Handle NVIDIA GSP firmware init hang with timeout fallback

- bee-nvidia-load: run insmod in background, poll /proc/devices for nvidiactl; if GSP init doesn't complete in 90s, kill insmod and retry with NVreg_EnableGpuFirmware=0. Handles EBUSY case with clear error. - Write /run/bee-nvidia-mode (gsp-on/gsp-off/gsp-stuck) for audit layer - Show GSP mode badge in sidebar: yellow for gsp-off, red for gsp-stuck - Report NvidiaGSPMode in RuntimeHealth with issue entries - Simplify GRUB menu: default (KMS+GSP), advanced submenu (GSP=off, nomodeset, fail-safe), remove load-to-RAM entry - Add pcmanfm, ristretto, mupdf, mousepad to desktop packages Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Fix service control buttons: sudo, real error output, UX feedback
2026-04-05 21:00:43 +03:00 · 2026-04-05 20:25:41 +03:00 · 2026-04-05 20:22:43 +03:00 · 2026-04-05 20:21:36 +03:00 · 2026-04-05 20:19:20 +03:00 · 2026-04-05 20:14:23 +03:00
28 changed files with 951 additions and 114 deletions
--- a/audit/internal/platform/kill_workers.go
+++ b/audit/internal/platform/kill_workers.go
@@ -15,6 +15,10 @@ var workerPatterns = []string{
 	"stress-ng",
 	"stressapptest",
 	"memtester",
+	// DCGM diagnostic workers — nvvs is spawned by dcgmi diag and survives
+	// if dcgmi is killed mid-run, leaving the GPU occupied (DCGM_ST_IN_USE).
+	"nvvs",
+	"dcgmi",
 }

 // KilledProcess describes a process that was sent SIGKILL.
--- a/audit/internal/platform/platform_stress.go
+++ b/audit/internal/platform/platform_stress.go
@@ -110,7 +110,7 @@ func (s *System) RunPlatformStress(
 			wg.Add(1)
 			go func() {
 				defer wg.Done()
-				gpuCmd := buildGPUStressCmd(loadCtx, vendor)
+				gpuCmd := buildGPUStressCmd(loadCtx, vendor, cycle.LoadSec)
 				if gpuCmd == nil {
 					return
 				}
@@ -392,6 +392,13 @@ func buildCPUStressCmd(ctx context.Context) (*exec.Cmd, error) {
 		cmdArgs = append(cmdArgs, "-M", strconv.Itoa(mb))
 	}
 	cmd := exec.CommandContext(ctx, path, cmdArgs...)
+	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
+	cmd.Cancel = func() error {
+		if cmd.Process != nil {
+			_ = syscall.Kill(-cmd.Process.Pid, syscall.SIGKILL)
+		}
+		return nil
+	}
 	cmd.Stdout = nil
 	cmd.Stderr = nil
 	if err := startLowPriorityCmd(cmd, 15); err != nil {
@@ -402,28 +409,28 @@ func buildCPUStressCmd(ctx context.Context) (*exec.Cmd, error) {

 // buildGPUStressCmd creates a GPU stress command appropriate for the detected vendor.
 // Returns nil if no GPU stress tool is available (CPU-only cycling still useful).
-func buildGPUStressCmd(ctx context.Context, vendor string) *exec.Cmd {
+func buildGPUStressCmd(ctx context.Context, vendor string, durSec int) *exec.Cmd {
 	switch strings.ToLower(vendor) {
 	case "amd":
-		return buildAMDGPUStressCmd(ctx)
+		return buildAMDGPUStressCmd(ctx, durSec)
 	case "nvidia":
-		return buildNvidiaGPUStressCmd(ctx)
+		return buildNvidiaGPUStressCmd(ctx, durSec)
 	}
 	return nil
 }

-func buildAMDGPUStressCmd(ctx context.Context) *exec.Cmd {
+func buildAMDGPUStressCmd(ctx context.Context, durSec int) *exec.Cmd {
 	rvsArgs, err := resolveRVSCommand()
 	if err != nil {
 		return nil
 	}
 	rvsPath := rvsArgs[0]
-	cfg := `actions:
+	cfg := fmt.Sprintf(`actions:
 - name: gst_platform
  device: all
  module: gst
  parallel: true
-  duration: 86400000
+  duration: %d`, durSec*1000) + `
  copy_matrix: false
  target_stress: 90
  matrix_size_a: 8640
@@ -433,13 +440,20 @@ func buildAMDGPUStressCmd(ctx context.Context) *exec.Cmd {
 	cfgFile := "/tmp/bee-platform-gst.conf"
 	_ = os.WriteFile(cfgFile, []byte(cfg), 0644)
 	cmd := exec.CommandContext(ctx, rvsPath, "-c", cfgFile)
+	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
+	cmd.Cancel = func() error {
+		if cmd.Process != nil {
+			_ = syscall.Kill(-cmd.Process.Pid, syscall.SIGKILL)
+		}
+		return nil
+	}
 	cmd.Stdout = nil
 	cmd.Stderr = nil
 	_ = startLowPriorityCmd(cmd, 10)
 	return cmd
 }

-func buildNvidiaGPUStressCmd(ctx context.Context) *exec.Cmd {
+func buildNvidiaGPUStressCmd(ctx context.Context, durSec int) *exec.Cmd {
 	path, err := satLookPath("bee-gpu-burn")
 	if err != nil {
 		path, err = satLookPath("bee-gpu-stress")
@@ -447,7 +461,17 @@ func buildNvidiaGPUStressCmd(ctx context.Context) *exec.Cmd {
 	if err != nil {
 		return nil
 	}
-	cmd := exec.CommandContext(ctx, path, "--seconds", "86400")
+	// Pass exact duration so bee-gpu-burn exits on its own when the cycle ends.
+	// Process group kill via Setpgid+Cancel is kept as a safety net for cases
+	// where the context is cancelled early (user stop, parent timeout).
+	cmd := exec.CommandContext(ctx, path, "--seconds", strconv.Itoa(durSec))
+	cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
+	cmd.Cancel = func() error {
+		if cmd.Process != nil {
+			_ = syscall.Kill(-cmd.Process.Pid, syscall.SIGKILL)
+		}
+		return nil
+	}
 	cmd.Stdout = nil
 	cmd.Stderr = nil
 	_ = startLowPriorityCmd(cmd, 10)
--- a/audit/internal/platform/runtime.go
+++ b/audit/internal/platform/runtime.go
@@ -173,6 +173,22 @@ func (s *System) collectGPURuntimeHealth(vendor string, health *schema.RuntimeHe

 	switch vendor {
 	case "nvidia":
+		if raw, err := os.ReadFile("/run/bee-nvidia-mode"); err == nil {
+			health.NvidiaGSPMode = strings.TrimSpace(string(raw))
+			if health.NvidiaGSPMode == "gsp-stuck" {
+				health.Issues = append(health.Issues, schema.RuntimeIssue{
+					Code:        "nvidia_gsp_stuck",
+					Severity:    "critical",
+					Description: "NVIDIA GSP firmware init timed out and the kernel module is stuck. Reboot and select 'GSP=off' in the boot menu.",
+				})
+			} else if health.NvidiaGSPMode == "gsp-off" {
+				health.Issues = append(health.Issues, schema.RuntimeIssue{
+					Code:        "nvidia_gsp_disabled",
+					Severity:    "warning",
+					Description: "NVIDIA GSP firmware disabled (fallback). Power management runs via CPU path — power draw readings may differ from reference hardware.",
+				})
+			}
+		}
 		health.DriverReady = strings.Contains(lsmodText, "nvidia ")
 		if !health.DriverReady {
 			health.Issues = append(health.Issues, schema.RuntimeIssue{
--- a/audit/internal/platform/sat.go
+++ b/audit/internal/platform/sat.go
@@ -25,6 +25,7 @@ var (
 	satLookPath     = exec.LookPath
 	satGlob         = filepath.Glob
 	satStat         = os.Stat
+	satFreeMemBytes = freeMemBytes

 	rocmSMIExecutableGlobs = []string{
 		"/opt/rocm/bin/rocm-smi",
@@ -381,6 +382,13 @@ func (s *System) RunNvidiaTargetedStressValidatePack(ctx context.Context, baseDi
 	if err != nil {
 		return "", err
 	}
+	// Kill any lingering nvvs/dcgmi processes from a previous interrupted run
+	// before starting — otherwise dcgmi diag fails with DCGM_ST_IN_USE (-34).
+	if killed := KillTestWorkers(); len(killed) > 0 && logFunc != nil {
+		for _, p := range killed {
+			logFunc(fmt.Sprintf("pre-flight: killed stale worker pid=%d name=%s", p.PID, p.Name))
+		}
+	}
 	return runAcceptancePackCtx(ctx, baseDir, "gpu-nvidia-targeted-stress", []satJob{
 		{name: "01-nvidia-smi-q.log", cmd: []string{"nvidia-smi", "-q"}},
 		{
@@ -407,6 +415,25 @@ func resolveDCGMGPUIndices(gpuIndices []int) ([]int, error) {
 	return all, nil
 }

+func memoryStressSizeArg() string {
+	if mb := envInt("BEE_VM_STRESS_SIZE_MB", 0); mb > 0 {
+		return fmt.Sprintf("%dM", mb)
+	}
+	availBytes := satFreeMemBytes()
+	if availBytes <= 0 {
+		return "80%"
+	}
+	availMB := availBytes / (1024 * 1024)
+	targetMB := (availMB * 2) / 3
+	if targetMB >= 256 {
+		targetMB = (targetMB / 256) * 256
+	}
+	if targetMB <= 0 {
+		return "80%"
+	}
+	return fmt.Sprintf("%dM", targetMB)
+}
+
 func (s *System) RunMemoryAcceptancePack(ctx context.Context, baseDir string, logFunc func(string)) (string, error) {
 	sizeMB := envInt("BEE_MEMTESTER_SIZE_MB", 128)
 	passes := envInt("BEE_MEMTESTER_PASSES", 1)
@@ -422,11 +449,9 @@ func (s *System) RunMemoryStressPack(ctx context.Context, baseDir string, durati
 	if seconds <= 0 {
 		seconds = envInt("BEE_VM_STRESS_SECONDS", 300)
 	}
-	// Use 80% of RAM by default; override with BEE_VM_STRESS_SIZE_MB.
-	sizeArg := "80%"
-	if mb := envInt("BEE_VM_STRESS_SIZE_MB", 0); mb > 0 {
-		sizeArg = fmt.Sprintf("%dM", mb)
-	}
+	// Base the default on current MemAvailable and keep headroom for the OS and
+	// concurrent stressors so mixed burn runs do not trip the OOM killer.
+	sizeArg := memoryStressSizeArg()
 	return runAcceptancePackCtx(ctx, baseDir, "memory-stress", []satJob{
 		{name: "01-free-before.log", cmd: []string{"free", "-h"}},
 		{name: "02-stress-ng-vm.log", cmd: []string{
--- a/audit/internal/platform/sat_test.go
+++ b/audit/internal/platform/sat_test.go
@@ -276,6 +276,37 @@ func TestEnvIntFallback(t *testing.T) {
 	}
 }

