Merge pull request 'Release v1.2.1 — stable Agent updates controls' (#53) from release-v1.2.1 into main

.gitea/workflows/ci.yml

@@ -3,7 +3,7 @@
 # Notes for anyone editing this file:
 #
 # Custom runner image
-#   Every job runs inside `gitea.dcglab.co.uk/steve/ci-runner-go`
+#   Every job runs inside `docker.dcglab.co.uk/public/ci-runner-go:latest`
 #   (recipe: https://gitea.dcglab.co.uk/steve/ci/src/branch/main/images/ci-runner-go).
 #   That image already ships:
 #     * Go on PATH at /usr/local/go/bin (so `actions/setup-go` is
@@ -11,8 +11,8 @@
 #       otherwise re-download Go on every job)
 #     * Node.js + npm (used by docs / e2e workflows)
 #     * Docker CLI, Buildx, Compose v2 (used by docker-build steps)
-#   When bumping the Go floor, push a new ci-runner-go image with
+#   The runner image is intentionally tracked via `latest`: it is shared CI
-#   the matching Go version and bump the date pin in IMAGE below.
+#   infrastructure, and updates should propagate to consuming repositories.
 #
 # Self-hosted runner expectations
 #   Each runner host bind-mounts persistent volumes for
@@ -71,10 +71,7 @@ jobs:
     name: Test (${{ matrix.name }})
     runs-on: ubuntu-latest
     container:
-      image: docker.dcglab.co.uk/ci-runner-go:2026-05-15
+      image: docker.dcglab.co.uk/public/ci-runner-go:latest
-      credentials:
-        username: ${{ secrets.ZOT_USERNAME }}
-        password: ${{ secrets.ZOT_PASSWORD }}
     strategy:
       fail-fast: false
       matrix:
@@ -110,10 +107,7 @@ jobs:
     name: Lint
     runs-on: ubuntu-latest
     container:
-      image: docker.dcglab.co.uk/ci-runner-go:2026-05-15
+      image: docker.dcglab.co.uk/public/ci-runner-go:latest
-      credentials:
-        username: ${{ secrets.ZOT_USERNAME }}
-        password: ${{ secrets.ZOT_PASSWORD }}
     steps:
       - uses: actions/checkout@v4
       - uses: golangci/golangci-lint-action@v7
@@ -130,10 +124,7 @@ jobs:
     name: Build (${{ matrix.goos }}/${{ matrix.goarch }})
     runs-on: ubuntu-latest
     container:
-      image: docker.dcglab.co.uk/ci-runner-go:2026-05-15
+      image: docker.dcglab.co.uk/public/ci-runner-go:latest
-      credentials:
-        username: ${{ secrets.ZOT_USERNAME }}
-        password: ${{ secrets.ZOT_PASSWORD }}
     strategy:
       fail-fast: false
       matrix:

.gitea/workflows/e2e.yml

@@ -31,7 +31,7 @@ jobs:
   e2e:
     name: Playwright vs docker-compose
     runs-on: ubuntu-latest
-    container: gitea.dcglab.co.uk/steve/ci-runner-go:2026-05-08
+    container: docker.dcglab.co.uk/public/ci-runner-go:latest
     timeout-minutes: 15
     steps:
       - uses: actions/checkout@v4

.gitea/workflows/release.yml

@@ -12,7 +12,9 @@
 #     plus install.sh / install.ps1 / the systemd unit baked in under
 #     /opt/restic-manager/dist (the read-only fallback path the server
 #     handlers use when <DataDir>/... is empty).
-#   * Pushes to zot OCI registry (docker.dcglab.co.uk).
+#   * Pushes to the public namespace in the zot OCI registry
+#     (docker.dcglab.co.uk/public/restic-manager). Pulls are anonymous;
+#     pushes remain authenticated.
 #
 # Tag fan-out
 #   * tag push: :vX.Y.Z, :X.Y, :X
@@ -29,7 +31,7 @@ on:
 
 env:
   REGISTRY: docker.dcglab.co.uk
-  IMAGE_NAME: restic-manager
+  IMAGE_NAME: public/restic-manager
 
