The controller no longer subscribes to the alive message: HostBeats, its handler, the heartbeat-gap duration and a fixed silent bound (issue 1183, AC3, AC4, AC5)

internal/link/bus.go

@@ -67,17 +67,12 @@ const (
 	// BuiltSubject is where a build's outcome lands, for results nobody was waiting for.
 	BuiltSubject = "mesh.control.built"
 
-	// AliveSubjects is every node's heartbeat. Core NATS, never a stream: a lost heartbeat is the
-	// next heartbeat, and a stream of them is the mesh's least valuable message competing for
-	// retention with its most valuable (design 25 §3).
-	AliveSubjects = "mesh.control.*.alive"
-
 	// ToolsAliveSubjects is every machine's node tools saying they are there (novox/hq to-be 45 §3,
 	// S11): core NATS like the host's, for the same reason.
 	ToolsAliveSubjects = "mesh.control.*.tools-alive"
 
 	// HealthSubjects is every machine's health statement between its reports (novox/hq ADR 0240): core
-	// NATS like the heartbeat, because a statement lost is said again within a minute while anything is
+	// NATS like the node tools' heartbeat, because a statement lost is said again within a minute while anything is
 	// not healthy, and the next report carries it whatever happens.
 	HealthSubjects = "mesh.control.*.health"
 )
@@ -89,9 +84,6 @@ func HealthSubject(node string) string { return "mesh.control." + node + ".healt
 // message the store-window guarantee is about (ADR 0083).
 func ReportSubject(node string) string { return "mesh.control." + node + ".report" }
 
-// AliveSubject is one node's heartbeat.
-func AliveSubject(node string) string { return "mesh.control." + node + ".alive" }
-
 // ToolsAliveSubject is one machine's node tools' heartbeat.
 func ToolsAliveSubject(node string) string { return "mesh.control." + node + ".tools-alive" }
 

internal/link/contracts.go

@@ -32,8 +32,6 @@ var Contracts = map[string]Contract{
 	KindSourceMoved: {Ordered: "by the merge's commit against what was built from it: a merge already acted on " +
 		"or older than the last look is history, not a second plan (issues 250, 266)",
 		Tests: []string{"TestAMergeOlderThanTheLastLookIsHistory", "TestAMergeMatchesTheSourcesBuiltFromIt"}},
-	KindHeartbeat: {Unordered: "a word that the machine is there: the newest heard is the newest said, and one " +
-		"lost is the next one"},
 	KindToolsHeartbeat: {Unordered: "a word that the node tools are there, as a machine's heartbeat"},
 	KindHealth: {Ordered: "by the time the node-engine looked (`at`, on the machine's clock): a statement older " +
 		"than the one kept for that machine is refused, so an event arriving after a newer report does not undo it " +

internal/link/contracts_test.go

@@ -15,7 +15,7 @@ import (
 // exists (novox/hq to-be 45 Phase 2, ADR 0227 rule 2).
 func TestEveryConsumedKindHasAContract(t *testing.T) {
 	// What the controller can be handed, from the subjects it is granted and the ones it derives.
-	subjects := []string{EnrolSubject, BuiltSubject, ReportSubject("anchor"), AliveSubject("anchor"),
+	subjects := []string{EnrolSubject, BuiltSubject, ReportSubject("anchor"),
 		ToolsAliveSubject("anchor"), HealthSubject("anchor"), "mesh.mod.postgres.event.provisioner.failing"}
 	subjects = append(subjects, broker.ControllerFollows...)
 	subjects = append(subjects, broker.ControllerHears...)
@@ -83,3 +83,17 @@ func TestEveryConsumedKindHasAContract(t *testing.T) {
 		}
 	}
 }
+
+// 1183/AC3: the controller no longer takes the node-engine's alive message: its subject is no kind, so nothing is
+// subscribed to it and nothing hears it. The tool runner's tools-alive stays.
+func TestTheAliveMessageIsNoKindTheControllerHears(t *testing.T) { // 1183/AC3
+	if kind, known := kindOfSubject("mesh.control.anchor.alive"); known {
+		t.Fatalf("the controller still takes the alive message as %s", kind)
+	}
+	if kind, known := kindOfSubject(ToolsAliveSubject("anchor")); !known || kind != KindToolsHeartbeat {
+		t.Fatalf("the tool runner's tools-alive is %q, %v", kind, known)
+	}
+	if _, ok := Contracts["heartbeat"]; ok {
+		t.Fatal("a contract still names the alive message")
+	}
+}

internal/link/enrolment.go

@@ -384,9 +384,8 @@ func (e Enrolment) Heard(ctx context.Context, report Report) (news bool, err err
 	}
 