+func TestMemoryStressSizeArgUsesAvailableMemory(t *testing.T) {
+	oldFreeMemBytes := satFreeMemBytes
+	satFreeMemBytes = func() int64 { return 96 * 1024 * 1024 * 1024 }
+	t.Cleanup(func() { satFreeMemBytes = oldFreeMemBytes })
+
+	if got := memoryStressSizeArg(); got != "65536M" {
+		t.Fatalf("sizeArg=%q want 65536M", got)
+	}
+}
+
+func TestMemoryStressSizeArgRespectsOverride(t *testing.T) {
+	oldFreeMemBytes := satFreeMemBytes
+	satFreeMemBytes = func() int64 { return 96 * 1024 * 1024 * 1024 }
+	t.Cleanup(func() { satFreeMemBytes = oldFreeMemBytes })
+	t.Setenv("BEE_VM_STRESS_SIZE_MB", "4096")
+
+	if got := memoryStressSizeArg(); got != "4096M" {
+		t.Fatalf("sizeArg=%q want 4096M", got)
+	}
+}
+
+func TestMemoryStressSizeArgFallsBackWhenFreeMemoryUnknown(t *testing.T) {
+	oldFreeMemBytes := satFreeMemBytes
+	satFreeMemBytes = func() int64 { return 0 }
+	t.Cleanup(func() { satFreeMemBytes = oldFreeMemBytes })
+
+	if got := memoryStressSizeArg(); got != "80%" {
+		t.Fatalf("sizeArg=%q want 80%%", got)
+	}
+}
+
 func TestClassifySATResult(t *testing.T) {
 	tests := []struct {
 		name   string
--- a/audit/internal/platform/services.go
+++ b/audit/internal/platform/services.go
@@ -61,7 +61,9 @@ func (s *System) ServiceState(name string) string {
 }

 func (s *System) ServiceDo(name string, action ServiceAction) (string, error) {
-	raw, err := exec.Command("systemctl", string(action), name).CombinedOutput()
+	// bee-web runs as the bee user; sudo is required to control system services.
+	// /etc/sudoers.d/bee grants bee NOPASSWD:ALL.
+	raw, err := exec.Command("sudo", "systemctl", string(action), name).CombinedOutput()
 	return string(raw), err
 }

--- a/audit/internal/schema/hardware.go
+++ b/audit/internal/schema/hardware.go
@@ -20,6 +20,7 @@ type RuntimeHealth struct {
 	ExportDir     string                 `json:"export_dir,omitempty"`
 	DriverReady   bool                   `json:"driver_ready,omitempty"`
 	CUDAReady     bool                   `json:"cuda_ready,omitempty"`
+	NvidiaGSPMode string                 `json:"nvidia_gsp_mode,omitempty"` // "gsp-on", "gsp-off", "gsp-stuck"
 	NetworkStatus string                 `json:"network_status,omitempty"`
 	Issues        []RuntimeIssue         `json:"issues,omitempty"`
 	Tools         []RuntimeToolStatus    `json:"tools,omitempty"`
--- a/audit/internal/webui/api.go
+++ b/audit/internal/webui/api.go
@@ -383,11 +383,13 @@ func (h *handler) handleAPIServicesAction(w http.ResponseWriter, r *http.Request
 		return
 	}
 	result, err := h.opts.App.ServiceActionResult(req.Name, action)
+	status := "ok"
 	if err != nil {
-		writeError(w, http.StatusInternalServerError, err.Error())
-		return
+		status = "error"
 	}
-	writeJSON(w, map[string]string{"status": "ok", "output": result.Body})
+	// Always return 200 with output so the frontend can display the actual
+	// systemctl error message instead of a generic "exit status 1".
+	writeJSON(w, map[string]string{"status": status, "output": result.Body})
 }

 // ── Network ───────────────────────────────────────────────────────────────────
--- a/audit/internal/webui/jobs.go
+++ b/audit/internal/webui/jobs.go
@@ -16,6 +16,7 @@ type jobState struct {
 	subs         []chan string
 	cancel       func() // optional cancel function; nil if job is not cancellable
 	logPath      string
+	serialPrefix string
 }

 // abort cancels the job if it has a cancel function and is not yet done.
@@ -36,6 +37,9 @@ func (j *jobState) append(line string) {
 	if j.logPath != "" {
 		appendJobLog(j.logPath, line)
 	}
+	if j.serialPrefix != "" {
+		taskSerialWriteLine(j.serialPrefix + line)
+	}
 	for _, ch := range j.subs {
 		select {
 		case ch <- line:
@@ -107,8 +111,11 @@ func (m *jobManager) get(id string) (*jobState, bool) {
 	return j, ok
 }