 # Force bash as the default shell — see ci.yml header.
 defaults:
@@ -41,10 +43,7 @@ jobs:
     name: Build + push image
     runs-on: ubuntu-latest
     container:
-      image: docker.dcglab.co.uk/ci-runner-go:2026-05-15
+      image: docker.dcglab.co.uk/public/ci-runner-go:latest
-      credentials:
-        username: ${{ secrets.ZOT_USERNAME }}
-        password: ${{ secrets.ZOT_PASSWORD }}
     steps:
       - uses: actions/checkout@v4
 

CHANGELOG.md

@@ -6,6 +6,49 @@ and the project follows [Semantic Versioning](https://semver.org/).
 
 ## [Unreleased]
 
+## [1.2.1] - 2026-08-22
+
+### Fixed
+
+- Made the Agent updates host-selection screen stable and responsive: removed
+  the redundant typed-count confirmation, corrected name/version/eligibility
+  filtering, and stopped replacing idle or terminal controls through polling.
+  Progress polling now runs only while a rollout is actively running. ([#50])
+
+## [1.2.0] - 2026-08-22
+
+### Added
+
+- Fleet-wide agent updates are now discoverable from Settings and the
+  dashboard, with arbitrary host subsets, name/tag/version/eligibility
+  filtering, explicit exclusion reasons, server-side membership validation,
+  and a canary-first pause after the first verified reconnect. Halted rolls
+  expose retry and resume actions. ([#43])
+- Running jobs with no recent persisted activity are detected server-side and
+  surfaced through deduplicated alerts. Kind-aware thresholds protect normal
+  long-running work, terminal jobs self-resolve, and Prometheus exports stuck
+  job count and oldest inactivity age. ([#41])
+- Snapshot projections now retain host-level refresh timestamps—including
+  authoritative empty reports—expose a derived stale flag, refresh after
+  backup, forget, and prune, and support an explicit operator reconciliation
+  endpoint. ([#40])
+
+### Fixed
+
+- Raised the bounded WebSocket read limit on both peers so ordinary large
+  snapshot reports and restic events no longer disconnect otherwise healthy
+  agents. Regression coverage exercises payloads beyond the library's former
+  32 KiB default in both directions. ([#44])
+- Fleet-update timeout verification now performs a final authoritative read of
+  the agent version delivered by the reconnect `hello`, avoiding a poll/deadline
+  race and reporting the last observed version when verification fails. ([#43])
+
+### Changed
+
+- CI runner images and release images use the anonymous-pull `public`
+  namespace at `docker.dcglab.co.uk`; registry authentication remains required
+  only for publishing.
+
 ## [1.1.1] - 2026-08-22
 
 ### Fixed
@@ -140,7 +183,8 @@ with a web UI, JSON API, and self-updating agent fleet.
   go vet, golangci-lint).
 - Threat model published (`docs/threat-model.md`).
 
-[Unreleased]: https://gitea.dcglab.co.uk/steve/restic-manager/compare/v1.1.1...HEAD
+[Unreleased]: https://gitea.dcglab.co.uk/steve/restic-manager/compare/v1.2.0...HEAD
+[1.2.0]: https://gitea.dcglab.co.uk/steve/restic-manager/compare/v1.1.1...v1.2.0
 [1.1.1]: https://gitea.dcglab.co.uk/steve/restic-manager/compare/v1.1.0...v1.1.1
 [1.1.0]: https://gitea.dcglab.co.uk/steve/restic-manager/releases/tag/v1.1.0
 [1.0.0]: https://gitea.dcglab.co.uk/steve/restic-manager/releases/tag/v1.0.0
@@ -148,3 +192,7 @@ with a web UI, JSON API, and self-updating agent fleet.
 [#37]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/37
 [#36]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/36
 [#34]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/34
+[#40]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/40
+[#41]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/41
+[#43]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/43
+[#44]: https://gitea.dcglab.co.uk/steve/restic-manager/issues/44

