hub v0.140.0: a failed operator mail is retried (1, 5, 15 min); a Docker set needs a passing memory-kill check on every ring-0 box (decision 157)

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0159rPz1ZhFKsS53msqPYxtS

hub/internal/web/system_test.go

@@ -1,6 +1,8 @@
 package web
 
 import (
+	"encoding/json"
+	"fmt"
 	"net/http"
 	"net/http/httptest"
 	"net/url"
@@ -80,19 +82,45 @@ func TestSystemPage_DockerButtonOnlyWhenReady(t *testing.T) {
 	// and drop it (the flake seen 2026-10-05 under full-suite load).
 	clock := time.Date(2026, 10, 5, 3, 0, 0, 0, time.UTC)
 	svc.Now = func() time.Time { return clock }
-	night := func() {
-		if err := svc.Ingest("full-1", osupdates.Report{RunID: time.Now().String(), Layer: "docker", Trigger: "night", Mode: "apply",
+	nightOOM := func(oom string) {
+		r := osupdates.Report{RunID: time.Now().String(), Layer: "docker", Trigger: "night", Mode: "apply",
+			Outcome: "nothing", Healthy: true, Installed: []osupdates.Package{{Name: "docker-ce", Version: "5:29.8.2-1", Origin: "Docker"}}}
+		if oom != "" {
+			r.OOMCheck = json.RawMessage(oom)
+		}
+		if err := svc.Ingest("full-1", r); err != nil {
+			t.Fatal(err)
+		}
+	}
+	pass := `{"result":"pass","oom_killed":true,"oom_event":true,"exit_code":137,"image":"felhom-controller","detail":""}`
+	nightOOM(pass)
+	if strings.Contains(getSystem(t, s), `action="/os/approve-docker"`) {
+		t.Fatal("button shown after ONE night")
+	}
+	nightOOM(pass)
+	if !strings.Contains(getSystem(t, s), `action="/os/approve-docker"`) {
+		t.Fatal("button missing after two healthy nights")
+	}
+}
+
+// Decision 157: two healthy nights WITHOUT the memory-kill check (an older agent) show no button, and the page says why.
+func TestSystemPage_DockerButtonHiddenWithoutTheMemoryKillCheck(t *testing.T) {
+	s, st, svc := systemServer(t)
+	_ = st.SetOSRing("full-1", 0)
+	clock := time.Date(2026, 10, 5, 3, 0, 0, 0, time.UTC)
+	svc.Now = func() time.Time { return clock }
+	for i := 0; i < 2; i++ {
+		if err := svc.Ingest("full-1", osupdates.Report{RunID: time.Now().String() + fmt.Sprint(i), Layer: "docker", Trigger: "night", Mode: "apply",
 			Outcome: "nothing", Healthy: true, Installed: []osupdates.Package{{Name: "docker-ce", Version: "5:29.8.2-1", Origin: "Docker"}}}); err != nil {
 			t.Fatal(err)
 		}
 	}
-	night()
-	if strings.Contains(getSystem(t, s), `action="/os/approve-docker"`) {
-		t.Fatal("button shown after ONE night")
+	page := getSystem(t, s)
+	if strings.Contains(page, `action="/os/approve-docker"`) {
+		t.Fatal("button shown although no ring-0 box reported the memory-kill check")
 	}
-	night()
-	if !strings.Contains(getSystem(t, s), `action="/os/approve-docker"`) {
-		t.Fatal("button missing after two healthy nights")
+	if !strings.Contains(page, "memory-kill check") {
+		t.Fatal("the page does not say why the button is missing")
 	}
 }