-func newTaskJobState(logPath string) *jobState {
+func newTaskJobState(logPath string, serialPrefix ...string) *jobState {
 	j := &jobState{logPath: logPath}
+	if len(serialPrefix) > 0 {
+		j.serialPrefix = serialPrefix[0]
+	}
 	if logPath == "" {
 		return j
 	}
--- a/audit/internal/webui/pages.go
+++ b/audit/internal/webui/pages.go
@@ -33,6 +33,9 @@ a{color:var(--accent);text-decoration:none}
 .sidebar-logo{padding:18px 16px 12px;font-size:18px;font-weight:700;color:#fff;letter-spacing:-.5px}
 .sidebar-logo span{color:rgba(255,255,255,.5);font-weight:400;font-size:12px;display:block;margin-top:2px}
 .sidebar-version{padding:0 16px 14px;font-size:11px;color:rgba(255,255,255,.45)}
+.sidebar-badge{margin:0 12px 12px;padding:5px 8px;border-radius:4px;font-size:11px;font-weight:600;text-align:center}
+.sidebar-badge-warn{background:#7a4f00;color:#f6c90e}
+.sidebar-badge-crit{background:#5c1a1a;color:#ff6b6b}
 .nav{flex:1}
 .nav-item{display:block;padding:10px 16px;color:rgba(255,255,255,.7);font-size:13px;border-left:3px solid transparent;transition:all .15s}
 .nav-item:hover{color:#fff;background:rgba(255,255,255,.08)}
@@ -107,6 +110,15 @@ func layoutNav(active string, buildLabel string) string {
 		buildLabel = "dev"
 	}
 	b.WriteString(`<div class="sidebar-version">Version ` + html.EscapeString(buildLabel) + `</div>`)
+	if raw, err := os.ReadFile("/run/bee-nvidia-mode"); err == nil {
+		gspMode := strings.TrimSpace(string(raw))
+		switch gspMode {
+		case "gsp-off":
+			b.WriteString(`<div class="sidebar-badge sidebar-badge-warn">NVIDIA GSP=off</div>`)
+		case "gsp-stuck":
+			b.WriteString(`<div class="sidebar-badge sidebar-badge-crit">NVIDIA GSP stuck — reboot</div>`)
+		}
+	}
 	b.WriteString(`<nav class="nav">`)
 	for _, item := range items {
 		cls := "nav-item"
@@ -2107,9 +2119,12 @@ func renderServicesInline() string {
 	return `<p style="font-size:13px;color:var(--muted);margin-bottom:10px">` + html.EscapeString(`bee-selfheal.timer is expected to be active; the oneshot bee-selfheal.service itself is not shown as a long-running service.`) + `</p>
 <div style="display:flex;justify-content:flex-end;gap:8px;flex-wrap:wrap;margin-bottom:8px"><button class="btn btn-sm btn-secondary" onclick="restartGPUDrivers()">Restart GPU Drivers</button><button class="btn btn-sm btn-secondary" onclick="loadServices()">&#8635; Refresh</button></div>
 <div id="svc-table"><p style="color:var(--muted);font-size:13px">Loading...</p></div>
-<div id="svc-out" style="display:none;margin-top:8px" class="card">
-  <div class="card-head">Output</div>
-  <div class="card-body" style="padding:10px"><div id="svc-terminal" class="terminal" style="max-height:150px"></div></div>
+<div id="svc-out" style="display:none;margin-top:12px">
+  <div style="display:flex;align-items:center;justify-content:space-between;margin-bottom:4px">
+    <span id="svc-out-label" style="font-size:12px;font-weight:600;color:var(--muted)">Output</span>
+    <span id="svc-out-status" style="font-size:12px"></span>
+  </div>
+  <div id="svc-terminal" class="terminal" style="max-height:220px;width:100%;box-sizing:border-box"></div>
 </div>
 <script>
 function loadServices() {
@@ -2125,9 +2140,9 @@ function loadServices() {
        '<div id="'+id+'" style="display:none;margin-top:6px"><pre style="font-size:11px;white-space:pre-wrap;word-break:break-all;max-height:200px;overflow-y:auto;background:#1b1c1d;padding:8px;border-radius:4px;color:#b5cea8">'+body+'</pre></div>' +
        '</td>' +
        '<td style="white-space:nowrap">' +
-        '<button class="btn btn-sm btn-secondary" onclick="svcAction(\''+s.name+'\',\'start\')">Start</button> ' +
-        '<button class="btn btn-sm btn-secondary" onclick="svcAction(\''+s.name+'\',\'stop\')">Stop</button> ' +
-        '<button class="btn btn-sm btn-secondary" onclick="svcAction(\''+s.name+'\',\'restart\')">Restart</button>' +
+        '<button class="btn btn-sm btn-secondary" id="btn-'+s.name+'-start"   onclick="svcAction(this,\''+s.name+'\',\'start\')">Start</button> ' +
+        '<button class="btn btn-sm btn-secondary" id="btn-'+s.name+'-stop"    onclick="svcAction(this,\''+s.name+'\',\'stop\')">Stop</button> ' +
+        '<button class="btn btn-sm btn-secondary" id="btn-'+s.name+'-restart" onclick="svcAction(this,\''+s.name+'\',\'restart\')">Restart</button>' +
        '</td></tr>';
    }).join('');
    document.getElementById('svc-table').innerHTML =
@@ -2138,16 +2153,45 @@ function toggleBody(id) {
  const el = document.getElementById(id);
  if (el) el.style.display = el.style.display==='none' ? 'block' : 'none';
 }
-function svcAction(name, action) {
+function svcAction(btn, name, action) {
+  var label = btn.textContent;
+  btn.disabled = true;
+  btn.textContent = '...';
+  var out = document.getElementById('svc-out');
+  var term = document.getElementById('svc-terminal');
+  var statusEl = document.getElementById('svc-out-status');
+  var labelEl = document.getElementById('svc-out-label');
+  out.style.display = 'block';
+  labelEl.textContent = action + ' ' + name;
+  term.textContent = 'Running...';
+  statusEl.textContent = '';
+  statusEl.style.color = '';
  fetch('/api/services/action',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({name,action})})
    .then(r=>r.json()).then(d => {
-      document.getElementById('svc-out').style.display='block';
-      document.getElementById('svc-terminal').textContent = d.output || d.error || action+' '+name;
-      setTimeout(loadServices, 1000);
+      term.textContent = d.output || d.error || '(no output)';
+      term.scrollTop = term.scrollHeight;
+      if (d.status === 'ok') {
+        statusEl.textContent = '✓ done';
+        statusEl.style.color = 'var(--ok-fg, #2c662d)';
+      } else {
+        statusEl.textContent = '✗ failed';
+        statusEl.style.color = 'var(--crit-fg, #9f3a38)';
+      }
+      btn.textContent = label;
+      btn.disabled = false;
+      setTimeout(loadServices, 800);
+    }).catch(e => {
+      term.textContent = 'Request failed: ' + e;
+      statusEl.textContent = '✗ error';
+      statusEl.style.color = 'var(--crit-fg, #9f3a38)';
+      btn.textContent = label;
+      btn.disabled = false;
    });
 }
 function restartGPUDrivers() {
-  svcAction('bee-nvidia', 'restart');
+  var btn = document.querySelector('[onclick*="restartGPUDrivers"]');
+  if (!btn) { svcAction({textContent:'',disabled:false}, 'bee-nvidia', 'restart'); return; }
+  svcAction(btn, 'bee-nvidia', 'restart');
 }
 loadServices();
 </script>`
@@ -2272,6 +2316,7 @@ function usbRefresh() {
  document.getElementById('usb-targets').innerHTML = '';
  document.getElementById('usb-msg').textContent = '';
  fetch('/api/export/usb').then(r=>r.json()).then(targets => {
+    window._usbTargets = Array.isArray(targets) ? targets : [];
    const st = document.getElementById('usb-status');
    const ct = document.getElementById('usb-targets');
    if (!targets || targets.length === 0) {
@@ -2280,7 +2325,7 @@ function usbRefresh() {
    }
    st.textContent = targets.length + ' device(s) found:';
    ct.innerHTML = '<table><tr><th>Device</th><th>FS</th><th>Size</th><th>Label</th><th>Model</th><th>Actions</th></tr>' +
-      targets.map(t => {
+      targets.map((t, idx) => {
        const dev = t.device || '';
        const label = t.label || '';
        const model = t.model || '';
@@ -2291,8 +2336,8 @@ function usbRefresh() {
          '<td>'+label+'</td>' +
          '<td style="font-size:12px;color:var(--muted)">'+model+'</td>' +
          '<td style="white-space:nowrap">' +
-            '<button class="btn btn-sm btn-primary" onclick="usbExport(\'audit\','+JSON.stringify(t)+',this)">Audit JSON</button> ' +
-            '<button class="btn btn-sm btn-secondary" onclick="usbExport(\'bundle\','+JSON.stringify(t)+',this)">Support Bundle</button>' +
+            '<button class="btn btn-sm btn-primary" onclick="usbExport(\'audit\','+idx+',this)">Audit JSON</button> ' +
+            '<button class="btn btn-sm btn-secondary" onclick="usbExport(\'bundle\','+idx+',this)">Support Bundle</button>' +
            '<div class="usb-row-msg" style="margin-top:6px;font-size:12px;color:var(--muted)"></div>' +
          '</td></tr>';
      }).join('') + '</table>';
@@ -2300,7 +2345,14 @@ function usbRefresh() {
    document.getElementById('usb-status').textContent = 'Error: ' + e;
  });
 }
-window.usbExport = function(type, target, btn) {
+window.usbExport = function(type, targetIndex, btn) {
+  const target = (window._usbTargets || [])[targetIndex];
+  if (!target) {
+    const msg = document.getElementById('usb-msg');
+    msg.style.color = 'var(--err,red)';
+    msg.textContent = 'Error: USB target not found. Refresh and try again.';
+    return;
+  }
  const msg = document.getElementById('usb-msg');
  const row = btn ? btn.closest('td') : null;
  const rowMsg = row ? row.querySelector('.usb-row-msg') : null;
--- a/audit/internal/webui/serial_console.go
+++ b/audit/internal/webui/serial_console.go
@@ -0,0 +1,41 @@
+package webui
+
+import (
+	"fmt"
+	"os"
+	"strings"
+	"time"
+)
+
+var taskSerialWriteLine = writeTaskSerialLine
+
+func writeTaskSerialLine(line string) {
+	line = strings.TrimSpace(line)
+	if line == "" {
+		return
+	}
+	payload := fmt.Sprintf("%s %s\n", time.Now().UTC().Format("2006-01-02 15:04:05Z"), line)
+	for _, path := range []string{"/dev/ttyS0", "/dev/ttyS1", "/dev/console"} {
+		f, err := os.OpenFile(path, os.O_WRONLY|os.O_APPEND, 0)
+		if err != nil {
+			continue
+		}
+		_, _ = f.WriteString(payload)
+		_ = f.Close()
+		return
+	}
+}
+
+func taskSerialPrefix(t *Task) string {
+	if t == nil {
+		return "[task] "
+	}
+	return fmt.Sprintf("[task %s %s] ", t.ID, t.Name)
+}
+
+func taskSerialEvent(t *Task, event string) {
+	if t == nil {
+		return
+	}
+	taskSerialWriteLine(fmt.Sprintf("%s%s", taskSerialPrefix(t), strings.TrimSpace(event)))
+}
--- a/audit/internal/webui/server.go
+++ b/audit/internal/webui/server.go
@@ -221,6 +221,11 @@ func NewHandler(opts HandlerOptions) http.Handler {
 	// ── Infrastructure ──────────────────────────────────────────────────────
 	mux.HandleFunc("GET /healthz", h.handleHealthz)
 	mux.HandleFunc("GET /api/ready", h.handleReady)
+	mux.HandleFunc("GET /loading", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Cache-Control", "no-store")
+		w.Header().Set("Content-Type", "text/html; charset=utf-8")
+		_, _ = w.Write([]byte(loadingPageHTML))
+	})

 	// ── Existing read-only endpoints (preserved for compatibility) ──────────
 	mux.HandleFunc("GET /audit.json", h.handleAuditJSON)
@@ -265,6 +270,8 @@ func NewHandler(opts HandlerOptions) http.Handler {
 	mux.HandleFunc("POST /api/tasks/{id}/cancel", h.handleAPITasksCancel)
 	mux.HandleFunc("POST /api/tasks/{id}/priority", h.handleAPITasksPriority)
 	mux.HandleFunc("GET /api/tasks/{id}/stream", h.handleAPITasksStream)
+	mux.HandleFunc("GET /api/tasks/{id}/charts", h.handleAPITaskChartsIndex)
+	mux.HandleFunc("GET /api/tasks/{id}/chart/", h.handleAPITaskChartSVG)
 	mux.HandleFunc("GET /tasks/{id}", h.handleTaskPage)