internal/alert/engine.go

@@ -61,9 +61,20 @@ type Engine struct {
 	stuckThresholds   map[string]time.Duration
 
 	closeOnce sync.Once
+	notifyWG  sync.WaitGroup
 	done      chan struct{}
 }
 
+func (e *Engine) dispatchNotification(ctx context.Context, payload notification.Payload) {
+	e.notifyWG.Add(1)
+	go func() {
+		defer e.notifyWG.Done()
+		e.hub.Dispatch(ctx, payload)
+	}()
+}
+
+func (e *Engine) waitNotifications() { e.notifyWG.Wait() }
+
 // NewEngine builds the engine. agentOfflineFloor + tickPeriod default
 // to 15min and 60s respectively when zero.
 func NewEngine(st *store.Store, hub *notification.Hub) *Engine {

internal/alert/rules.go

@@ -60,7 +60,7 @@ func (e *Engine) raiseAndNotify(ctx context.Context, hostID, kind, dedupKey, sev
 	if err == nil {
 		hostName = host.Name
 	}
-	go e.hub.Dispatch(ctx, notification.Payload{
+	e.dispatchNotification(ctx, notification.Payload{
 		Event:    notification.EventRaised,
 		AlertID:  id,
 		Severity: severity,
@@ -85,7 +85,7 @@ func (e *Engine) Acknowledge(ctx context.Context, alertID, userID string, when t
 		return nil //nolint:nilerr
 	}
 	p := alertPayload(ctx, e.store, notification.EventAcknowledged, a)
-	go e.hub.Dispatch(context.WithoutCancel(ctx), p)
+	e.dispatchNotification(context.WithoutCancel(ctx), p)
 	return nil
 }
 
@@ -99,7 +99,7 @@ func (e *Engine) Resolve(ctx context.Context, alertID string, when time.Time) er
 		return nil
 	}
 	p := alertPayload(ctx, e.store, notification.EventResolved, a)
-	go e.hub.Dispatch(context.WithoutCancel(ctx), p)
+	e.dispatchNotification(context.WithoutCancel(ctx), p)
 	return nil
 }
 
@@ -164,7 +164,7 @@ func (e *Engine) resolveAndNotify(ctx context.Context, hostID, kind, dedupKey st
 		if a.Kind != kind || a.DedupKey != dedupKey {
 			continue
 		}
-		go e.hub.Dispatch(ctx, notification.Payload{
+		e.dispatchNotification(ctx, notification.Payload{
 			Event:    notification.EventResolved,
 			AlertID:  a.ID,
 			Severity: a.Severity,

internal/alert/rules_test.go

@@ -25,6 +25,7 @@ func setupEngine(t *testing.T) (*Engine, *store.Store, string) {
 	aead, _ := crypto.NewAEAD(key)
 	hub := notification.NewHub(st, aead, "https://rm.example")
 	eng := NewEngine(st, hub)
+	t.Cleanup(eng.waitNotifications)
 	hostID := ulid.Make().String()
 	if err := st.CreateHost(context.Background(), store.Host{
 		ID: hostID, Name: "alfa-01", OS: "linux", Arch: "amd64",

internal/server/http/fleet_update_test.go

@@ -6,7 +6,9 @@ import (
 	"bytes"
 	"context"
 	"encoding/json"
+	"io"
 	stdhttp "net/http"
+	"strings"
 	"sync"
 	"testing"
 	"time"
@@ -19,6 +21,67 @@ import (
 	"gitea.dcglab.co.uk/steve/restic-manager/internal/version"
 )
 