 	// A bare word that a node is there is not an account of what the machine did or holds: it
-	// moves last_seen and touches nothing else. This arrives every minute (link.AliveEvery),
-	// while a real report is rare, so recording it as one would overwrite the node's last real
-	// apply with an empty one — wiping the declaration digest that decides whether the node is
+	// moves last_seen and touches nothing else. Recording one as a report would overwrite the
+	// node's last real apply with an empty one — wiping the declaration digest that decides whether the node is
 	// current, the carried ports a push assigns around, and the clean-or-failed outcome — and a
 	// node that had just caught up would read as behind within the minute. The alive path calls
 	// this with only a node name; a real report always carries an account (something applied, or
@@ -394,10 +393,6 @@ func (e Enrolment) Heard(ctx context.Context, report Report) (news bool, err err
 	if report.Applied == nil && report.Refused == "" && len(report.Failed) == 0 {
 		if report.Superseded != "" {
 			log.Printf("%s set aside declaration %s for the newer %s", report.Node, report.Declared, report.Superseded)
-		} else if err := e.Inventory.RecordHeartbeatGap(ctx, report.Node, node.LastSeen); err != nil {
-			// The silence before this word, which the machine-silent bound will be set from (novox/hq
-			// to-be 45 Phase 0). A measurement lost is said and costs the report nothing.
-			log.Printf("the silence before %s's word could not be recorded: %v", report.Node, err)
 		}
 		return false, e.Inventory.Seen(ctx, node.ID)
 	}

internal/link/protocol.go

@@ -27,7 +27,6 @@ const ControlQueue = "control"
 const (
 	KeyEnrol  = "enrol"
 	KeyReport = "report"
-	KeyAlive  = "alive"
 )
 
 // QueueFor is the queue a node consumes from — the only one it may read.
@@ -134,19 +133,6 @@ type Signed struct {
 	Signature   []byte `json:"signature"`
 }
 
-// Alive is a node saying nothing except that it is there.
-//
-// How long a node has been out of touch is a fact only the mesh can hold — nobody else is
-// watching — and without it a node running last month's assignments looks exactly like one that
-// is current.
-type Alive struct {
-	Node string `json:"node"`
-	// IntervalSeconds is how often the node says it is there (novox/hq to-be 45 §3, S1): the
-	// watchdog's bound is three of them. Zero from a host older than that, which is read as the
-	// interval hosts have always used.
-	IntervalSeconds int `json:"interval_seconds,omitempty"`
-}
-
 // ToolsAlive is a machine's node tools saying they are there (novox/hq to-be 45 §3, S11): the
 // runtime every module's tools and every held seat's verbs are served by. Its own word, apart from the
 // host's, because a host heard and a runtime gone is a machine nobody can ask anything.

internal/link/receive.go

@@ -27,7 +27,6 @@ import (
 const (
 	KindEnrolment = "enrolment"
 	KindReport    = "report"
-	KindHeartbeat = "heartbeat"
 	// KindToolsHeartbeat is a machine's node tools saying they are there (novox/hq to-be 45 S11).
 	KindToolsHeartbeat = "tools-heartbeat"
 	// KindHealth is a machine's statement of its long-running resources' health (novox/hq ADR 0240).