 	// Services
@@ -1207,37 +1214,106 @@ const loadingPageHTML = `<!DOCTYPE html>
 <html lang="en">
 <head>
 <meta charset="UTF-8">
-<title>EASY-BEE</title>
+<title>EASY-BEE — Starting</title>
 <style>
 *{margin:0;padding:0;box-sizing:border-box}
 html,body{height:100%;background:#0f1117;display:flex;align-items:center;justify-content:center;font-family:'Courier New',monospace;color:#e2e8f0}
-.logo{font-size:13px;line-height:1.4;color:#f6c90e;margin-bottom:48px;white-space:pre}
-.spinner{width:48px;height:48px;border:4px solid #2d3748;border-top-color:#f6c90e;border-radius:50%;animation:spin .8s linear infinite;margin:0 auto 24px}
+.wrap{text-align:center;width:420px}
+.logo{font-size:11px;line-height:1.4;color:#f6c90e;margin-bottom:6px;white-space:pre;text-align:left}
+.subtitle{font-size:12px;color:#a0aec0;text-align:left;margin-bottom:24px;padding-left:2px}
+.spinner{width:36px;height:36px;border:3px solid #2d3748;border-top-color:#f6c90e;border-radius:50%;animation:spin .8s linear infinite;margin:0 auto 14px}
+.spinner.hidden{display:none}
@keyframes spin{to{transform:rotate(360deg)}}
-.status{font-size:14px;color:#a0aec0;letter-spacing:.05em}
+.status{font-size:13px;color:#a0aec0;margin-bottom:20px;min-height:18px}
+table{width:100%;border-collapse:collapse;font-size:12px;margin-bottom:20px;display:none}
+td{padding:3px 6px;text-align:left}
+td:first-child{color:#718096;width:55%}
+.ok{color:#68d391}
+.run{color:#f6c90e}
+.fail{color:#fc8181}
+.dim{color:#4a5568}
+.btn{background:#1a202c;color:#a0aec0;border:1px solid #2d3748;padding:7px 18px;font-size:12px;cursor:pointer;font-family:inherit;display:none}
+.btn:hover{border-color:#718096;color:#e2e8f0}
 </style>
 </head>
 <body>
-<div style="text-align:center">
+<div class="wrap">
  <div class="logo">  ███████╗ █████╗ ███████╗██╗   ██╗      ██████╗ ███████╗███████╗
  ██╔════╝██╔══██╗██╔════╝╚██╗ ██╔╝      ██╔══██╗██╔════╝██╔════╝
  █████╗  ███████║███████╗ ╚████╔╝ █████╗██████╔╝█████╗  █████╗
  ██╔══╝  ██╔══██║╚════██║  ╚██╔╝  ╚════╝██╔══██╗██╔══╝  ██╔══╝
  ███████╗██║  ██║███████║   ██║         ██████╔╝███████╗███████╗
  ╚══════╝╚═╝  ╚═╝╚══════╝   ╚═╝         ╚═════╝ ╚══════╝╚══════╝</div>
-  <div class="spinner"></div>
-  <div class="status" id="s">Starting up...</div>
+  <div class="subtitle">Hardware Audit LiveCD</div>
+  <div class="spinner" id="spin"></div>
+  <div class="status" id="st">Connecting to bee-web...</div>
+  <table id="tbl"></table>
+  <button class="btn" id="btn" onclick="go()">Open app now</button>
 </div>
 <script>
-function probe(){
-  fetch('/api/ready',{cache:'no-store'})
-    .then(function(r){
-      if(r.ok){window.location.replace('/');}
-      else{setTimeout(probe,1000);}
+(function(){
+var gone = false;
+function go(){ if(!gone){gone=true;window.location.replace('/');} }
+
+function icon(s){
+  if(s==='active')   return '<span class="ok">&#9679; active</span>';
+  if(s==='failed')   return '<span class="fail">&#10005; failed</span>';
+  if(s==='activating'||s==='reloading') return '<span class="run">&#9675; starting</span>';
+  if(s==='inactive') return '<span class="dim">&#9675; inactive</span>';
+  return '<span class="dim">'+s+'</span>';
+}
+
+function allSettled(svcs){
+  for(var i=0;i<svcs.length;i++){
+    var s=svcs[i].state;
+    if(s!=='active'&&s!=='failed'&&s!=='inactive') return false;
+  }
+  return true;
+}
+
+var pollTimer=null;
+
+function pollServices(){
+  fetch('/api/services',{cache:'no-store'})
+    .then(function(r){return r.json();})
+    .then(function(svcs){
+      if(!svcs||!svcs.length) return;
+      var tbl=document.getElementById('tbl');
+      tbl.style.display='';
+      var html='';
+      for(var i=0;i<svcs.length;i++)
+        html+='<tr><td>'+svcs[i].name+'</td><td>'+icon(svcs[i].state)+'</td></tr>';
+      tbl.innerHTML=html;
+      if(allSettled(svcs)){
+        clearInterval(pollTimer);
+        document.getElementById('spin').className='spinner hidden';
+        document.getElementById('st').textContent='Ready \u2014 opening...';
+        setTimeout(go,800);
+      }
    })
-    .catch(function(){setTimeout(probe,1000);});
+    .catch(function(){});
+}
+
+function probe(){
+  fetch('/healthz',{cache:'no-store'})
+    .then(function(r){
+      if(r.ok){
+        document.getElementById('st').textContent='bee-web running \u2014 checking services...';
+        document.getElementById('btn').style.display='';
+        pollServices();
+        pollTimer=setInterval(pollServices,1500);
+      } else {
+        document.getElementById('st').textContent='bee-web starting (status '+r.status+')...';
+        setTimeout(probe,500);
+      }
+    })
+    .catch(function(){
+      document.getElementById('st').textContent='Waiting for bee-web to start...';
+      setTimeout(probe,500);
+    });
 }
 probe();
+})();
 </script>
 </body>
 </html>`
--- a/audit/internal/webui/server_test.go
+++ b/audit/internal/webui/server_test.go
@@ -723,6 +723,111 @@ func TestTaskDetailPageRendersSavedReport(t *testing.T) {
 	}
 }

+func TestTaskDetailPageRendersCancelForRunningTask(t *testing.T) {
+	globalQueue.mu.Lock()
+	origTasks := globalQueue.tasks
+	globalQueue.tasks = []*Task{{
+		ID:        "task-live-1",
+		Name:      "CPU SAT",
+		Target:    "cpu",
+		Status:    TaskRunning,
+		CreatedAt: time.Now(),
+	}}
+	globalQueue.mu.Unlock()
+	t.Cleanup(func() {
+		globalQueue.mu.Lock()
+		globalQueue.tasks = origTasks
+		globalQueue.mu.Unlock()
+	})
+
+	handler := NewHandler(HandlerOptions{Title: "Bee Hardware Audit"})
+	rec := httptest.NewRecorder()
+	handler.ServeHTTP(rec, httptest.NewRequest(http.MethodGet, "/tasks/task-live-1", nil))
+	if rec.Code != http.StatusOK {
+		t.Fatalf("status=%d", rec.Code)
+	}
+	body := rec.Body.String()
+	if !strings.Contains(body, `Cancel</button>`) {
+		t.Fatalf("task detail page missing cancel button: %s", body)
+	}
+	if !strings.Contains(body, `function cancelTaskDetail(id)`) {
+		t.Fatalf("task detail page missing cancel handler: %s", body)
+	}
+	if !strings.Contains(body, `/api/tasks/' + id + '/cancel`) {
+		t.Fatalf("task detail page missing cancel endpoint: %s", body)
+	}
+	if !strings.Contains(body, `id="task-live-charts"`) {
+		t.Fatalf("task detail page missing live charts container: %s", body)
+	}
+	if !strings.Contains(body, `/api/tasks/' + taskId + '/charts`) {
+		t.Fatalf("task detail page missing live charts index endpoint: %s", body)
+	}
+}
+
+func TestTaskChartSVGUsesTaskTimeWindow(t *testing.T) {
+	dir := t.TempDir()
+	metricsPath := filepath.Join(dir, "metrics.db")
+	prevMetricsPath := taskReportMetricsDBPath
+	taskReportMetricsDBPath = metricsPath
+	t.Cleanup(func() { taskReportMetricsDBPath = prevMetricsPath })
+
+	db, err := openMetricsDB(metricsPath)
+	if err != nil {
+		t.Fatalf("openMetricsDB: %v", err)
+	}
+	base := time.Now().UTC()
+	samples := []platform.LiveMetricSample{
+		{Timestamp: base.Add(-3 * time.Minute), PowerW: 100},
+		{Timestamp: base.Add(-2 * time.Minute), PowerW: 200},
+		{Timestamp: base.Add(-1 * time.Minute), PowerW: 300},
+	}
+	for _, sample := range samples {
+		if err := db.Write(sample); err != nil {
+			t.Fatalf("Write: %v", err)
+		}
+	}
+	_ = db.Close()
+
+	started := base.Add(-2*time.Minute - 5*time.Second)
+	done := base.Add(-1*time.Minute + 5*time.Second)
+	globalQueue.mu.Lock()
+	origTasks := globalQueue.tasks
+	globalQueue.tasks = []*Task{{
+		ID:        "task-chart-1",
+		Name:      "Power Window",
+		Target:    "cpu",
+		Status:    TaskDone,
+		CreatedAt: started.Add(-10 * time.Second),
+		StartedAt: &started,
+		DoneAt:    &done,
+	}}
+	globalQueue.mu.Unlock()
+	t.Cleanup(func() {
+		globalQueue.mu.Lock()
+		globalQueue.tasks = origTasks
+		globalQueue.mu.Unlock()
+	})
+
+	handler := NewHandler(HandlerOptions{Title: "Bee Hardware Audit"})
+	req := httptest.NewRequest(http.MethodGet, "/api/tasks/task-chart-1/chart/server-power.svg", nil)
+	req.SetPathValue("id", "task-chart-1")
+	rec := httptest.NewRecorder()
+	handler.ServeHTTP(rec, req)
+	if rec.Code != http.StatusOK {
+		t.Fatalf("status=%d body=%s", rec.Code, rec.Body.String())
+	}
+	body := rec.Body.String()
+	if !strings.Contains(body, "System Power") {
+		t.Fatalf("task chart missing expected title: %s", body)
+	}
+	if !strings.Contains(body, "min 200") {
+		t.Fatalf("task chart stats should start from in-window sample: %s", body)
+	}
+	if strings.Contains(body, "min 100") {
+		t.Fatalf("task chart should not include pre-task sample in stats: %s", body)
+	}
+}
+
 func TestViewerRendersLatestSnapshot(t *testing.T) {
 	dir := t.TempDir()
 	path := filepath.Join(dir, "audit.json")
--- a/audit/internal/webui/task_page.go
+++ b/audit/internal/webui/task_page.go
@@ -1,11 +1,15 @@
 package webui

 import (
+	"encoding/json"
 	"fmt"
 	"html"
 	"net/http"
 	"os"
 	"strings"
+	"time"
+
+	"bee/audit/internal/platform"
 )

 func (h *handler) handleTaskPage(w http.ResponseWriter, r *http.Request) {
@@ -22,6 +26,51 @@ func (h *handler) handleTaskPage(w http.ResponseWriter, r *http.Request) {
 	_, _ = w.Write([]byte(body))
 }

+func (h *handler) handleAPITaskChartsIndex(w http.ResponseWriter, r *http.Request) {
+	task, samples, _, _, ok := h.taskSamplesForRequest(r)
+	if !ok {
+		http.NotFound(w, r)
+		return
+	}
+	type taskChartIndexEntry struct {
+		Title string `json:"title"`
+		File  string `json:"file"`
+	}
+	entries := make([]taskChartIndexEntry, 0)
+	for _, spec := range taskChartSpecsForSamples(samples) {
+		title, _, ok := renderTaskChartSVG(spec.Path, samples, taskTimelineForTask(task))
+		if !ok {
+			continue
+		}
+		entries = append(entries, taskChartIndexEntry{Title: title, File: spec.File})
+	}
+	w.Header().Set("Cache-Control", "no-store")
+	w.Header().Set("Content-Type", "application/json; charset=utf-8")
+	_ = json.NewEncoder(w).Encode(entries)
+}
+
+func (h *handler) handleAPITaskChartSVG(w http.ResponseWriter, r *http.Request) {
+	task, samples, _, _, ok := h.taskSamplesForRequest(r)
+	if !ok {
+		http.NotFound(w, r)
+		return
+	}
+	file := strings.TrimPrefix(r.URL.Path, "/api/tasks/"+task.ID+"/chart/")
+	path, ok := taskChartPathFromFile(file)
+	if !ok {
+		http.NotFound(w, r)
+		return
+	}
+	title, buf, hasData := renderTaskChartSVG(path, samples, taskTimelineForTask(task))
+	if !hasData || len(buf) == 0 || strings.TrimSpace(title) == "" {
+		http.Error(w, "metrics history unavailable", http.StatusServiceUnavailable)
+		return
+	}
+	w.Header().Set("Content-Type", "image/svg+xml")
+	w.Header().Set("Cache-Control", "no-store")
+	_, _ = w.Write(buf)
+}
+
 func renderTaskDetailPage(opts HandlerOptions, task Task) string {
 	title := task.Name
 	if strings.TrimSpace(title) == "" {
@@ -30,6 +79,9 @@ func renderTaskDetailPage(opts HandlerOptions, task Task) string {
 	var body strings.Builder
 	body.WriteString(`<div style="display:flex;align-items:center;gap:12px;margin-bottom:16px;flex-wrap:wrap">`)
 	body.WriteString(`<a class="btn btn-secondary btn-sm" href="/tasks">Back to Tasks</a>`)
+	if task.Status == TaskRunning || task.Status == TaskPending {
+		body.WriteString(`<button class="btn btn-danger btn-sm" onclick="cancelTaskDetail('` + html.EscapeString(task.ID) + `')">Cancel</button>`)
+	}
 	body.WriteString(`<span style="font-size:12px;color:var(--muted)">Artifacts are saved in the task folder under <code>./tasks</code>.</span>`)
 	body.WriteString(`</div>`)