+func TestFleetUpdateIdlePageDoesNotPollOrRequireTypedConfirmation(t *testing.T) {
+	t.Parallel()
+	_, baseURL, st := newTestServerWithUI(t)
+	cookie := loginAsAdmin(t, st)
+	_ = makeHost(t, st, "fleet-filter-host")
+
+	req, _ := stdhttp.NewRequest("GET", baseURL+"/settings/fleet-update", nil)
+	req.AddCookie(cookie)
+	res, err := stdhttp.DefaultClient.Do(req)
+	if err != nil {
+		t.Fatalf("get fleet update page: %v", err)
+	}
+	defer res.Body.Close()
+	if res.StatusCode != stdhttp.StatusOK {
+		t.Fatalf("status: got %d, want 200", res.StatusCode)
+	}
+	raw, _ := io.ReadAll(res.Body)
+	body := string(raw)
+
+	for _, want := range []string{
+		`id="fleet-filter-name"`,
+		`id="fleet-filter-version"`,
+		`id="fleet-filter-state"`,
+		"r.style.display=",
+	} {
+		if !strings.Contains(body, want) {
+			t.Errorf("page missing %q", want)
+		}
+	}
+	if strings.Contains(body, "Type selected count to confirm") || strings.Contains(body, `id="fleet-update-confirm"`) {
+		t.Error("page still renders the typed-count confirmation")
+	}
+	if strings.Contains(body, `hx-trigger="every 3s`) {
+		t.Error("idle selection page must not poll and replace operator input")
+	}
+}
+
+func TestFleetUpdateRunningPagePollsForProgress(t *testing.T) {
+	t.Parallel()
+	_, baseURL, st := newTestServerWithUI(t)
+	cookie, userID := loginAsAdminWithID(t, st)
+	hostID := makeHost(t, st, "fleet-running-host")
+	if err := st.CreateFleetUpdate(context.Background(), store.FleetUpdate{
+		ID: ulid.Make().String(), StartedByUserID: userID, TargetVersion: version.Version,
+	}, []string{hostID}); err != nil {
+		t.Fatalf("create fleet update: %v", err)
+	}
+
+	req, _ := stdhttp.NewRequest("GET", baseURL+"/settings/fleet-update", nil)
+	req.AddCookie(cookie)
+	res, err := stdhttp.DefaultClient.Do(req)
+	if err != nil {
+		t.Fatalf("get fleet update page: %v", err)
+	}
+	defer res.Body.Close()
+	raw, _ := io.ReadAll(res.Body)
+	if body := string(raw); !strings.Contains(body, `hx-trigger="every 3s`) {
+		t.Error("running rollout page must poll for progress")
+	}
+}
+
 // fakeFleetWorker stands in for *fleetupdate.Worker in HTTP tests.
 // It records what was passed to Start/Cancel and lets tests inject
 // canned errors. Satisfies the FleetWorker interface in

web/templates/pages/fleet_update.html

@@ -28,5 +28,15 @@
 
   {{template "fleet_update_inner" .}}
 
+  <script>
+    function fleetBoxes(){return Array.from(document.querySelectorAll('.fleet-host:not(:disabled)'))}
+    function fleetReview(){const n=fleetBoxes().filter(x=>x.checked).length;document.getElementById('fleet-selected-count').textContent=n;document.getElementById('fleet-timeout').textContent=n+' × 95s';document.getElementById('fleet-update-start-btn').disabled=n===0}
+    function fleetFilter(){const q=document.getElementById('fleet-filter-name').value.toLowerCase(),v=document.getElementById('fleet-filter-version').value.toLowerCase(),s=document.getElementById('fleet-filter-state').value;document.querySelectorAll('.fleet-candidate').forEach(r=>r.style.display=r.dataset.name.toLowerCase().includes(q)&&r.dataset.version.toLowerCase().includes(v)&&(s==='all'||r.dataset.state===s)?'':'none')}
+    function fleetSelectVisible(on){document.querySelectorAll('.fleet-candidate').forEach(r=>{const x=r.querySelector('.fleet-host:not(:disabled)');if(x&&r.style.display!=='none')x.checked=on});fleetReview()}
+    function fleetSelectOnly(id){fleetBoxes().forEach(x=>x.checked=x.value===id);fleetReview()}
+    async function fleetStart(e){e.preventDefault();const ids=fleetBoxes().filter(x=>x.checked).map(x=>x.value),out=document.getElementById('fleet-start-error');out.textContent='';const res=await fetch('/api/fleet/update',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({host_ids:ids,canary_first:document.getElementById('fleet-canary').checked})});if(res.ok){location.reload();return false}const body=await res.json();out.textContent=body.message||body.code||'Unable to start';return false}
+    if(document.getElementById('fleet-selected-count'))fleetReview()
+  </script>
+
 </div>
 {{end}}

