenBack), EpochReleased); err != nil {
return nil, err
}
}
rows, err := tx.Query(ctx,
`update controller_epoch set ended = now(), how = $2
where ended is null and epoch < $1
+71 -9
View File
@@ -34,9 +34,11 @@ import (
"encoding/json"
"errors"
"fmt"
"strconv"
"sync"
"time"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
)
@@ -115,6 +117,7 @@ type Options struct {
// Lease is one instance's hold, or its wait for one.
type Lease struct {
kv jetstream.KeyValue
js jetstream.JetStream
stream jetstream.Stream
ttl time.Duration
o Options
@@ -139,6 +142,9 @@ type Lease struct {
lostWhy error
lost chan struct{}
taken time.Time
// givenBack is the epoch whose holder gave back the key this instance took — the epoch its delete marker
// names on the bus — and zero when the key was left to expire, or freed by anything else (novox/hq issue 906).
givenBack uint64
// update writes the key at a revision: the bucket's own, or a test's bus that does not answer.
update func(ctx context.Context, key string, value []byte, last uint64) (uint64, error)
}
@@ -184,7 +190,7 @@ func Open(ctx context.Context, js jetstream.JetStream, bucket string, o Options)
return nil, fmt.Errorf("a lease renewed every %s with a margin of %s does not fit a key that lives %s",
o.RenewEvery, o.Margin, status.TTL())
}
return &Lease{kv: kv, stream: stream, ttl: status.TTL(), o: o, lost: make(chan struct{}), update: kv.Update}, nil
return &Lease{kv: kv, js: js, stream: stream, ttl: status.TTL(), o: o, lost: make(chan struct{}), update: kv.Update}, nil
}
// TTL is how long the key lives unrenewed.
@@ -271,6 +277,7 @@ func (l *Lease) TryTake(ctx context.Context) (uint64, error) {
} else if found {
return 0, &heldBy{h}
}
givenBack := l.wasGivenBack(ctx)
if err := l.aboveTheFloor(ctx); err != nil {
return 0, err
}
@@ -307,13 +314,51 @@ func (l *Lease) TryTake(ctx context.Context) (uint64, error) {
"back to be taken again", revision, floor)
}
l.mu.Lock()
l.held, l.epoch, l.revision, l.taken = true, revision, revision, now
l.held, l.epoch, l.revision, l.taken, l.givenBack = true, revision, revision, now, givenBack
l.renewedAt(anchor)
l.mu.Unlock()
l.o.Say("took the controller lease at epoch %d (%s)", revision, h.Instance)
return revision, nil
}
// ReleasedHeader is the header of the delete marker a holder's Release writes, naming the epoch it gave back:
// the one mark that says a handover. A delete marker without it — a command's own lease given back, a take
// given back over the floor — names no epoch, so it says nothing about the holder before (novox/hq issue 906).
const ReleasedHeader = "Mesh-Lease-Released"
// wasGivenBack is the epoch the free key was given back by: the epoch its last message on the bus — the delete
// marker a holder's Release writes — names. Zero when the key expired (the bucket's age removed every message),
// when the marker names no epoch, and when it cannot be read: a takeover nobody can prove was a handover is
// said as one that was not (novox/hq issue 906, ADR 0558 §3).
func (l *Lease) wasGivenBack(ctx context.Context) uint64 {
if l.stream == nil {
return 0
}
last, err := l.stream.GetLastMsgForSubject(ctx, l.subject())
if err != nil || last == nil {
return 0
}
if op := last.Header.Get("KV-Operation"); op != "DEL" && op != "PURGE" {
return 0
}
epoch, err := strconv.ParseUint(last.Header.Get(ReleasedHeader), 10, 64)
if err != nil {
return 0
}
return epoch
}
// subject is the lease key's subject in its bucket's stream.
func (l *Lease) subject() string { return "$KV." + l.kv.Bucket() + "." + Key }
// GivenBack is the epoch whose holder gave back the key this instance took, zero when it was not given back:
// what the takeover was, read from the bus (novox/hq issue 906).
func (l *Lease) GivenBack() uint64 {
l.mu.Lock()
defer l.mu.Unlock()
return l.givenBack
}
// floor is the highest epoch issued, zero without a floor.
func (l *Lease) floor(ctx context.Context) (uint64, error) {
if l.o.Floor == nil {
@@ -588,9 +633,11 @@ func (l *Lease) lose(why error) error {
return l.lostWhy
}
// Release gives the lease back, so the next candidate takes it at once rather than after its age.
// Only the key this instance last wrote is deleted: one that moved is somebody else's.
func (l *Lease) Release(ctx context.Context) {
// Release gives the lease back, so the next candidate takes it at once rather than after its age, and answers
// the epoch it gave back — zero when it gave nothing back: it held none, had lost it, or the bus would not
// delete the key. Only the key this instance last wrote is deleted: one that moved is somebody else's. The
// gate closes before the key goes, so nothing this instance starts after overlaps its successor.
func (l *Lease) Release(ctx context.Context) uint64 {
l.mu.Lock()
held, revision, epoch := l.held && l.lostWhy == nil, l.revision, l.epoch
l.held = false
@@ -600,11 +647,11 @@ func (l *Lease) Release(ctx context.Context) {
}
l.mu.Unlock()
if !held {
return
return 0
}
releasing, cancel := context.WithTimeout(context.WithoutCancel(ctx), 5*time.Second)
defer cancel()
err := l.kv.Delete(releasing, Key, jetstream.LastRevision(revision))
err := l.giveBack(releasing, epoch, revision)
if err != nil && isWrongLast(err) {
// A renewal was in flight as this began and landed first: the key moved, and is still this
// instance's if it still names it at this epoch. Deleted at the revision it has now.
@@ -612,14 +659,29 @@ func (l *Lease) Release(ctx context.Context) {
var h Holder
if json.Unmarshal(entry.Value(), &h) == nil && h.Instance == l.o.Holder.Instance &&
(h.Epoch == epoch || h.Epoch == 0 && entry.Revision() == epoch) {
err = l.kv.Delete(releasing, Key, jetstream.LastRevision(entry.Revision()))
err = l.giveBack(releasing, epoch, entry.Revision())
}
}
}
if err != nil {
l.o.Say("the controller lease (epoch %d) could not be given back, so the next controller waits for "+
"it to expire: %v", epoch, err)
return
return 0
}
l.o.Say("gave the controller lease back (epoch %d)", epoch)
return epoch
}
// giveBack deletes the key at the revision this instance last wrote, as the bucket's own delete does — a
// marker with KV-Operation DEL, compare-and-set on the key's last revision — and names the epoch given back in
// the marker, so the next holder can tell this handover from any other delete (ReleasedHeader).
func (l *Lease) giveBack(ctx context.Context, epoch, revision uint64) error {
if l.js == nil {
return l.kv.Delete(ctx, Key, jetstream.LastRevision(revision))
}
msg := nats.NewMsg(l.subject())
msg.Header.Set("KV-Operation", "DEL")
msg.Header.Set(ReleasedHeader, strconv.FormatUint(epoch, 10))
_, err := l.js.PublishMsg(ctx, msg, jetstream.WithExpectLastSequencePerSubject(revision))
return err
}
+17 -3
View File
@@ -133,7 +133,12 @@ func TestTheSecondControllerWaitsAndTakesAHigherEpochOnHandover(t *testing.T) {
// a gives it back: b takes it at once, not after the key's age, at a higher epoch.