@@ -46,16 +98,52 @@ func renderTaskDetailPage(opts HandlerOptions, task Task) string {
 	}

 	if task.Status == TaskRunning || task.Status == TaskPending {
+		body.WriteString(`<div class="card"><div class="card-head">Live Charts</div><div class="card-body">`)
+		body.WriteString(`<div id="task-live-charts" style="display:flex;flex-direction:column;gap:16px;color:var(--muted);font-size:13px">Loading charts...</div>`)
+		body.WriteString(`</div></div>`)
 		body.WriteString(`<div class="card"><div class="card-head">Live Logs</div><div class="card-body">`)
 		body.WriteString(`<div id="task-live-log" class="terminal" style="max-height:none;white-space:pre-wrap">Connecting...</div>`)
 		body.WriteString(`</div></div>`)
 		body.WriteString(`<script>
+function cancelTaskDetail(id) {
+  fetch('/api/tasks/' + id + '/cancel', {method:'POST'}).then(function(){ window.location.reload(); });
+}
+function loadTaskLiveCharts(taskId) {
+  fetch('/api/tasks/' + taskId + '/charts').then(function(r){ return r.json(); }).then(function(charts){
+    const host = document.getElementById('task-live-charts');
+    if (!host) return;
+    if (!Array.isArray(charts) || charts.length === 0) {
+      host.innerHTML = 'Waiting for metric samples...';
+      return;
+    }
+    host.innerHTML = charts.map(function(chart) {
+      return '<div class="card" style="margin:0">' +
+        '<div class="card-head">' + chart.title + '</div>' +
+        '<div class="card-body" style="padding:12px">' +
+        '<img data-task-chart="1" data-base-src="/api/tasks/' + taskId + '/chart/' + chart.file + '" src="/api/tasks/' + taskId + '/chart/' + chart.file + '?t=' + Date.now() + '" style="width:100%;display:block;border-radius:6px" alt="' + chart.title + '">' +
+        '</div></div>';
+    }).join('');
+  }).catch(function(){
+    const host = document.getElementById('task-live-charts');
+    if (host) host.innerHTML = 'Task charts are unavailable.';
+  });
+}
+function refreshTaskLiveCharts() {
+  document.querySelectorAll('img[data-task-chart="1"]').forEach(function(img){
+    const base = img.dataset.baseSrc;
+    if (!base) return;
+    img.src = base + '?t=' + Date.now();
+  });
+}
 var _taskDetailES = new EventSource('/api/tasks/` + html.EscapeString(task.ID) + `/stream');
 var _taskDetailTerm = document.getElementById('task-live-log');
+var _taskChartTimer = null;
 _taskDetailES.onopen = function(){ _taskDetailTerm.textContent = ''; };
 _taskDetailES.onmessage = function(e){ _taskDetailTerm.textContent += e.data + "\n"; _taskDetailTerm.scrollTop = _taskDetailTerm.scrollHeight; };
-_taskDetailES.addEventListener('done', function(){ _taskDetailES.close(); setTimeout(function(){ window.location.reload(); }, 1000); });
-_taskDetailES.onerror = function(){ _taskDetailES.close(); };
+_taskDetailES.addEventListener('done', function(){ if (_taskChartTimer) clearInterval(_taskChartTimer); _taskDetailES.close(); setTimeout(function(){ window.location.reload(); }, 1000); });
+_taskDetailES.onerror = function(){ if (_taskChartTimer) clearInterval(_taskChartTimer); _taskDetailES.close(); };
+loadTaskLiveCharts('` + html.EscapeString(task.ID) + `');
+_taskChartTimer = setInterval(function(){ refreshTaskLiveCharts(); loadTaskLiveCharts('` + html.EscapeString(task.ID) + `'); }, 2000);
 </script>`)
 	}

@@ -83,3 +171,37 @@ func taskArtifactDownloadLink(task Task, absPath string) string {
 	}
 	return fmt.Sprintf(`/export/file?path=%s`, absPath)
 }
+
+func (h *handler) taskSamplesForRequest(r *http.Request) (Task, []platform.LiveMetricSample, time.Time, time.Time, bool) {
+	id := r.PathValue("id")
+	taskPtr, ok := globalQueue.findByID(id)
+	if !ok {
+		return Task{}, nil, time.Time{}, time.Time{}, false
+	}
+	task := *taskPtr
+	start, end := taskTimeWindow(&task)
+	samples, err := loadTaskMetricSamples(start, end)
+	if err != nil {
+		return task, nil, start, end, true
+	}
+	return task, samples, start, end, true
+}
+
+func taskTimelineForTask(task Task) []chartTimelineSegment {
+	start, end := taskTimeWindow(&task)
+	return []chartTimelineSegment{{Start: start, End: end, Active: true}}
+}
+
+func taskChartPathFromFile(file string) (string, bool) {
+	file = strings.TrimSpace(file)
+	for _, spec := range taskDashboardChartSpecs {
+		if spec.File == file {
+			return spec.Path, true
+		}
+	}
+	if strings.HasPrefix(file, "gpu-") && strings.HasSuffix(file, "-overview.svg") {
+		id := strings.TrimSuffix(strings.TrimPrefix(file, "gpu-"), "-overview.svg")
+		return "gpu/" + id + "-overview", true
+	}
+	return "", false
+}
--- a/audit/internal/webui/task_report.go
+++ b/audit/internal/webui/task_report.go
@@ -53,6 +53,18 @@ var taskDashboardChartSpecs = []taskChartSpec{
 	{Path: "gpu-all-temp", File: "gpu-all-temp.svg"},
 }

+func taskChartSpecsForSamples(samples []platform.LiveMetricSample) []taskChartSpec {
+	specs := make([]taskChartSpec, 0, len(taskDashboardChartSpecs)+len(taskGPUIndices(samples)))
+	specs = append(specs, taskDashboardChartSpecs...)
+	for _, idx := range taskGPUIndices(samples) {
+		specs = append(specs, taskChartSpec{
+			Path: fmt.Sprintf("gpu/%d-overview", idx),
+			File: fmt.Sprintf("gpu-%d-overview.svg", idx),
+		})
+	}
+	return specs
+}
+
 func writeTaskReportArtifacts(t *Task) error {
 	if t == nil {
 		return nil
@@ -136,7 +148,7 @@ func writeTaskCharts(dir string, start, end time.Time, samples []platform.LiveMe
 	timeline := []chartTimelineSegment{{Start: start, End: end, Active: true}}
 	var charts []taskReportChart
 	inline := make(map[string]string)
-	for _, spec := range taskDashboardChartSpecs {
+	for _, spec := range taskChartSpecsForSamples(samples) {
 		title, svg, ok := renderTaskChartSVG(spec.Path, samples, timeline)
 		if !ok || len(svg) == 0 {
 			continue
@@ -148,24 +160,17 @@ func writeTaskCharts(dir string, start, end time.Time, samples []platform.LiveMe
 		charts = append(charts, taskReportChart{Title: title, File: spec.File})
 		inline[spec.File] = string(svg)
 	}
-
-	for _, idx := range taskGPUIndices(samples) {
-		file := fmt.Sprintf("gpu-%d-overview.svg", idx)
-		svg, ok, err := renderGPUOverviewChartSVG(idx, samples, timeline)
-		if err != nil || !ok || len(svg) == 0 {
-			continue
-		}
-		path := filepath.Join(dir, file)
-		if err := os.WriteFile(path, svg, 0644); err != nil {
-			continue
-		}
-		charts = append(charts, taskReportChart{Title: gpuDisplayLabel(idx) + " Overview", File: file})
-		inline[file] = string(svg)
-	}
 	return charts, inline
 }

 func renderTaskChartSVG(path string, samples []platform.LiveMetricSample, timeline []chartTimelineSegment) (string, []byte, bool) {
+	if idx, sub, ok := parseGPUChartPath(path); ok && sub == "overview" {
+		buf, hasData, err := renderGPUOverviewChartSVG(idx, samples, timeline)
+		if err != nil || !hasData {
+			return "", nil, false
+		}
+		return gpuDisplayLabel(idx) + " Overview", buf, true
+	}
 	datasets, names, labels, title, yMin, yMax, ok := chartDataFromSamples(path, samples)
 	if !ok {
 		return "", nil, false
@@ -227,13 +232,11 @@ func renderTaskReportFragment(report taskReport, charts map[string]string, logTe
 	b.WriteString(`</div></div></div>`)

 	if len(report.Charts) > 0 {
-		b.WriteString(`<div class="grid2">`)
 		for _, chart := range report.Charts {
 			b.WriteString(`<div class="card"><div class="card-head">` + html.EscapeString(chart.Title) + `</div><div class="card-body" style="padding:12px">`)
 			b.WriteString(charts[chart.File])
 			b.WriteString(`</div></div>`)
 		}
-		b.WriteString(`</div>`)
 	} else {
 		b.WriteString(`<div class="alert alert-info">No metric samples were captured during this task window.</div>`)
 	}
--- a/audit/internal/webui/tasks.go
+++ b/audit/internal/webui/tasks.go
@@ -258,6 +258,7 @@ func (q *taskQueue) enqueue(t *Task) {
 	q.prune()
 	q.persistLocked()
 	q.mu.Unlock()
+	taskSerialEvent(t, "queued")
 	select {
 	case q.trigger <- struct{}{}:
 	default:
@@ -435,7 +436,7 @@ func (q *taskQueue) worker() {
 				t.StartedAt = &now
 				t.DoneAt = nil
 				t.ErrMsg = ""
-				j := newTaskJobState(t.LogPath)
+				j := newTaskJobState(t.LogPath, taskSerialPrefix(t))
 				t.job = j
 				batch = append(batch, t)
 			}
@@ -520,6 +521,11 @@ func (q *taskQueue) finalizeTaskRun(t *Task, j *jobState) {
 	if err := writeTaskReportArtifacts(t); err != nil {
 		appendJobLog(t.LogPath, "WARN: task report generation failed: "+err.Error())
 	}
+	if t.ErrMsg != "" {
+		taskSerialEvent(t, "finished with status="+t.Status+" error="+t.ErrMsg)
+		return
+	}
+	taskSerialEvent(t, "finished with status="+t.Status)
 }