internal/link/receive_nats.go

@@ -114,12 +114,8 @@ func (n *natsInbound) Receive(ctx context.Context, act func(context.Context, Con
 	// Heartbeats, on core NATS and off any stream (design 25 §3). Their own subscription because
 	// they are their own guarantee: a lost one is the next one.
 	beats := make(chan *nats.Msg, Prefetch)
-	alive, err := conn.ChanSubscribe(AliveSubjects, beats)
-	if err != nil {
-		return fmt.Errorf("subscribing to heartbeats: %w", err)
-	}
-	defer func() { _ = alive.Unsubscribe() }()
-	// And each machine's node tools, on the same channel: as cheap, and lost the same way.
+	// Each machine's node tools, as cheap and lost the same way. The node-engine's alive message is not subscribed to: a
+	// machine's last word is the watchman's presence (novox/hq ADR 0433 §3, issue 1183).
 	toolsAlive, err := conn.ChanSubscribe(ToolsAliveSubjects, beats)
 	if err != nil {
 		return fmt.Errorf("subscribing to the node tools' heartbeats: %w", err)
@@ -283,8 +279,6 @@ func kindOfSubject(subject string) (string, bool) {
 		switch rest {
 		case "report":
 			return KindReport, true
-		case "alive":
-			return KindHeartbeat, true
 		case "tools-alive":
 			return KindToolsHeartbeat, true
 		case "health":

internal/link/receive_nats_integration_test.go

@@ -157,24 +157,31 @@ func TestNatsAnEnrolmentIsAnsweredOnTheAddressInItsPayload(t *testing.T) {
 	}
 }
 
-func TestNatsAHeartbeatIsHeardAndNothingIsKept(t *testing.T) {
+// 1183/AC3, AC5: a node-engine still sending its alive message is harmless: it errors nothing, nothing is recorded and
+// nothing is persisted; the tool runner's tools-alive is still heard.
+func TestNatsAnAliveMessageIsHarmlessAndToolsAliveIsStillHeard(t *testing.T) { // 1183/AC3
 	js := aBus(t)
 	store := &counted{}
 	_, stop := servingOn(t, js, store)
 	defer stop()
 
-	// Given time to subscribe: a core subscription that is not yet up misses what is published,
-	// which is the guarantee a heartbeat has and not a fault.
-	eventually(t, "the heartbeat subscription coming up", func() bool {
-		body, _ := json.Marshal(Alive{Node: "anchor"})
-		_ = js.Conn().Publish(AliveSubject("anchor"), body)
+	// Given time to subscribe: a core subscription that is not yet up misses what is published.
+	eventually(t, "the tools-alive subscription coming up", func() bool {
+		if err := js.Conn().Publish("mesh.control.anchor.alive", []byte(`{"node":"anchor"}`)); err != nil {
+			t.Fatalf("an alive message could not be published: %v", err)
+		}
+		body, _ := json.Marshal(ToolsAlive{Node: "anchor"})
+		_ = js.Conn().Publish(ToolsAliveSubject("anchor"), body)
 		_ = js.Conn().Flush()
-		return store.count() >= 1
+		_, heard := ToolsBeats.Of("anchor")
+		return heard
 	})
+	if n := store.count(); n != 0 {
+		t.Fatalf("an alive message was recorded %d time(s)", n)
+	}
 	info, err := js.Context().StreamInfo("CONTROL")
 	if err != nil || info.State.Msgs != 0 {
-		t.Fatalf("a heartbeat was persisted, and the mesh's least valuable message now competes "+
-			"for retention with its most valuable: %+v, %v", info, err)
+		t.Fatalf("an alive message was persisted: %+v, %v", info, err)
 	}
 }
 

internal/link/receive_nats_test.go

@@ -125,9 +125,6 @@ type enrolsAs struct{ reply EnrolReply }
 
 func (e enrolsAs) Enrol(context.Context, EnrolRequest) (EnrolReply, error) { return e.reply, nil }
 
-// A heartbeat is core NATS: it reaches the controller and nothing is persisted, so the stream the
-// reports live in stays empty.
-
 // The two events the controller follows arrive over one durable consumer with two filters, and it
 // can acknowledge them.
 //