stop()
handedOver := time.Now()
a.Release(t.Context())
if given := a.Release(t.Context()); given != epochA {
t.Fatalf("a's release answers epoch %d given back, want %d", given, epochA)
}
if again := a.Release(t.Context()); again != 0 {
t.Fatalf("a second release gives epoch %d back again", again)
}
select {
case epochB := <-took:
if epochB <= epochA {
@@ -142,6 +147,9 @@ func TestTheSecondControllerWaitsAndTakesAHigherEpochOnHandover(t *testing.T) {
if waited := time.Since(handedOver); waited > testTTL {
t.Fatalf("b took the lease %s after a gave it back: it waited out the key's age", waited)
}
if given := b.GivenBack(); given != epochA {
t.Fatalf("b reads from the bus that epoch %d gave the key back, want a's %d (novox/hq issue 906)", given, epochA)
}
case <-time.After(2 * testTTL):
t.Fatal("b never took the lease a gave back")
}
@@ -182,6 +190,9 @@ func TestAHolderThatStopsRenewingStopsActingBeforeItIsTakenOver(t *testing.T) {
if epochB <= epochA {
t.Fatalf("b took epoch %d after a's %d", epochB, epochA)
}
if given := b.GivenBack(); given != 0 {
t.Fatalf("b reads a key that expired as given back by epoch %d (novox/hq issue 906)", given)
}
// And a, the moment it tries to renew, knows it lost: the key moved.
if err := a.Renew(t.Context()); err == nil {
t.Fatal("a renewed a lease b holds")
@@ -226,8 +237,11 @@ func TestARenewalRefusedIsALossAtOnce(t *testing.T) {
if _, err := a.Epoch(); !errors.Is(err, ErrNotHeld) {
t.Fatalf("a acts after its renewal was refused: %v", err)
}
// And giving back a lease it lost deletes nothing of the one who holds it now.
a.Release(t.Context())
// And giving back a lease it lost deletes nothing of the one who holds it now, and answers no epoch given
// back, so none is recorded released (novox/hq issue 906).
if given := a.Release(t.Context()); given != 0 {
t.Fatalf("a says it gave back epoch %d of a lease it lost", given)
}
if h, found, err := Current(t.Context(), kv); err != nil || !found || h.Instance != "intruder" {
t.Fatalf("after a gave back what it lost, the key holds %+v (%v, %v)", h, found, err)
}
+1
View File
@@ -82,6 +82,7 @@
"dead-letters",
"root-free",
"judge-directories",
"placement-export",
"data",
"build",
"artifacts",
+2 -2
View File
@@ -7,9 +7,9 @@ The copy carries the catalogue's private details: domains, first names in module
private addresses. That is acceptable while mesh-controller is private; if it ever goes public, this copy is
a second place to clean besides the catalogue itself.
mesh-catalog 3a1017f5594217092a654dd1e2d72226766bc16a modules/*/module.json, modules/*/files, modules/*/templates, predefined.json
mesh-catalog 8ae8e88dddc86c646de4d1be35919d109ab4c779 modules/*/module.json, modules/*/files, modules/*/templates, predefined.json
mesh-host 9d9e3a46c4f5943cfffb413d962ce48b75e2fd39 examples/foundation-first-node.lock
mesh-sdk 10b6e202320f38301c67a6822b0734e0b870ea30 conformance/events/module-event.json
mesh-sdk 6e4a3014cbde924dcd24158369cbe9ad287ce2c9 conformance/events/module-event.json
To move them, from this repository's root, with the two repositories checked out beside it:
@@ -0,0 +1 @@
${seat:mesh-artifact-store:reach}
@@ -1,26 +1,52 @@
{
"module": "artifact-store-tools",
"version": "1",
"claims": [
{
"name": "mesh-artifact-sweep",
"scope": "mesh",
"serves": [
"artifacts",
"held",
"collect",
"mirrors",
"images"
]
}
],
"requires": [
"artifact-store"
],
"invokes": [
"seat:mesh-controller.artifacts",
"seat:mesh-controller.collect",
"seat:mesh-hand-acts.hand-act",
"mesh-catalog.catalog_modules",
"mesh-catalog.catalog_module"
],
"reads": [
"mesh-delivery-assigner.wants",
"mesh-delivery-builder.builds",
"mesh-delivery-builder.build-log",
"mesh-delivery-indexer.manifests",
"node-engine.have",
"mesh-controller.lease"
],
"state": [
"findings"
],
"tools": [
"artifact_store_repositories",
"artifact_store_usage",
"artifact_store_references",
"artifact_store_collect"
],
"config": {
"older-store-names": {
"type": "string",
"description": "The names the artifact store had before, as host names or host:port separated by spaces, or none when it had no other. artifact_store_references reads every definition under these as well as under the store's current name, so an image a definition pins under one is said to be named by a definition (novox/hq issue 323).",
"pattern": "none|[A-Za-z0-9.:-]+( [A-Za-z0-9.:-]+)*",
"default": "none"
}
},
"config": {
"older-store-names": {
"type": "string",
"description": "The names the artifact store had before, as host names or host:port separated by spaces, or none when it had no other. artifact_store_references reads every definition under these as well as under the store's current name, so an image a definition pins under one is said to be named by a definition (novox/hq issue 323).",
"pattern": "none|[A-Za-z0-9.:-]+( [A-Za-z0-9.:-]+)*",
"default": "none"
}
},
"resources": [
{
"id": "state",
@@ -36,6 +62,14 @@
"mode": "0600",
"trusted": false,
"template": "files/store-names"
},
{
"id": "store-address",
"type": "file",
"path": "${dir:state}/store-address",
"mode": "0600",
"trusted": false,
"template": "files/store-address"
}
],
"build": {
@@ -52,7 +86,8 @@
],
"env": {
"MESH_ARTIFACT_STORE_CONTAINER": "mesh-registry",
"MESH_ARTIFACT_STORE_NAMES_FILE": "${dir:state}/store-names"
"MESH_ARTIFACT_STORE_NAMES_FILE": "${dir:state}/store-names",
"MESH_ARTIFACT_STORE_ADDRESS_FILE": "${dir:state}/store-address"
}
},
{
@@ -8,7 +8,10 @@
"provides": [
{
"name": "rootless-image-build",
"reach": "machine"
"reach": "machine",
"credential": {
"form": "none"
}
}
],
"resources": [
@@ -7,7 +7,9 @@
"disk_filesystems",
"disk_folders",
"disk_devices",
"cpu_top"
"cpu_top",
"disk_readings",
"disk_measure"
],
"build": {
"artifacts": [
@@ -23,5 +25,16 @@
]
}
]
}
},
"state": [
"readings"
],
"reads": [
"mesh-config.resolved",
"mesh-delivery-assigner.placement",
"mesh-delivery-builder.builds",
"mesh-delivery-indexer.manifests",
"mesh-delivery-indexer.modules",
"node-engine.have"
]
}
@@ -4,7 +4,10 @@
"provides": [
{
"name": "artifact-store",
"scope": "mesh"
"scope": "mesh",
"credential": {
"form": "none"