 // setCPUGovernor writes the given governor to all CPU scaling_governor sysfs files.
@@ -858,6 +864,7 @@ func (h *handler) handleAPITasksCancel(w http.ResponseWriter, r *http.Request) {
 		now := time.Now()
 		t.DoneAt = &now
 		globalQueue.persistLocked()
+		taskSerialEvent(t, "finished with status="+t.Status)
 		writeJSON(w, map[string]string{"status": "cancelled"})
 	case TaskRunning:
 		if t.job != nil {
@@ -867,6 +874,7 @@ func (h *handler) handleAPITasksCancel(w http.ResponseWriter, r *http.Request) {
 		now := time.Now()
 		t.DoneAt = &now
 		globalQueue.persistLocked()
+		taskSerialEvent(t, "finished with status="+t.Status)
 		writeJSON(w, map[string]string{"status": "cancelled"})
 	default:
 		writeError(w, http.StatusConflict, "task is not running or pending")
@@ -907,6 +915,7 @@ func (h *handler) handleAPITasksCancelAll(w http.ResponseWriter, _ *http.Request
 		case TaskPending:
 			t.Status = TaskCancelled
 			t.DoneAt = &now
+			taskSerialEvent(t, "finished with status="+t.Status)
 			n++
 		case TaskRunning:
 			if t.job != nil {
@@ -914,6 +923,7 @@ func (h *handler) handleAPITasksCancelAll(w http.ResponseWriter, _ *http.Request
 			}
 			t.Status = TaskCancelled
 			t.DoneAt = &now
+			taskSerialEvent(t, "finished with status="+t.Status)
 			n++
 		}
 	}
@@ -932,6 +942,7 @@ func (h *handler) handleAPITasksKillWorkers(w http.ResponseWriter, _ *http.Reque
 		case TaskPending:
 			t.Status = TaskCancelled
 			t.DoneAt = &now
+			taskSerialEvent(t, "finished with status="+t.Status)
 			cancelled++
 		case TaskRunning:
 			if t.job != nil {
@@ -939,6 +950,7 @@ func (h *handler) handleAPITasksKillWorkers(w http.ResponseWriter, _ *http.Reque
 			}
 			t.Status = TaskCancelled
 			t.DoneAt = &now
+			taskSerialEvent(t, "finished with status="+t.Status)
 			cancelled++
 		}
 	}
--- a/audit/internal/webui/tasks_test.go
+++ b/audit/internal/webui/tasks_test.go
@@ -325,6 +325,49 @@ func TestFinalizeTaskRunCreatesReportFolderAndArtifacts(t *testing.T) {
 	}
 }

+func TestTaskLifecycleMirrorsToSerialConsole(t *testing.T) {
+	var lines []string
+	prev := taskSerialWriteLine
+	taskSerialWriteLine = func(line string) { lines = append(lines, line) }
+	t.Cleanup(func() { taskSerialWriteLine = prev })
+
+	dir := t.TempDir()
+	q := &taskQueue{
+		statePath: filepath.Join(dir, "tasks-state.json"),
+		logsDir:   filepath.Join(dir, "tasks"),
+		trigger:   make(chan struct{}, 1),
+	}
+	task := &Task{
+		ID:        "task-serial-1",
+		Name:      "CPU SAT",
+		Target:    "cpu",
+		Status:    TaskPending,
+		CreatedAt: time.Now().UTC(),
+	}
+
+	q.enqueue(task)
+	started := time.Now().UTC()
+	task.Status = TaskRunning
+	task.StartedAt = &started
+	job := newTaskJobState(task.LogPath, taskSerialPrefix(task))
+	job.append("Starting CPU SAT...")
+	job.append("CPU stress duration: 60s")
+	job.finish("")
+	q.finalizeTaskRun(task, job)
+
+	joined := strings.Join(lines, "\n")
+	for _, needle := range []string{
+		"queued",
+		"Starting CPU SAT...",
+		"CPU stress duration: 60s",
+		"finished with status=done",
+	} {
+		if !strings.Contains(joined, needle) {
+			t.Fatalf("serial mirror missing %q in %q", needle, joined)
+		}
+	}
+}
+
 func TestResolveBurnPreset(t *testing.T) {
 	tests := []struct {
 		profile string
--- a/iso/builder/auto/config
+++ b/iso/builder/auto/config
@@ -32,7 +32,7 @@ lb config noauto \
    --memtest memtest86+ \
    --iso-volume "EASY_BEE_${BEE_GPU_VENDOR_UPPER:-NVIDIA}" \
    --iso-application "EASY-BEE-${BEE_GPU_VENDOR_UPPER:-NVIDIA}" \
-    --bootappend-live "boot=live components video=1920x1080 console=tty0 console=ttyS0,115200n8 loglevel=6 systemd.show_status=1 username=bee user-fullname=Bee modprobe.blacklist=nouveau,snd_hda_intel,snd_hda_codec_realtek,snd_hda_codec_generic,soundcore" \
+    --bootappend-live "boot=live components video=1920x1080 console=tty0 console=ttyS0,115200n8 loglevel=3 systemd.show_status=1 username=bee user-fullname=Bee modprobe.blacklist=nouveau,snd_hda_intel,snd_hda_codec_realtek,snd_hda_codec_generic,soundcore" \
    --apt-recommends false \
    --chroot-squashfs-compression-type zstd \
    "${@}"
--- a/iso/builder/config/bootloaders/grub-pc/grub.cfg
+++ b/iso/builder/config/bootloaders/grub-pc/grub.cfg
@@ -7,36 +7,29 @@ echo "  █████╗  ███████║███████╗ ╚
 echo "  ██╔══╝  ██╔══██║╚════██║  ╚██╔╝  ╚════╝██╔══██╗██╔══╝  ██╔══╝"
 echo "  ███████╗██║  ██║███████║   ██║         ██████╔╝███████╗███████╗"
 echo "  ╚══════╝╚═╝  ╚═╝╚══════╝   ╚═╝         ╚═════╝ ╚══════╝╚══════╝"
+echo "  Hardware Audit LiveCD"
 echo ""

 menuentry "EASY-BEE" {
+    linux   @KERNEL_LIVE@ @APPEND_LIVE@ bee.nvidia.mode=normal net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
+    initrd  @INITRD_LIVE@
+}
+
+submenu "EASY-BEE (advanced options) -->" {
+    menuentry "EASY-BEE — GSP=off" {
+        linux   @KERNEL_LIVE@ @APPEND_LIVE@ bee.nvidia.mode=gsp-off net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
+        initrd  @INITRD_LIVE@
+    }
+
+    menuentry "EASY-BEE — nomodeset" {
        linux   @KERNEL_LIVE@ @APPEND_LIVE@ nomodeset bee.nvidia.mode=normal net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
        initrd  @INITRD_LIVE@
    }

-menuentry "EASY-BEE (graphics/KMS)" {
-    linux   @KERNEL_LIVE@ @APPEND_LIVE@ bee.display=kms bee.nvidia.mode=normal net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
+    menuentry "EASY-BEE — fail-safe" {
+        linux   @KERNEL_LIVE@ @APPEND_LIVE@ nomodeset bee.nvidia.mode=gsp-off noapic noapm nodma nomce nolapic nosmp vga=normal net.ifnames=0 biosdevname=0
        initrd  @INITRD_LIVE@
    }
-
-menuentry "EASY-BEE (load to RAM)" {
-    linux   @KERNEL_LIVE@ @APPEND_LIVE@ toram nomodeset bee.nvidia.mode=normal net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
-    initrd  @INITRD_LIVE@
-}
-
-menuentry "EASY-BEE (NVIDIA GSP=off)" {
-    linux   @KERNEL_LIVE@ @APPEND_LIVE@ nomodeset bee.nvidia.mode=gsp-off net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
-    initrd  @INITRD_LIVE@
-}
-
-menuentry "EASY-BEE (graphics/KMS, GSP=off)" {
-    linux   @KERNEL_LIVE@ @APPEND_LIVE@ bee.display=kms bee.nvidia.mode=gsp-off net.ifnames=0 biosdevname=0 mitigations=off transparent_hugepage=always numa_balancing=disable nowatchdog nosoftlockup
-    initrd  @INITRD_LIVE@
-}
-
-menuentry "EASY-BEE (fail-safe)" {
-    linux   @KERNEL_LIVE@ @APPEND_LIVE@ bee.nvidia.mode=gsp-off memtest noapic noapm nodma nomce nolapic nosmp vga=normal net.ifnames=0 biosdevname=0
-    initrd  @INITRD_LIVE@
 }