web/templates/partials/fleet_update_inner.html

@@ -7,11 +7,11 @@
 */}}
 {{define "fleet_update_inner"}}
 {{$page := .Page}}
-<div id="fleet-update-panel" class="mt-5"
+<div id="fleet-update-panel" class="mt-5"{{if and $page.Active (eq $page.Active.Status "running")}}
      hx-get="{{$page.PollURL}}"
      hx-trigger="every 3s [document.visibilityState==='visible']"
      hx-select="#fleet-update-panel"
-     hx-swap="outerHTML">
+     hx-swap="outerHTML"{{end}}>
 
 {{if and $page.Active (eq $page.Active.Status "running")}}
 
@@ -172,20 +172,9 @@
     </div>
     <form class="mt-3 flex items-center gap-3" onsubmit="return fleetStart(event)">
 	  <label class="text-[11.5px] text-ink-mute"><input id="fleet-canary" type="checkbox" checked> Pause after first host for canary review</label>
-      <label class="text-[11.5px] text-ink-mute">Type selected count to confirm:</label>
+      <button id="fleet-update-start-btn" class="btn btn-amber">Start agent update</button>
-      <input id="fleet-update-confirm" class="field mono text-[12.5px]" style="width:80px;padding:5px 8px" oninput="fleetReview()" autocomplete="off">
-      <button id="fleet-update-start-btn" class="btn btn-amber" disabled>Start agent update</button>
       <span id="fleet-start-error" class="text-bad text-[12px]"></span>
     </form>
-    <script>
-      function fleetBoxes(){return Array.from(document.querySelectorAll('.fleet-host:not(:disabled)'))}
-      function fleetReview(){const n=fleetBoxes().filter(x=>x.checked).length;document.getElementById('fleet-selected-count').textContent=n;document.getElementById('fleet-timeout').textContent=n+' × 95s';document.getElementById('fleet-update-start-btn').disabled=n===0||document.getElementById('fleet-update-confirm').value!==String(n)}
-      function fleetFilter(){const q=document.getElementById('fleet-filter-name').value.toLowerCase(),v=document.getElementById('fleet-filter-version').value.toLowerCase(),s=document.getElementById('fleet-filter-state').value;document.querySelectorAll('.fleet-candidate').forEach(r=>r.hidden=!(r.dataset.name.toLowerCase().includes(q)&&r.dataset.version.toLowerCase().includes(v)&&(s==='all'||r.dataset.state===s)))}
-      function fleetSelectVisible(on){document.querySelectorAll('.fleet-candidate:not([hidden]) .fleet-host:not(:disabled)').forEach(x=>x.checked=on);fleetReview()}
-	  function fleetSelectOnly(id){fleetBoxes().forEach(x=>x.checked=x.value===id);fleetReview()}
-	  async function fleetStart(e){e.preventDefault();const ids=fleetBoxes().filter(x=>x.checked).map(x=>x.value),out=document.getElementById('fleet-start-error');out.textContent='';const res=await fetch('/api/fleet/update',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({host_ids:ids,canary_first:document.getElementById('fleet-canary').checked})});if(res.ok){location.reload();return false}const body=await res.json();out.textContent=body.message||body.code||'Unable to start';return false}
-      fleetReview()
-    </script>
   {{end}}
 </div>
 {{end}}