}
}
],
"claims": [
@@ -1,14 +1,15 @@
# Written by the mesh (module dnsmasq, novox/hq issue 348). Replaced on every push.
#
# The resolver binds its addresses once, at start (bind-interfaces), so it starts only once this
# machine's private address is there: it waits up to 80 seconds for it, and a start that found it
# missing is tried again every ten seconds rather than left failed.
# machine's private address is there: a start that finds it missing fails after at most two seconds and
# is tried again every ten seconds (Restart=on-failure, no start limit) until the tunnel has brought it.
# It does not order itself After= or Wants= network-online.target, and its wait is short: the packaged
# unit is Before= network-online.target and nss-lookup.target, and the tunnel is ordered after them, so
# a long wait would hold the tunnel back and a drop-in ordering would be a cycle (issue 1126).
[Unit]
After=network-online.target
Wants=network-online.target
StartLimitIntervalSec=0
[Service]
ExecStartPre=/usr/bin/sh -c 'for i in $(seq 1 80); do ip -4 -o addr show to ${machine:address} | grep -q . && exit 0; sleep 1; done; echo "${machine:address} is not on this machine" >&2; exit 1'
ExecStartPre=/usr/bin/sh -c 'for i in $(seq 1 2); do ip -4 -o addr show to ${machine:address} | grep -q . && exit 0; sleep 1; done; echo "${machine:address} is not on this machine" >&2; exit 1'
Restart=on-failure
RestartSec=10s
@@ -9,7 +9,10 @@
{
"name": "wildcard-resolution",
"scope": "mesh",
"identity": false
"identity": false,
"credential": {
"form": "none"
}
}
],
"requires": [
@@ -7,7 +7,7 @@
"privileged"
],
"invokes": [
"seat:mesh-controller.images"
"seat:mesh-artifact-sweep.images"
],
"claims": [
{
@@ -40,6 +40,7 @@
],
"emits": [
"repo.created",
"repo.writable",
"issue.opened",
"pull.merged",
"pull.updated",
@@ -181,6 +182,12 @@
"trusted": true,
"default": {}
},
"check-gates-merge": {
"type": "boolean",
"description": "Whether gitea_merge_pull_request judges mesh/check before it merges. True (the default): a head merges only on a passing mesh/check judged against the trunk's head (ADR 0447, ADR 0534 §4). False (the operator, 2026-10-25 22:56: \"run the check locally only, but not as a gate, it loses us too much time\"): a head the operator's Approve names in approved_heads merges whatever its mesh/check says (pending, failed, missing or judged on an older trunk); a head that moved, a branch no rule guards and a merge git refuses are still refused. Set together with mesh-delivery-gate's check-gates-merge.",
"trusted": true,
"default": true
},
"main-mergers": {
"type": "list",
"description": "The forge logins that may merge into main beside the tools' account, on every repository of the organisation (ADR 0534 §2): the operator's own, so the operator can always merge by hand at the forge. The tools' account is always on main's merge whitelist; no agent's account belongs here.",
@@ -316,7 +323,8 @@
"takes-config": [
"url",
"keep-main-rules",
"main-mergers"
"main-mergers",
"check-gates-merge"
]
},
{
@@ -391,6 +399,7 @@
"provides": [
{
"name": "npm-package-registry",
"credential": {"form": "pair"},
"scope": "mesh",
"identity": {
"max": 40,
@@ -399,6 +408,7 @@
},
{
"name": "git",
"credential": {"form": "none"},
"scope": "mesh"
},
{
@@ -411,6 +421,7 @@
},
{
"name": "forge-account",
"credential": {"form": "none"},
"scope": "mesh"
}
],
@@ -537,6 +548,7 @@
"MESH_GITEA_ADMIN_PASSWORD_FILE": "${dir:state}/admin.secret",
"MESH_GITEA_ORIGINS_FILE": "${dir:mesh-state}/origins",
"MESH_GITEA_SSH_PORT": "222",
"MESH_GITEA_HOST_KEY_FILE": "${dir:data}/ssh/ssh_host_ed25519_key.pub",
"MESH_FORGE_HOST_FILE": "/etc/gitea/forge-host"
}
}
@@ -554,11 +566,14 @@
"announced"
],
"reads": [
"mesh-controller.assignments",
"mesh-provisioner.plans",
"mesh-provisioner.provisions",
"ssh-client.keys",
"mesh-build-go.keys",
"mesh-build-node.keys",
"mesh-build-image.keys",
"mesh-delivery-checker.owed"
"mesh-delivery-checker.owed",
"mesh-delivery-indexer.modules",
"mesh-delivery-gate.source"
]
}
@@ -7,7 +7,10 @@
"provides": [
{
"name": "go-toolchain",
"reach": "machine"
"reach": "machine",
"credential": {
"form": "none"
}
}
],
"resources": [
@@ -7,6 +7,9 @@
"scope": "mesh",
"identity": {
"in": "an InfluxDB v1 authorization"
},
"credential": {
"form": "pair"
}
}
],
@@ -0,0 +1,4 @@
# Written by the mesh's kernel module (systemd-modules-load reads it at boot).
# tun: buildah's network for a RUN line (pasta) and every other user-space tunnel open /dev/net/tun.
# Loaded at boot, so an upgrade that removes the running kernel's modules cannot keep it out later.
tun
@@ -100,6 +100,25 @@
"id": "linux-lts-headers",
"type": "package",
"package": "linux-lts-headers"
},
{
"id": "modules-load",
"type": "file",
"path": "/etc/modules-load.d/mesh-kernel.conf",
"mode": "0644",
"template": "files/modules-load.conf"
},
{
"id": "modules-keep",
"type": "package",
"package": "kernel-modules-hook"
},
{
"id": "modules-cleanup",
"type": "service",
"unit": "linux-modules-cleanup.service",
"state": "stopped",
"boot": "enabled"
}
],
"build": {
@@ -9,6 +9,7 @@
"provides": [
{
"name": "oidc-client",
"credential": {"form": "pair"},
"scope": "mesh",
"identity": {
"max": 255,
@@ -240,6 +241,7 @@
"bound"
],
"reads": [
"mesh-controller.assignments"
"mesh-provisioner.plans",
"mesh-provisioner.provisions"
]
}
@@ -665,7 +665,14 @@
"path": "${dir:mesh-state}/config.json",
"mode": "0600",
"template": "files/config.json",
"takes-config": ["domain", "proxy-address", "sitename", "website", "url", "container"],
"takes-config": [
"domain",
"proxy-address",
"sitename",
"website",
"url",
"container"
],
"merge": "json"
},
{
@@ -758,6 +765,9 @@
"identity": {
"max": 64,
"in": "a mailbox's local part"
},
"credential": {
"form": "pair"
}
}
],
@@ -1,3 +1,5 @@
{
"capacity": "shanks=3,ace=2,g14=0,novox=0"
"capacity": 0,
"cores-per-step": 6,
"build-hours": ""
}
@@ -10,7 +10,7 @@
"state": [
"toolchains",
"keys",
"steps"
{"name": "steps", "ttl-seconds": 172800}, "pauses"
],
"claims": [
{
@@ -20,14 +20,27 @@
}
],
"reads": [
"mesh-delivery-checker.kept"
"mesh-delivery-checker.kept",
"mesh-delivery-indexer.manifests"
],
"config": {
"capacity": {
"type": "number",
"description": "How many asks of the Go build seat this module runs at once on this node (novox/hq ADR 0525 §5); 0 pulls nothing. It is set per node in the mesh's configuration; a node with no value builds nothing.",
"minimum": 0,
"default": 0
},
"build-hours": {
"type": "string",
"description": "How many asks of the Go build seat this module runs at once on each node, as node=asks pairs separated by commas, such as shanks=3,ace=2 (novox/hq ADR 0525 §5); a node set to 0 pulls nothing, a node not named runs one at a time. The control-node and the laptop build nothing.",