 if [ "${grub_platform}" = "efi" ]; then
--- a/iso/builder/config/hooks/normal/9000-bee-setup.hook.chroot
+++ b/iso/builder/config/hooks/normal/9000-bee-setup.hook.chroot
@@ -31,6 +31,7 @@ systemctl enable bee-audit.service
 systemctl enable bee-web.service
 systemctl enable bee-sshsetup.service
 systemctl enable bee-selfheal.timer
+systemctl enable bee-boot-status.service
 systemctl enable ssh.service
 systemctl enable lightdm.service 2>/dev/null || true
 systemctl enable qemu-guest-agent.service 2>/dev/null || true
@@ -60,6 +61,7 @@ chmod +x /usr/local/bin/bee-smoketest  2>/dev/null || true
 chmod +x /usr/local/bin/bee            2>/dev/null || true
 chmod +x /usr/local/bin/bee-log-run    2>/dev/null || true
 chmod +x /usr/local/bin/bee-selfheal      2>/dev/null || true
+chmod +x /usr/local/bin/bee-boot-status  2>/dev/null || true
 if [ "$GPU_VENDOR" = "nvidia" ]; then
    chmod +x /usr/local/bin/bee-nvidia-load 2>/dev/null || true
    chmod +x /usr/local/bin/bee-gpu-burn 2>/dev/null || true
--- a/iso/builder/config/hooks/normal/9001-wallpaper.hook.chroot
+++ b/iso/builder/config/hooks/normal/9001-wallpaper.hook.chroot
@@ -0,0 +1,72 @@
+#!/bin/sh
+# 9001-wallpaper.hook.chroot — generate /usr/share/bee/wallpaper.png inside chroot
+set -e
+echo "=== generating bee wallpaper ==="
+mkdir -p /usr/share/bee
+
+python3 - <<'PYEOF'
+from PIL import Image, ImageDraw, ImageFont
+import os
+
+W, H = 1920, 1080
+
+LOGO = """\
+  \u2588\u2588\u2588\u2588\u2588\u2588\u2557 \u2588\u2588\u2588\u2588\u2588\u2557 \u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557\u2588\u2588\u2557   \u2588\u2588\u2557      \u2588\u2588\u2588\u2588\u2588\u2588\u2557 \u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557
+  \u2588\u2588\u2554\u2550\u2550\u2550\u2550\u255d\u2588\u2588\u2554\u2550\u2550\u2588\u2588\u2557\u2588\u2588\u2554\u2550\u2550\u2550\u2550\u255d\u255a\u2588\u2588\u2557 \u2588\u2588\u2554\u255d      \u2588\u2588\u2554\u2550\u2550\u2588\u2588\u2557\u2588\u2588\u2554\u2550\u2550\u2550\u2550\u255d\u2588\u2588\u2554\u2550\u2550\u2550\u2550\u255d
+  \u2588\u2588\u2588\u2588\u2588\u2557  \u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2551\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557 \u255a\u2588\u2588\u2588\u2588\u2554\u255d \u2588\u2588\u2588\u2588\u2588\u2557\u2588\u2588\u2588\u2588\u2588\u2588\u2554\u255d\u2588\u2588\u2588\u2588\u2588\u2557  \u2588\u2588\u2588\u2588\u2588\u2557
+  \u2588\u2588\u2554\u2550\u2550\u255d  \u2588\u2588\u2554\u2550\u2550\u2588\u2588\u2551\u255a\u2550\u2550\u2550\u2550\u2588\u2588\u2551  \u255a\u2588\u2588\u2554\u255d  \u255a\u2550\u2550\u2550\u2550\u255d\u2588\u2588\u2554\u2550\u2550\u2588\u2588\u2557\u2588\u2588\u2554\u2550\u2550\u255d  \u2588\u2588\u2554\u2550\u2550\u255d
+  \u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557\u2588\u2588\u2551  \u2588\u2588\u2551\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2551   \u2588\u2588\u2551         \u2588\u2588\u2588\u2588\u2588\u2588\u2554\u255d\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2557
+  \u255a\u2550\u2550\u2550\u2550\u2550\u2550\u255d\u255a\u2550\u255d  \u255a\u2550\u255d\u255a\u2550\u2550\u2550\u2550\u2550\u2550\u255d   \u255a\u2550\u255d         \u255a\u2550\u2550\u2550\u2550\u2550\u255d \u255a\u2550\u2550\u2550\u2550\u2550\u2550\u255d\u255a\u2550\u2550\u2550\u2550\u2550\u2550\u255d
+  Hardware Audit LiveCD"""
+
+# Find a monospace font that supports box-drawing characters
+FONT_CANDIDATES = [
+    '/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf',
+    '/usr/share/fonts/truetype/liberation/LiberationMono-Regular.ttf',
+    '/usr/share/fonts/truetype/freefont/FreeMono.ttf',
+    '/usr/share/fonts/truetype/noto/NotoMono-Regular.ttf',
+]
+
+font_path = None
+for p in FONT_CANDIDATES:
+    if os.path.exists(p):
+        font_path = p
+        break
+
+SIZE = 22
+if font_path:
+    font_logo = ImageFont.truetype(font_path, SIZE)
+    font_sub  = ImageFont.truetype(font_path, SIZE)
+else:
+    font_logo = ImageFont.load_default()
+    font_sub  = font_logo
+
+img  = Image.new('RGB', (W, H), (0, 0, 0))
+draw = ImageDraw.Draw(img)
+
+# Measure logo block
+lines = LOGO.split('\n')
+bbox = draw.textbbox((0, 0), LOGO, font=font_logo)
+text_w = bbox[2] - bbox[0]
+text_h = bbox[3] - bbox[1]
+
+x = (W - text_w) // 2
+y = (H - text_h) // 2
+
+# Draw logo lines: first 6 in amber, last line (subtitle) dimmer
+logo_lines  = lines[:6]
+sub_line    = lines[6] if len(lines) > 6 else ''
+
+cy = y
+for line in logo_lines:
+    draw.text((x, cy), line, font=font_logo, fill=(0xf6, 0xc9, 0x0e))
+    cy += SIZE + 2
+cy += 8
+if sub_line:
+    draw.text((x, cy), sub_line, font=font_sub, fill=(0x80, 0x68, 0x18))
+
+img.save('/usr/share/bee/wallpaper.png', optimize=True)
+print('wallpaper written: /usr/share/bee/wallpaper.png')
+PYEOF
+
+echo "=== wallpaper done ==="
--- a/iso/builder/config/package-lists/bee.list.chroot
+++ b/iso/builder/config/package-lists/bee.list.chroot
@@ -60,9 +60,15 @@ qrencode
 # Local desktop (openbox + chromium kiosk)
 openbox
 tint2
+feh
+python3-pil
 xorg
 xterm
 chromium
+mousepad
+pcmanfm
+ristretto
+mupdf
 xserver-xorg-video-fbdev
 xserver-xorg-video-vesa
 lightdm
--- a/iso/overlay/etc/systemd/system/bee-boot-status.service
+++ b/iso/overlay/etc/systemd/system/bee-boot-status.service
@@ -0,0 +1,18 @@
+[Unit]
+Description=Bee: boot status display
+After=systemd-user-sessions.service
+Before=getty@tty1.service
+
+[Service]
+Type=oneshot
+RemainAfterExit=no
+ExecStart=/usr/local/bin/bee-boot-status
+TTYPath=/dev/tty1
+StandardInput=tty
+StandardOutput=tty
+StandardError=tty
+TTYReset=yes
+TTYVHangup=yes
+
+[Install]
+WantedBy=multi-user.target
--- a/iso/overlay/etc/systemd/system/getty@tty1.service.d/wait-bee.conf
+++ b/iso/overlay/etc/systemd/system/getty@tty1.service.d/wait-bee.conf
@@ -0,0 +1,2 @@
+[Unit]
+After=bee-boot-status.service
--- a/iso/overlay/etc/systemd/system/lightdm.service.d/bee-display-mode.conf
+++ b/iso/overlay/etc/systemd/system/lightdm.service.d/bee-display-mode.conf
@@ -1,6 +1,4 @@
 [Unit]
-Wants=bee-preflight.service
-After=bee-preflight.service