internal/link/serve.go

@@ -205,8 +205,8 @@ func (s *Server) Close() {
 
 // Serve acts on what arrives until the context ends.
 func (s *Server) Serve(ctx context.Context) error {
-	s.log.Printf("consuming what nodes say: %s, %s, %s, %s",
-		KindEnrolment, KindReport, KindHeartbeat, KindBuilt)
+	s.log.Printf("consuming what nodes say: %s, %s, %s",
+		KindEnrolment, KindReport, KindBuilt)
 	if s.upgrader != nil {
 		s.log.Printf("following %s", KindModuleMoved)
 	}
@@ -226,8 +226,6 @@ func (s *Server) act(ctx context.Context, m Control) {
 		s.enrolling(ctx, m)
 	case KindReport:
 		s.reported(ctx, m)
-	case KindHeartbeat:
-		s.heartbeat(m)
 	case KindToolsHeartbeat:
 		s.toolsHeartbeat(m)
 	case KindHealth:
@@ -324,28 +322,6 @@ func (s *Server) outstanding(ctx context.Context, node string) string {
 	return digest
 }
 
-// heartbeat records that a node was heard from, and nothing else.
-//
-// Deliberately silent: a node saying it is there every minute would fill the log with the ordinary
-// case, and a log where the ordinary case is loud is a log nobody reads. Not held for the store
-// either — the next heartbeat is a minute away, and a heartbeat kept for two minutes to be written
-// late says nothing the one after it will not say better.
-func (s *Server) heartbeat(m Control) {
-	var alive Alive
-	if err := json.Unmarshal(m.Body(), &alive); err != nil || alive.Node == "" {
-		_ = m.Drop()
-		return
-	}
-	// The interval it says, for the watchdog's bound (novox/hq to-be 45 S1).
-	HostBeats.Heard(alive.Node, time.Now(), time.Duration(alive.IntervalSeconds)*time.Second, "")
-	if s.listener != nil {
-		if _, err := s.listener.Heard(context.Background(), Report{Node: alive.Node}); err != nil {
-			s.log.Printf("could not record that %s is here: %v", alive.Node, err)
-		}
-	}
-	_ = m.Took()
-}
-
 // toolsHeartbeat records that a machine's node tools were heard from (novox/hq to-be 45 S11), and
 // nothing else: in this process's memory, for the watchdog — the next one is a minute away.
 func (s *Server) toolsHeartbeat(m Control) {

internal/link/stale_report_test.go

@@ -121,15 +121,3 @@ func TestAReportCarryingWhatOnlyTheNodeKnowsIsActedOnHoweverOldItIs(t *testing.T
 		}
 	}
 }
-
-// A heartbeat is not held for the store: the next one is a minute away, and one kept for two
-// minutes to be written late says nothing the one after it will not say better.
-func TestAHeartbeatIsNotHeldForTheStore(t *testing.T) {
-	s, in := serving()
-	s.listener = heardWith{err: errors.Join(ErrTryAgain, errors.New("starting up"))}
-	to := &settled{}
-	s.act(context.Background(), in.sends(t, to, KindHeartbeat, Alive{Node: "anchor"}))
-	if !to.acked || len(in.held) != 0 {
-		t.Fatalf("a heartbeat was held for the store: %+v, %d held", *to, len(in.held))
-	}
-}

internal/link/watched.go

@@ -21,8 +21,7 @@ import (
 //
 // **In memory, on purpose.** A heartbeat is the least valuable message the mesh sends — the next one
 // is a minute away — and what a watchdog needs of it is the newest moment and the interval it said.
-// A machine's last word is kept in the store as well (`last_seen`), which is what S1 reads; these are
-// what the store does not keep: the interval, the node tools' word, and when the event loop last took
+// What the store does not keep: the node tools' word and its interval, and when the event loop last took
 // a message.
 
 // Beat is one emitter's newest heartbeat.
@@ -43,11 +42,9 @@ type Beats struct {
 // NewBeats is an empty record, started now.
 func NewBeats() *Beats { return &Beats{started: time.Now(), heard: map[string]Beat{}} }
 
-// HostBeats are the node-engines' heartbeats (S1); ToolsBeats the node tools' (S11).
-var (
-	HostBeats  = NewBeats()
-	ToolsBeats = NewBeats()
-)
+// ToolsBeats are the node tools' heartbeats (S11). The node-engine's alive message is not heard: a machine's last word is
+// the watchman's presence (novox/hq ADR 0433 §3).
+var ToolsBeats = NewBeats()
 
 // Heard records one heartbeat.
 func (b *Beats) Heard(node string, at time.Time, every time.Duration, version string) {