"pattern": "|[a-z0-9_-]+=[0-9]+(,[a-z0-9_-]+=[0-9]+)*",
"default": "shanks=3,ace=2,g14=0,novox=0"
"description": "When this node may take this module's build asks: one window HH:MM-HH:MM in the node's local time (an end before its start crosses midnight, 18:00-06:00). Outside it the node takes no new step and finishes the ones it runs; an empty value builds at any hour. It is set per node in the mesh's configuration.",
"pattern": "|([01][0-9]|2[0-3]):[0-5][0-9]-([01][0-9]|2[0-3]):[0-5][0-9]",
"default": ""
},
"cores-per-step": {
"type": "number",
"description": "How many processors one Go step of this module needs (a full check of a large repository takes what it is given: three at about 4 cores each ran past the 30-minute step bound on 2026-10-26). The capacity above is an upper bound: on each node the module runs at most floor(the cores of its build slice's CPU quota / this) steps at once, the quota read live from its cgroup (issue 1806), so a lowered build CPU share never starves its steps. Above 0.",
"minimum": 0.1,
"default": 6
}
},
"tools": [
@@ -38,6 +51,7 @@
"build_go_release",
"build_go_step",
"build_go_steps",
"build_go_step_end",
"build_go_toolchains",
"build_go_status",
"build_go_health"
@@ -105,7 +119,7 @@
"path": "${dir:state}/settings.json",
"mode": "0600",
"owner": "mesh-build-go",
"takes-config": ["capacity"],
"takes-config": ["capacity", "cores-per-step", "build-hours"],
"merge": "json",
"template": "files/settings.json"
},
@@ -1,3 +1,5 @@
{
"capacity": "shanks=3,ace=2,g14=0,novox=0"
"capacity": 0,
"cores-per-step": 2,
"build-hours": ""
}
@@ -9,7 +9,7 @@
],
"state": [
"keys",
"steps"
{"name": "steps", "ttl-seconds": 172800}, "pauses"
],
"claims": [
{
@@ -20,10 +20,22 @@
],
"config": {
"capacity": {
"type": "number",
"description": "How many asks of the image build seat this module runs at once on this node (novox/hq ADR 0525 §5); 0 pulls nothing. It is set per node in the mesh's configuration; a node with no value builds nothing.",
"minimum": 0,
"default": 0
},
"build-hours": {
"type": "string",
"description": "How many asks of the image build seat this module runs at once on each node, as node=asks pairs separated by commas, such as shanks=3,ace=2 (novox/hq ADR 0525 §5); a node set to 0 pulls nothing, a node not named runs one at a time. The control-node and the laptop build nothing.",
"pattern": "|[a-z0-9_-]+=[0-9]+(,[a-z0-9_-]+=[0-9]+)*",
"default": "shanks=3,ace=2,g14=0,novox=0"
"description": "When this node may take this module's build asks: one window HH:MM-HH:MM in the node's local time (an end before its start crosses midnight, 18:00-06:00). Outside it the node takes no new step and finishes the ones it runs; an empty value builds at any hour. It is set per node in the mesh's configuration.",
"pattern": "|([01][0-9]|2[0-3]):[0-5][0-9]-([01][0-9]|2[0-3]):[0-5][0-9]",
"default": ""
},
"cores-per-step": {
"type": "number",
"description": "How many processors one image step of this module needs (buildah, whose RUN lines are mostly fetches). The capacity above is an upper bound: on each node the module runs at most floor(the cores of its build slice's CPU quota / this) steps at once, the quota read live from its cgroup (issue 1806), so a lowered build CPU share never starves its steps. Above 0.",
"minimum": 0.1,
"default": 2
}
},
"tools": [
@@ -33,6 +45,7 @@
"build_image_release",
"build_image_step",
"build_image_steps",
"build_image_step_end",
"build_image_status",
"build_image_health"
],
@@ -77,7 +90,7 @@
"path": "${dir:state}/settings.json",
"mode": "0600",
"owner": "mesh-build-image",
"takes-config": ["capacity"],
"takes-config": ["capacity", "cores-per-step", "build-hours"],
"merge": "json",
"template": "files/settings.json"
},
@@ -1,3 +1,5 @@
{
"capacity": "shanks=3,ace=2,g14=0,novox=0"
"capacity": 0,
"cores-per-step": 2,
"build-hours": ""
}
@@ -10,7 +10,7 @@
"state": [
"toolchains",
"keys",
"steps"
{"name": "steps", "ttl-seconds": 172800}, "pauses"
],
"claims": [
{
@@ -21,10 +21,22 @@
],
"config": {
"capacity": {
"type": "number",
"description": "How many asks of the Node build seat this module runs at once on this node (novox/hq ADR 0525 §5); 0 pulls nothing. It is set per node in the mesh's configuration; a node with no value builds nothing.",
"minimum": 0,
"default": 0
},
"build-hours": {
"type": "string",
"description": "How many asks of the Node build seat this module runs at once on each node, as node=asks pairs separated by commas, such as shanks=3,ace=2 (novox/hq ADR 0525 §5); a node set to 0 pulls nothing, a node not named runs one at a time. The control-node and the laptop build nothing.",
"pattern": "|[a-z0-9_-]+=[0-9]+(,[a-z0-9_-]+=[0-9]+)*",
"default": "shanks=3,ace=2,g14=0,novox=0"
"description": "When this node may take this module's build asks: one window HH:MM-HH:MM in the node's local time (an end before its start crosses midnight, 18:00-06:00). Outside it the node takes no new step and finishes the ones it runs; an empty value builds at any hour. It is set per node in the mesh's configuration.",
"pattern": "|([01][0-9]|2[0-3]):[0-5][0-9]-([01][0-9]|2[0-3]):[0-5][0-9]",
"default": ""
},
"cores-per-step": {
"type": "number",
"description": "How many processors one Node step of this module needs (tsc, esbuild and a test runner; a median step takes 15 s). The capacity above is an upper bound: on each node the module runs at most floor(the cores of its build slice's CPU quota / this) steps at once, the quota read live from its cgroup (issue 1806), so a lowered build CPU share never starves its steps. Above 0.",
"minimum": 0.1,
"default": 2
}
},
"tools": [
@@ -34,6 +46,7 @@
"build_node_release",
"build_node_step",
"build_node_steps",
"build_node_step_end",
"build_node_toolchains",
"build_node_status",
"build_node_health"
@@ -99,7 +112,7 @@
"path": "${dir:state}/settings.json",
"mode": "0600",
"owner": "mesh-build-node",
"takes-config": ["capacity"],
"takes-config": ["capacity", "cores-per-step", "build-hours"],
"merge": "json",
"template": "files/settings.json"
},
@@ -25,9 +25,7 @@
"secrets": {
"postgres-database": "${dir:state}/database.secret"
},
"consumes": [
"mesh-controller.built-before"
],
"consumes": [],
"emits": [
"registered",
"upgraded",
@@ -92,5 +90,12 @@
"from": "files"
}
]
}
},
"invokes": [
"mesh-delivery-builder.builder_builds_since"
],
"reads": [
"mesh-delivery-builder.builds",
"mesh-delivery-indexer.manifests"
]
}
@@ -0,0 +1,87 @@
{
"module": "mesh-conditions",
"version": "1",
"runs-as": "mesh-conditions",
"claims": [
{
"name": "mesh-conditions",
"scope": "mesh",
"serves": [
"conditions",
"silence",
"compare",
"take-over",
"hand-back",
"resync"
]
}
],
"uses": [
"operator-channel"
],
"reads": [
"mesh-delivery-watchman.findings",
"mesh-self-check.findings",
"mesh-healers.findings",
"mesh-delivery-assigner.findings",
"messenger.findings",
"mesh-provisioner.findings",
"artifact-store-tools.findings",
"mesh-healers.attempts",
"mesh-controller.conditions",
"mesh-controller.lease",
"mesh-delivery-assigner.wants",
"node-engine.have",
"mesh-delivery-assigner.placement"
],
"state": [
"conditions",
{
"name": "history",
"ttl-seconds": 7776000
},
"takeover",
"held",
{
"name": "bounds",
"key-ttl": true
},
"asked"
],
"own-secrets": {
"broker": "${dir:state}/broker"
},
"secrets-owner": "mesh-conditions",
"resources": [
{
"id": "account",
"type": "user",
"name": "mesh-conditions",
"shell": "/usr/bin/nologin",
"home": "/var/lib/mesh-conditions"
},
{
"id": "state",
"type": "directory",
"mode": "0700",
"owner": "mesh-conditions",
"place": ".",
"unknown": "program"
}
],
"build": {
"artifacts": [
{
"name": "tools",
"kind": "bundle",
"language": "go",
"system": "arch",
"from": "cmd/mesh-conditions",
"binary": "mesh-conditions",
"loads": [
"mesh-conditions"
]
}
]
}
}
@@ -7,7 +7,10 @@
"name": "mesh-config",
"scope": "mesh",
"serves": [
"config"
"config",
"compare",
"take-over",
"hand-back"
]
}
],
@@ -23,14 +26,30 @@
"gitea.forge.pull-request.closed"
],
"invokes": [
"seat:mesh-hand-acts.hand-act",
"seat:mesh-hand-acts.warranted",
"seat:mesh-controller.hand-act",
"seat:mesh-controller.warranted",
"gitea.gitea_get_pull_request"
],
"reads": [
"mesh-controller.assignments",
"mesh-delivery-assigner.placement",
"mesh-delivery-assigner.wants",
"mesh-delivery-builder.builds",
"mesh-delivery-indexer.modules",
"mesh-delivery-indexer.manifests",
"mesh-controller.predefined",
"mesh-controller.declared-config",
"mesh-controller.predefined"
"mesh-delivery-indexer.predefined",
"mesh-delivery-indexer.config-candidates",
"mesh-controller.lease"
],
"tools": [
"config_compare",
"config_take_over",
"config_hand_back",
"config_declared",
"config_resync"
],
"state": [
"values",
@@ -38,7 +57,9 @@
"predefined",
"proposals",
"pulls",
"kept"
"kept",
"declared",
"takeover"
],
"logs": [
"history"
@@ -1,4 +1,5 @@
{
"refusal-stall-floor-minutes": 10,
"author-new-head-minutes": 120
"author-new-head-minutes": 120,
"check-gates-merge": true
}
@@ -14,7 +14,8 @@
"table",
"stalled",
"recheck",
"close"
"close",
"settle"
]
}
],
@@ -26,13 +27,17 @@
"gitea.forge.pull-request.edited",
"gitea.forge.pull-request.closed",
"mesh-delivery-forge-bridge.forge.**",
"mesh-config.config-changed"
"mesh-config.config-changed",
"gitea.repo.writable"
],
"emits": [
"transition",
"group",
"merged"
],
"uses": [
"operator-channel"
],
"invokes": [
"gitea.gitea_commit_statuses",
"gitea.gitea_contains",
@@ -57,14 +62,23 @@
{
"name": "source",
"history": 64
},
"conditions",
{
"name": "turns",
"history": 5
}
],
"reads": [
"mesh-delivery-teller.answers",
"mesh-delivery-checker.verdicts"
"mesh-delivery-checker.verdicts",
"mesh-delivery-view.reach"
],
"tools": [
"delivery_status"
"delivery_hand_now",
"delivery_merge_again",
"delivery_status",
"delivery_turns"
],
"config": {
"refusal-stall-floor-minutes": {
@@ -82,6 +96,12 @@
"maximum": 10080,
"trusted": false,
"default": 120
},
"check-gates-merge": {
"type": "boolean",
"description": "Whether mesh/check gates the merge. True (the default): the operator's Approve is asked, and a merge handed, only on a passing verdict judged against the trunk's head, and a trunk's move puts a verdict out of date (ADR 0447, ADR 0505 §4). False (the operator, 2026-10-25 22:56: \"run the check locally only, but not as a gate, it loses us too much time\"; developers run the unit and integration levels locally before a pull request): a pull request reviewed at its head is ready, and its Approve is asked once the check at its heads (a group: of all its heads together) has a verdict, a pass or a failure said in the ask, or after 60 minutes with none, said so (the operator, 2026-10-27 23:54: \"if you don't want me to approve it, don't send the approval request\"); an approved merge waits for no verdict; no merge is refused for a check judged on an older trunk; a later verdict is shown on the pull request and the delivery, and asks a pending Approve again in its words; a group still merges all its members in one pass. Set it false together with the gitea module's check-gates-merge, which lets its merge tool merge an approved head without a passing mesh/check.",
"trusted": false,
"default": true
}
},
"resources": [
@@ -99,7 +119,8 @@
"mode": "0600",
"takes-config": [
"refusal-stall-floor-minutes",
"author-new-head-minutes"
"author-new-head-minutes",
"check-gates-merge"
],
"merge": "json",
"template": "files/settings.json"
@@ -0,0 +1,65 @@
{
"module": "mesh-hand-acts",
"version": "1",
"runs-as": "mesh-hand-acts",
"claims": [
{
"name": "mesh-hand-acts",
"scope": "mesh",
"serves": [
"hand-act",
"hand-acts",
"warranted"
]
}
],
"reads": [
"mesh-controller.hand-acts",
"messenger.asks"
],
"logs": [
{
"name": "acts",
"max-age-days": 400
}
],
"state": [
"warrants"
],
"own-secrets": {
"broker": "${dir:state}/broker"
},
"secrets-owner": "mesh-hand-acts",
"resources": [
{
"id": "account",
"type": "user",
"name": "mesh-hand-acts",
"shell": "/usr/bin/nologin",
"home": "/var/lib/mesh-hand-acts"
},
{
"id": "state",
"type": "directory",
"mode": "0700",
"owner": "mesh-hand-acts",
"place": ".",
"unknown": "program"
}
],
"build": {
"artifacts": [
{
"name": "tools",
"kind": "bundle",
"language": "go",
"system": "arch",
"from": "cmd/mesh-hand-acts",
"binary": "mesh-hand-acts",
"loads": [
"mesh-hand-acts"
]
}
]
}
}
@@ -0,0 +1,82 @@
{
"module": "mesh-healers",
"version": "1",
"runs-as": "mesh-healers",
"claims": [
{
"name": "mesh-healers",
"scope": "mesh",
"serves": [
"healers",
"heal",
"drill"
]
}
],
"consumes": [
"mesh-conditions.condition-raised",
"mesh-conditions.condition-changed",
"mesh-hand-acts.act-recorded"
],
"invokes": [
"mesh-grants.grants_assert",
"seat:mesh-hand-acts.hand-act",
"seat:mesh-controller.drill"
],
"reads": [
"mesh-controller.lease",
"mesh-conditions.conditions",
"mesh-controller.hand-acts"
],
"state": [
{
"name": "attempts",
"key-ttl": true
},
{
"name": "would",
"key-ttl": true
},
"findings",
{
"name": "wakes",
"key-ttl": true
}
],
"own-secrets": {
"broker": "${dir:state}/broker"
},
"secrets-owner": "mesh-healers",
"resources": [
{
"id": "account",
"type": "user",
"name": "mesh-healers",
"shell": "/usr/bin/nologin",
"home": "/var/lib/mesh-healers"
},
{
"id": "state",
"type": "directory",
"mode": "0700",
"owner": "mesh-healers",
"place": ".",
"unknown": "program"
}
],
"build": {
"artifacts": [
{
"name": "tools",
"kind": "bundle",
"language": "go",
"system": "arch",
"from": "cmd/mesh-healers",
"binary": "mesh-healers",
"loads": [
"mesh-healers"
]
}
]
}
}
@@ -13,12 +13,17 @@
"move",
"note",
"link",
"milestones",
"milestone-set",
"milestone-move",
"import",
"groom",
"set",
"tracking",
"reserve",
"fix-times"
"fix-times",
"learn-fixes",
"changes"
]
}
],
@@ -33,12 +38,17 @@
"move",
"note",
"link",
"milestones",
"milestone-set",
"milestone-move",
"import",
"groom",
"set",
"tracking",
"reserve",
"fix-times"
"fix-times",
"learn-fixes",
"changes"
]
}
],
@@ -52,7 +62,9 @@
"triaged",
"ready",
"ready-to-verify",
"reserved"
"reserved",
"milestone-set",
"milestone-moved"
],
"consumes": [
"mesh-issues.opened",
@@ -66,6 +78,7 @@
"seat:operator-channel.notify",
"gitea.gitea_pull_reviews",
"gitea.gitea_contains",
"gitea.gitea_get_pull_request",
"records.records_status",
"records.records_list",
"records.records_read",
@@ -81,7 +94,11 @@
},
"counters",
"grooming",
"reservations"
"reservations",
{
"name": "milestones",
"history": 5
}
],
"config": {
"first-number": {
@@ -32,7 +32,8 @@
"propose",
"import",
"return",
"withdraw"
"withdraw",
"progress"
]
}
],
@@ -52,13 +53,15 @@
"propose",
"import",
"return",
"withdraw"
"withdraw",
"progress"
]
}
],
"tools": [
"knowledge_health",
"knowledge_highest"
"knowledge_highest",
"knowledge_marks"
],
"emits": [
"reserved",
@@ -81,7 +84,14 @@
"records.records_status",
"records.records_list",
"records.records_read",
"seat:operator-channel.notify"
"seat:operator-channel.notify",
"seat:issue-tracker.list",
"seat:issue-tracker.get",
"seat:issue-tracker.milestones",
"seat:issue-tracker.changes",
"seat:mesh-delivery.deliveries",
"seat:mesh-delivery.show",
"seat:mesh-delivery.groups"
],
"config": {
"judge-model": {
@@ -11,11 +11,36 @@
]
}
],
"uses": [
"operator-channel"
],
"reads": [
"mesh-delivery-assigner.placement",
"mesh-delivery-assigner.wants",
"mesh-delivery-indexer.modules",
"mesh-delivery-indexer.manifests",
"mesh-delivery-builder.builds",
"node-engine.have",
"mesh-config.resolved",
"gitea.bound",
"keycloak.bound",
"postgres.bound",
"route-adapter.bound",
"mesh-vault.bound"
],
"tools": [
"provisioner_provisions",
"provisioner_plans",
"provisioner_compose",
"provisioner_decide"
],
"state": [
"registry",
"needs",
"applied",
"collect"
"collect",
"provisions",
"plans"
],
"own-secrets": {
"broker": "${dir:state}/broker"
@@ -0,0 +1,87 @@
{
"module": "mesh-self-check",
"version": "1",
"runs-as": "mesh-self-check",
"claims": [
{
"name": "mesh-self-check",
"scope": "mesh",
"serves": [
"doctor",
"probes"
]
}
],
"invokes": [
"nats.nats_streams",
"nats.nats_closed_connections",
"seat:mesh-watchdogs.signals",
"seat:node-intrusion-prevention.banned"
],
"reads": [
"mesh-delivery-assigner.placement",
"mesh-delivery-assigner.seat-holders",
"mesh-delivery-watchman.presence",
"mesh-host.health",
"disk-load.readings",
"restic.good",
"mesh-delivery-indexer.modules",
"mesh-delivery-indexer.manifests"
],
"state": [
"findings",
"verdict",
{
"name": "verdicts",
"ttl-seconds": 604800
}
],
"own-secrets": {
"broker": "${dir:state}/broker"
},
"secrets-owner": "mesh-self-check",
"resources": [
{
"id": "account",
"type": "user",
"name": "mesh-self-check",
"shell": "/usr/bin/nologin",
"home": "/var/lib/mesh-self-check"
},
{
"id": "state",
"type": "directory",
"mode": "0700",
"owner": "mesh-self-check",
"place": ".",
"unknown": "program"
}
],
"build": {
"artifacts": [
{
"name": "tools",
"kind": "bundle",
"language": "go",
"system": "arch",
"from": "cmd/mesh-self-check",
"binary": "mesh-self-check",
"loads": [
"mesh-self-check"
]
}
]
},
"config": {
"probes": {
"type": "list",
"description": "The ids of the probes that run, each judging one thing of the mesh (mesh-self-check.probes lists them with what each asserts). A probe not listed runs nothing and raises nothing, and what it raised before is cleared. D8 and DA are in the code and off by default.",
"trusted": true,
"default": [
"D13",
"D-fill"
],
"runtime": true
}
}
}
@@ -5,6 +5,9 @@
"provides": [
{
"name": "secret",
"credential": {
"form": "pair"
},
"scope": "mesh",
"identity": {
"in": "a vault entry"
@@ -14,10 +17,27 @@
"emits": [
"provisioned",
"rotated",
"secret-changed"
"secret-changed",
"account-hashed",
"own-replaced",
"account-issued",
"signed"
],
"reads": [
"mesh-controller.assignments"
"node-engine.have",
"mesh-grants.accounts",
"mesh-controller.lease",
"mesh-delivery-assigner.placement",
"mesh-provisioner.plans",
"mesh-provisioner.provisions"
],
"state": [
{
"name": "bus-accounts",
"history": 2
},
"signers",
"bound"
],
"receives": {
"secret": "${dir:grants}/mesh.json"
@@ -135,7 +155,8 @@
"MESH_VAULT_LEDGER": "${dir:ledger}",
"MESH_VAULT_ROOT": "${dir:root}",
"MESH_VAULT_STORE": "${dir:store}",
"MESH_VAULT_STATE": "${dir:state}"
"MESH_VAULT_STATE": "${dir:state}",
"MESH_VAULT_SECRET_ASK_CALLERS": "node-tools"
}
}
]
@@ -145,6 +166,10 @@
"secret_holders",
"secret_verify",
"secret_handed",
"secret_own",
"account_hash",
"account_adopt",
"account_retire",
"secret_asks",
"secret_vault_export",
"provisioner_retirement",
@@ -152,7 +177,12 @@
"provisioner_retire_reject",
"provisioner_delete",
"vault_status",
"vault_health"
"vault_health",
"own_drift",
"own_take_switch",
"vault_signers",
"vault_signed",
"signers_reassert"
],
"claims": [
{
@@ -166,7 +196,10 @@
"rotate",
"move-in",
"node-key",
"move-asks"
"move-asks",
"own",
"issue-account",
"sign"
]
}
]
@@ -3,11 +3,15 @@
"version": "1",
"slug": "watch",
"consumes": [
"mesh-controller.doctor-heartbeat"
"mesh-controller.doctor-heartbeat",
"mesh-self-check.checked"
],
"reads": [
"mesh-conditions.takeover",
"mesh-delivery-assigner.seat-holders"
],
"invokes": [
"mesh-watcher.watcher_ping",
"seat:mesh-controller.seats"
"mesh-watcher.watcher_ping"
],
"own-secrets": {
"telegram-token": "${dir:state}/telegram-token"
@@ -39,7 +43,9 @@
"type": "file",
"path": "${dir:state}/settings.json",
"mode": "0600",
"takes-config": ["telegram-chat-id"],
"takes-config": [
"telegram-chat-id"
],
"merge": "json",
"template": "files/settings.json"
}
@@ -77,6 +77,9 @@
"mesh-controller.condition-raised",
"mesh-controller.condition-changed",
"mesh-controller.condition-cleared",
"mesh-conditions.condition-raised",
"mesh-conditions.condition-changed",
"mesh-conditions.condition-cleared",
"mesh-delivery-view.moving",
"mesh-delivery-view.settled",
"mesh-delivery-view.stuck",
@@ -90,9 +93,13 @@
],
"invokes": [
"seat:mesh-controller.conditions",
"seat:mesh-controller.root-free",
"seat:mesh-conditions.conditions",
"seat:node-engine.root-free",
"seat:issue-tracker.tracking"
],
"reads": [
"mesh-conditions.takeover"
],
"state": [
"open",
"sent",
@@ -12,6 +12,9 @@
"identity": {
"max": 20,
"in": "an S3 access key"
},
"credential": {
"form": "pair"
}
}
],
@@ -36,8 +39,8 @@
]
}
],
"invokes": [
"seat:mesh-controller.seats"
"reads": [
"mesh-delivery-assigner.seat-holders"
],
"contributes": {
"route": {
@@ -12,6 +12,9 @@
"identity": {
"max": 63,
"in": "a MongoDB database name"
},
"credential": {
"form": "pair"
}
}
],
@@ -8,6 +8,9 @@
"scope": "mesh",
"identity": {
"in": "a Mosquitto client and topic prefix"
},
"credential": {
"form": "pair"
}
}
],
@@ -5,11 +5,9 @@
"nfs-share"
],
"reads": [
"disk-load.readings",
"nfs-server.exports"
],
"invokes": [
"seat:mesh-controller.data"
],
"capabilities": [
"package-manager",
"service-manager"
@@ -38,20 +36,20 @@
"per-machine": true
}
],
"config": {
"shares": {
"type": "string",
"description": "The shares of another machine this machine mounts: space-separated name=mountpoint, each optionally :ro, or none (novox/hq ADR 0263 rule 5).",
"trusted": true,
"default": "none"
},
"sources": {
"type": "string",
"description": "The occasional sources this machine mounts: space-separated name=smb://[<user>@]<host>/<share>@<mountpoint> or name=disk:<uuid>@<mountpoint>, each optionally :ro, or none (novox/hq ADR 0263 rule 6); an SMB source's username is in its entry, its password the secret smb-password-<name> (novox/hq ADR 0283).",
"trusted": true,
"default": "none"
}
"config": {
"shares": {
"type": "string",
"description": "The shares of another machine this machine mounts: space-separated name=mountpoint, each optionally :ro, or none (novox/hq ADR 0263 rule 5).",
"trusted": true,
"default": "none"
},
"sources": {
"type": "string",
"description": "The occasional sources this machine mounts: space-separated name=smb://[<user>@]<host>/<share>@<mountpoint> or name=disk:<uuid>@<mountpoint>, each optionally :ro, or none (novox/hq ADR 0263 rule 6); an SMB source's username is in its entry, its password the secret smb-password-<name> (novox/hq ADR 0283).",
"trusted": true,
"default": "none"
}
},
"own-secrets": {
"smb-password-*": {
"path": "${dir:state}/smb-password-*.secret",
@@ -12,6 +12,9 @@
"identity": {
"max": 128,
"in": "a SQL Server login and database name"
},
"credential": {
"form": "pair"
}
}
],
@@ -8,7 +8,10 @@
"provides": [
{
"name": "mesh-bus",
"scope": "mesh"
"scope": "mesh",
"credential": {
"form": "none"
}
}
],
"claims": [
@@ -19,12 +22,16 @@
],
"bus-users": "/var/lib/nats-module/conf/accounts.conf",
"own-secrets": {
"broker": "${dir:mesh-state}/broker"
"broker": "${dir:mesh-state}/broker",
"broker-nats-tools": "${dir:tools-secret}/broker-nats-tools"
},
"capabilities": [
"container-runtime"
],
"emits": [],
"emits": [
"consumer-removed",
"bucket-keys-removed"
],
"consumes": [],
"listens": [
{
@@ -45,7 +52,9 @@
"nats_buckets",
"nats_users",
"nats_user_can",
"nats_snapshot_verify"
"nats_snapshot_verify",
"nats_consumer_remove",
"nats_bucket_remove"
],
"guards": [
8222
@@ -86,6 +95,13 @@
"mode": "0700",
"unknown": "program"
},
{
"id": "tools-secret",
"type": "directory",
"path": "/var/lib/mesh/nats-tools",
"mode": "0711",
"unknown": "program"
},
{
"id": "snapshots",
"type": "directory",
@@ -190,11 +206,18 @@
"loads": [
"nats-tools"
],
"bus": {
"removes": [
"DEAD_LETTER_NOTICES.controller",
"mesh-host_acts"
]
},
"env": {
"MESH_NATS_MONITOR": "http://127.0.0.1:8222",
"MESH_NATS_CONTAINER": "mesh-broker-nats",
"MESH_NATS_USERS_FILE": "/etc/nats/accounts.conf",
"MESH_NATS_SNAPSHOTS_DIR": "${dir:snapshots}"
"MESH_NATS_SNAPSHOTS_DIR": "${dir:snapshots}",
"NATS_TOOLS_CREDENTIAL": "${dir:tools-secret}/broker-nats-tools"
}
},
{
@@ -72,7 +72,7 @@
},
"user": {
"type": "string",
"description": "The Nextcloud account the module's tools act as; empty is admin.",
"description": "The Nextcloud account the module's tools act as; empty is mesh-admin, the account the module makes and keeps for itself. Any other account is used as given and never repaired.",
"trusted": true,
"default": ""
}
@@ -13,7 +13,10 @@
{
"name": "nfs-share",
"scope": "mesh",
"identity": false
"identity": false,
"credential": {
"form": "none"
}
}
],
"data": {
@@ -45,26 +48,24 @@
}
],
"reads": [
"disk-load.readings",
"mesh-delivery-assigner.seat-holders",
"mounts.shares"
],
"invokes": [
"seat:mesh-controller.seats",
"seat:mesh-controller.data"
],
"config": {
"shares": {
"type": "string",
"description": "The folders this machine exports, as space-separated name=path; the operator's, with no default (novox/hq ADR 0263).",
"trusted": true,
"required": true
},
"grants": {
"type": "string",
"description": "Which nodes may mount which share, and how: share=node:rw|ro[,node:rw|ro], entries separated by spaces, or none to export to no node whatever a node wants (novox/hq ADR 0263).",
"trusted": true,
"default": "none"
}
"config": {
"shares": {
"type": "string",
"description": "The folders this machine exports, as space-separated name=path; the operator's, with no default (novox/hq ADR 0263).",
"trusted": true,
"required": true
},
"grants": {
"type": "string",
"description": "Which nodes may mount which share, and how: share=node:rw|ro[,node:rw|ro], entries separated by spaces, or none to export to no node whatever a node wants (novox/hq ADR 0263).",
"trusted": true,
"default": "none"
}
},
"facts": {
"machines": {
"path": "/etc/nfs-server/machines",
@@ -7,7 +7,10 @@
"provides": [
{
"name": "node-toolchain",
"reach": "machine"
"reach": "machine",
"credential": {
"form": "none"
}
}
],
"resources": [
@@ -7,7 +7,10 @@
"provides": [
{
"name": "model-access",
"scope": "node"
"scope": "node",
"credential": {
"form": "none"
}
}
],
"listens": [
@@ -9,6 +9,7 @@
"provides": [
{
"name": "postgres-database",
"credential": {"form": "pair"},
"scope": "mesh",
"identity": {
"max": 63,
@@ -195,6 +196,7 @@
"bound"
],
"reads": [
"mesh-controller.assignments"
"mesh-provisioner.plans",
"mesh-provisioner.provisions"
]
}
@@ -314,6 +314,48 @@
"mode": "0644",
"template": "templates/systemd/system/suspend-then-hibernate.target.d/mesh-power.conf"
},
{
"id": "reboot-drop-ins",
"type": "directory",
"path": "/etc/systemd/system/reboot.target.d",
"mode": "0755",
"unknown": "program"
},
{
"id": "reboot-bound",
"type": "file",
"path": "/etc/systemd/system/reboot.target.d/mesh-power.conf",
"mode": "0644",
"template": "templates/systemd/system/reboot.target.d/mesh-power.conf"
},
{
"id": "poweroff-drop-ins",
"type": "directory",
"path": "/etc/systemd/system/poweroff.target.d",
"mode": "0755",
"unknown": "program"
},
{
"id": "poweroff-bound",
"type": "file",
"path": "/etc/systemd/system/poweroff.target.d/mesh-power.conf",
"mode": "0644",
"template": "templates/systemd/system/poweroff.target.d/mesh-power.conf"
},
{
"id": "halt-drop-ins",
"type": "directory",
"path": "/etc/systemd/system/halt.target.d",
"mode": "0755",
"unknown": "program"
},
{
"id": "halt-bound",
"type": "file",
"path": "/etc/systemd/system/halt.target.d/mesh-power.conf",
"mode": "0644",
"template": "templates/systemd/system/halt.target.d/mesh-power.conf"
},
{
"id": "udev-rule",
"type": "file",
@@ -0,0 +1,8 @@
# Written by the mesh (module power, novox/hq ADR 0211; issue 1126). Replaced on every push.
#
# A shutdown is bounded at ten minutes, not systemd's thirty. A unit started again during the shutdown
# (on ace, 2026-10-28, the bus and the authorization manager) is never stopped a second time, and the
# shutdown then waits on this job's timeout with nothing on the console. The target keeps its own
# forcing action: the last phase still ends every process, unmounts and syncs.
[Unit]
JobTimeoutSec=10min
@@ -0,0 +1,8 @@
# Written by the mesh (module power, novox/hq ADR 0211; issue 1126). Replaced on every push.
#
# A shutdown is bounded at ten minutes, not systemd's thirty. A unit started again during the shutdown
# (on ace, 2026-10-28, the bus and the authorization manager) is never stopped a second time, and the
# shutdown then waits on this job's timeout with nothing on the console. The target keeps its own
# forcing action: the last phase still ends every process, unmounts and syncs.
[Unit]
JobTimeoutSec=10min
@@ -0,0 +1,8 @@
# Written by the mesh (module power, novox/hq ADR 0211; issue 1126). Replaced on every push.
#
# A shutdown is bounded at ten minutes, not systemd's thirty. A unit started again during the shutdown
# (on ace, 2026-10-28, the bus and the authorization manager) is never stopped a second time, and the
# shutdown then waits on this job's timeout with nothing on the console. The target keeps its own
# forcing action: the last phase still ends every process, unmounts and syncs.
[Unit]
JobTimeoutSec=10min
@@ -7,6 +7,9 @@
"scope": "mesh",
"identity": {
"in": "a Redis ACL user and key prefix"
},
"credential": {
"form": "pair"
}
}
],
@@ -41,40 +41,40 @@
}
]
},
"config": {
"keep-daily": {
"type": "number",
"description": "How many nightly restore points are kept, a count from 1 to 9999: two weeks undo a mistake noticed within a fortnight. Read by the night, which thins nothing while it does not read (novox/hq issue 416); lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 14
},
"keep-weekly": {
"type": "number",
"description": "How many weekly restore points are kept behind the dailies, a count from 1 to 9999, refused by the night like keep-daily; lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 8
},
"keep-monthly": {
"type": "number",
"description": "How many monthly restore points are kept behind the weeklies, a count from 1 to 9999, refused by the night like keep-daily; lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 6
},
"keep-hand-taken-days": {
"type": "number",
"description": "How many days a restore point taken by hand (node-backup.now, or the mesh's before a module moves) is kept at least, a count from 1 to 9999; it is kept until a later good night of its module exists, and this many days in any case, so the next thinning never takes the one taken against a risky change (novox/hq issue 453). Lowering it lets go of the older ones taken by hand at the next thinning, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 7
}
"config": {
"keep-daily": {
"type": "number",
"description": "How many nightly restore points are kept, a count from 1 to 9999: two weeks undo a mistake noticed within a fortnight. Read by the night, which thins nothing while it does not read (novox/hq issue 416); lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 14
},
"keep-weekly": {
"type": "number",
"description": "How many weekly restore points are kept behind the dailies, a count from 1 to 9999, refused by the night like keep-daily; lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 8
},
"keep-monthly": {
"type": "number",
"description": "How many monthly restore points are kept behind the weeklies, a count from 1 to 9999, refused by the night like keep-daily; lowering it deletes the restore points beyond the new count at the next night, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 6
},
"keep-hand-taken-days": {
"type": "number",
"description": "How many days a restore point taken by hand (node-backup.now, or the mesh's before a module moves) is kept at least, a count from 1 to 9999; it is kept until a later good night of its module exists, and this many days in any case, so the next thinning never takes the one taken against a risky change (novox/hq issue 453). Lowering it lets go of the older ones taken by hand at the next thinning, and that cannot be undone.",
"minimum": 1,
"maximum": 9999,
"trusted": true,
"default": 7
}
},
"resources": [
{
"id": "state",
@@ -139,22 +139,17 @@
"MESH_BACKUP_DATA": "${dir:state}/data.conf"
},
"health": {
"kind": "tool",
"tool": "backup_health",
"interval": "60s",
"timeout": "20s",
"looks": 2,
"grace": "60s"
"kind": "unit"
}
},
{
"id": "measure",
"id": "awake",
"type": "process",
"name": "restic-measure",
"name": "restic-awake",
"artifact": "tools",
"run": [
"./restic-backups",
"measure"
"awake"
],
"user": "${machine:account}",
"env": {
@@ -165,7 +160,12 @@
"MESH_BACKUP_DATA": "${dir:state}/data.conf"
},
"health": {
"kind": "unit"
"kind": "tool",
"tool": "backup_health",
"interval": "60s",
"timeout": "20s",
"looks": 2,
"grace": "60s"
}
}
],
@@ -195,5 +195,11 @@
"from": "files"
}
]
}
},
"reads": [
"disk-load.readings"
],
"state": [
"good"
]
}
@@ -9,6 +9,7 @@
"provides": [
{
"name": "route",
"credential": {"form": "none"},
"scope": "mesh",
"identity": false
}
@@ -129,6 +130,7 @@
"bound"
],
"reads": [
"mesh-controller.assignments"
"mesh-provisioner.plans",
"mesh-provisioner.provisions"
]
}
@@ -9,7 +9,10 @@
"provides": [
{
"name": "greeting",
"scope": "mesh"
"scope": "mesh",
"credential": {
"form": "none"
}
}
],
"serves": {
@@ -7,7 +7,10 @@
"provides": [
{
"name": "internal-acme-ca",
"scope": "mesh"
"scope": "mesh",
"credential": {
"form": "none"
}
}
],
"serves": {
@@ -8,7 +8,10 @@
"provides": [
{
"name": "split-dns",
"reach": "machine"
"reach": "machine",
"credential": {
"form": "none"
}
}
],
"requires": [
@@ -32,7 +32,10 @@
"provides": [
{
"name": "analytics",
"scope": "mesh"
"scope": "mesh",
"credential": {
"form": "none"
}
}
],
"serves": {
@@ -9,7 +9,10 @@
"provides": [
{
"name": "x11-display",
"reach": "machine"
"reach": "machine",
"credential": {
"form": "none"
}
}
],
"claims": [