 [Service]
 ExecStartPre=/usr/local/bin/bee-display-mode
--- a/iso/overlay/usr/local/bin/bee-boot-status
+++ b/iso/overlay/usr/local/bin/bee-boot-status
@@ -0,0 +1,89 @@
+#!/bin/sh
+# bee-boot-status — boot progress display on tty1.
+# Shows live service status until all bee services are done or failed,
+# then exits so getty can show the login prompt.
+
+CRITICAL="bee-preflight bee-nvidia bee-audit"
+ALL="bee-sshsetup ssh bee-network bee-nvidia bee-preflight bee-audit bee-web"
+
+svc_state() { systemctl is-active "$1.service" 2>/dev/null || echo "inactive"; }
+
+svc_icon() {
+    case "$(svc_state "$1")" in
+        active)       printf '\033[32m[  OK  ]\033[0m' ;;
+        failed)       printf '\033[31m[ FAIL ]\033[0m' ;;
+        activating)   printf '\033[33m[  ..  ]\033[0m' ;;
+        deactivating) printf '\033[33m[ stop ]\033[0m' ;;
+        inactive)     printf '\033[90m[      ]\033[0m' ;;
+        *)            printf '\033[90m[  ?   ]\033[0m' ;;
+    esac
+}
+
+svc_detail() {
+    local svc="$1" state
+    state="$(svc_state "$svc")"
+    case "$state" in
+        failed)
+            local res
+            res="$(systemctl show -p Result "$svc.service" 2>/dev/null | cut -d= -f2)"
+            [ -n "$res" ] && [ "$res" != "success" ] && printf '  \033[31m(%s)\033[0m' "$res"
+            ;;
+        activating)
+            local line
+            line="$(journalctl -u "$svc.service" -n 1 --no-pager --output=cat 2>/dev/null | cut -c1-55)"
+            [ -n "$line" ] && printf '  \033[90m%s\033[0m' "$line"
+            ;;
+    esac
+}
+
+all_critical_done() {
+    for svc in $CRITICAL; do
+        case "$(svc_state "$svc")" in
+            active|failed|inactive) ;;
+            *) return 1 ;;
+        esac
+    done
+    return 0
+}
+
+while true; do
+    # move to top-left and clear screen
+    printf '\033[H\033[2J'
+
+    printf '\n'
+    printf '  \033[33m███████╗ █████╗ ███████╗██╗   ██╗      ██████╗ ███████╗███████╗\033[0m\n'
+    printf '  \033[33m██╔════╝██╔══██╗██╔════╝╚██╗ ██╔╝      ██╔══██╗██╔════╝██╔════╝\033[0m\n'
+    printf '  \033[33m█████╗  ███████║███████╗ ╚████╔╝ █████╗██████╔╝█████╗  █████╗\033[0m\n'
+    printf '  \033[33m██╔══╝  ██╔══██║╚════██║  ╚██╔╝  ╚════╝██╔══██╗██╔══╝  ██╔══╝\033[0m\n'
+    printf '  \033[33m███████╗██║  ██║███████║   ██║         ██████╔╝███████╗███████╗\033[0m\n'
+    printf '  \033[33m╚══════╝╚═╝  ╚═╝╚══════╝   ╚═╝         ╚═════╝ ╚══════╝╚══════╝\033[0m\n'
+    printf '  Hardware Audit LiveCD\n'
+    printf '\n'
+
+    for svc in $ALL; do
+        printf '  %s  %-20s%s\n' "$(svc_icon "$svc")" "$svc" "$(svc_detail "$svc")"
+    done
+    printf '\n'
+
+    # Network
+    ips="$(ip -4 addr show scope global 2>/dev/null | awk '/inet /{printf "  %-16s %s\n", $NF, $2}')"
+    if [ -n "$ips" ]; then
+        printf '  \033[1mNetwork:\033[0m\n'
+        printf '%s\n' "$ips"
+        printf '\n'
+    fi
+
+    if all_critical_done; then
+        printf '  \033[1;32mSystem ready.\033[0m  Audit is running in the background.\n'
+        first_ip="$(ip -4 addr show scope global 2>/dev/null | awk '/inet /{print $2}' | cut -d/ -f1 | head -1)"
+        if [ -n "$first_ip" ]; then
+            printf '  Web UI: \033[1mhttp://%s/\033[0m\n' "$first_ip"
+        fi
+        printf '\n'
+        sleep 3
+        break
+    fi
+
+    printf '  \033[90mStarting up...\033[0m\n'
+    sleep 3
+done
--- a/iso/overlay/usr/local/bin/bee-nvidia-load
+++ b/iso/overlay/usr/local/bin/bee-nvidia-load
@@ -50,11 +50,93 @@ load_module() {
        log "WARN: not found: $ko"
        return 1
    fi
-    if insmod "$ko" "$@"; then
+    if timeout 90 insmod "$ko" "$@"; then
        log "loaded: $mod $*"
        return 0
    fi
-    log "WARN: failed to load: $mod"
+    log "WARN: failed to load: $mod (exit $?)"
+    dmesg | tail -n 10 | sed 's/^/  dmesg: /' || true
+    return 1
+}
+
+nvidia_is_functional() {
+    grep -q ' nvidiactl$' /proc/devices 2>/dev/null
+}
+
+load_module_with_gsp_fallback() {
+    ko="$NVIDIA_KO_DIR/nvidia.ko"
+    if [ ! -f "$ko" ]; then
+        log "ERROR: not found: $ko"
+        return 1
+    fi
+
+    # Run insmod in background — on some converted SXM→PCIe cards GSP enters an
+    # infinite crash/reload loop and insmod never returns. We check for successful
+    # initialization by polling /proc/devices for nvidiactl instead of waiting for
+    # insmod to exit.
+    log "loading nvidia (GSP enabled, timeout 90s)"
+    insmod "$ko" &
+    _insmod_pid=$!
+
+    _waited=0
+    while [ $_waited -lt 90 ]; do
+        if nvidia_is_functional; then
+            log "loaded: nvidia (GSP enabled, ${_waited}s)"
+            echo "gsp-on" > /run/bee-nvidia-mode
+            return 0
+        fi
+        # Check if insmod exited with an error before timeout
+        if ! kill -0 "$_insmod_pid" 2>/dev/null; then
+            wait "$_insmod_pid"
+            _rc=$?
+            if [ $_rc -ne 0 ]; then
+                log "nvidia load failed (exit $_rc)"
+                dmesg | tail -n 10 | sed 's/^/  dmesg: /' || true
+                return 1
+            fi
+            # insmod exited 0 but nvidiactl not yet in /proc/devices — give it a moment
+            sleep 2
+            if nvidia_is_functional; then
+                log "loaded: nvidia (GSP enabled, ${_waited}s)"
+                return 0
+            fi
+            log "insmod exited 0 but nvidiactl missing — treating as failure"
+            return 1
+        fi
+        sleep 1
+        _waited=$((_waited + 1))
+    done
+
+    # GSP init timed out — kill the hanging insmod and attempt gsp-off fallback
+    log "nvidia GSP init timed out after 90s"
+    kill "$_insmod_pid" 2>/dev/null || true
+    wait "$_insmod_pid" 2>/dev/null || true
+
+    # Attempt to unload the partially-initialized module
+    if ! rmmod nvidia 2>/dev/null; then
+        # Module is stuck in the kernel — cannot reload with different params.
+        # User must reboot and select bee.nvidia.mode=gsp-off at boot menu.
+        log "ERROR: rmmod nvidia failed (EBUSY) — module stuck in kernel"
+        log "ERROR: reboot and select 'EASY-BEE (advanced) -> GSP=off' in boot menu"
+        echo "gsp-stuck" > /run/bee-nvidia-mode
+        return 1
+    fi
+
+    sleep 2
+    log "retrying with NVreg_EnableGpuFirmware=0"
+    log "WARNING: GSP disabled — power management will run via CPU path, not GPU firmware"
+
+    if insmod "$ko" NVreg_EnableGpuFirmware=0; then
+        if nvidia_is_functional; then
+            log "loaded: nvidia (GSP disabled)"
+            echo "gsp-off" > /run/bee-nvidia-mode
+            return 0
+        fi
+        log "insmod gsp-off exited 0 but nvidiactl missing"
+        return 1
+    fi
+
+    log "nvidia load failed (GSP=off)"
    dmesg | tail -n 10 | sed 's/^/  dmesg: /' || true
    return 1
 }
@@ -70,7 +152,7 @@ load_host_module() {

 case "$nvidia_mode" in
    normal|full)
-        if ! load_module nvidia; then
+        if ! load_module_with_gsp_fallback; then
            exit 1
        fi
        # nvidia-modeset on some server kernels needs ACPI video helper symbols
--- a/iso/overlay/usr/local/bin/bee-openbox-session
+++ b/iso/overlay/usr/local/bin/bee-openbox-session
@@ -7,16 +7,24 @@ xset s off
 xset -dpms
 xset s noblank

+# Set desktop background.
+if [ -f /usr/share/bee/wallpaper.png ]; then
+    feh --bg-fill /usr/share/bee/wallpaper.png
+else
+    xsetroot -solid '#f6c90e'
+fi
+
 tint2 &

-# Wait up to 120s for bee-web to bind. The web server starts immediately now
-# (audit is deferred), so this should succeed in a few seconds on most hardware.
-i=0
-while [ $i -lt 120 ]; do
-    if curl -sf http://localhost/healthz >/dev/null 2>&1; then break; fi
+# Wait up to 60s for bee-web before opening Chromium.
+# Without this Chromium gets connection-refused and shows a blank page.
+_i=0
+while [ $_i -lt 60 ]; do
+    curl -sf http://localhost/healthz >/dev/null 2>&1 && break
    sleep 1
-    i=$((i+1))
+    _i=$((_i+1))
 done
+unset _i

 chromium \
    --disable-infobars \
@@ -24,7 +32,8 @@ chromium \
    --no-first-run \
    --disable-session-crashed-bubble \
    --disable-features=TranslateUI \
+    --user-data-dir=/tmp/bee-chrome \
    --start-maximized \
-    http://localhost/ &
+    http://localhost/loading &

 exec openbox
Author	SHA1	Message	Date
Michael Chus	7a21c370e4	Handle NVIDIA GSP firmware init hang with timeout fallback - bee-nvidia-load: run insmod in background, poll /proc/devices for nvidiactl; if GSP init doesn't complete in 90s, kill insmod and retry with NVreg_EnableGpuFirmware=0. Handles EBUSY case with clear error. - Write /run/bee-nvidia-mode (gsp-on/gsp-off/gsp-stuck) for audit layer - Show GSP mode badge in sidebar: yellow for gsp-off, red for gsp-stuck - Report NvidiaGSPMode in RuntimeHealth with issue entries - Simplify GRUB menu: default (KMS+GSP), advanced submenu (GSP=off, nomodeset, fail-safe), remove load-to-RAM entry - Add pcmanfm, ristretto, mupdf, mousepad to desktop packages Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 21:00:43 +03:00
Michael Chus	a493e3ab5b	Fix service control buttons: sudo, real error output, UX feedback - services.go: use sudo systemctl so bee user can control system services - api.go: always return 200 with output field even on error, so the frontend shows the actual systemctl message instead of "exit status 1" - pages.go: button shows "..." while pending then restores label; output panel is full-width under the table with ✓/✗ status indicator; output auto-scrolls to bottom Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 20:25:41 +03:00
Michael Chus	19b4803ec7	Pass exact cycle duration to GPU stress instead of 86400s sentinel bee-gpu-burn now receives --seconds <LoadSec> so it exits naturally when the cycle ends, rather than relying solely on context cancellation to kill it. Process group kill (Setpgid+Cancel) is kept as a safety net for early cancellation (user stop, context timeout). Same fix for AMD RVS which now gets duration_ms = LoadSec * 1000. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 20:22:43 +03:00
Michael Chus	1bdfb1e9ca	Fix nvidia-targeted-stress failing with DCGM_ST_IN_USE (-34) nvvs (DCGM validation suite) survives when dcgmi is killed mid-run, leaving the GPU occupied. The next dcgmi diag invocation then fails with "affected resource is in use". Two-part fix: - Add nvvs and dcgmi to KillTestWorkers patterns so they are cleaned up by the global cancel handler - Call KillTestWorkers at the start of RunNvidiaTargetedStressValidatePack to clear any stale processes before dcgmi diag runs Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 20:21:36 +03:00
Michael Chus	c5d6b30177	Fix platform thermal cycling leaving GPU load running after test ends bee-gpu-burn is a shell script that spawns bee-gpu-burn-worker children. exec.CommandContext default cancel only kills the shell parent; the worker processes survive and keep loading the GPU indefinitely. Fix: set Setpgid=true and a custom Cancel that sends SIGKILL to the entire process group (-pid), same pattern already used in runSATCommandCtx. Applied to Nvidia, AMD, and CPU stress commands for consistency. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 20:19:20 +03:00
Michael Chus	5b9015451e	Add live task charts and fix USB export actions	2026-04-05 20:14:23 +03:00
Michael Chus	d1a6863ceb	Use amber fallback wallpaper color (#f6c90e) instead of black Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:30:41 +03:00
Michael Chus	f9aa05de8e	Add wallpaper: black background with amber EASY-BEE ASCII art logo - Add feh and python3-pil to package list - Add chroot hook that generates /usr/share/bee/wallpaper.png using PIL: black background, EASY-BEE box-drawing logo in amber (#f6c90e), "Hardware Audit LiveCD" subtitle in dim amber — matches motd exactly - bee-openbox-session: set wallpaper with feh --bg-fill, fall back to xsetroot -solid black if wallpaper not found Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:29:42 +03:00
Michael Chus	a9ccea8cca	Fix black desktop and Chromium blank page on startup - Set xsetroot solid background (#12100a, dark amber) so openbox doesn't show bare black before Chromium opens - Re-add healthz wait loop before launching Chromium: without it Chromium opens localhost/loading before bee-web is up and gets connection-refused which renders as a blank white page Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:25:32 +03:00
Michael Chus	fc5c985fb5	Reset tty1 properly when bee-boot-status exits Add TTYReset=yes and TTYVHangup=yes so systemd restores the terminal to a clean state before handing tty1 to getty. Without this the screen went black with no cursor after the status display finished. Also remove DefaultDependencies=no which was too aggressive. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:22:01 +03:00
Michael Chus	5eb3baddb4	Fix bee-boot-status blank screen caused by variable buffering Command substitution in sh strips trailing newlines, so accumulating output in a variable via $(...) lost all line breaks. Reverted to direct printf calls which work correctly. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:21:10 +03:00
Michael Chus	a6ac13b5d3	Improve bee-boot-status: slower refresh, more detail - Refresh every 3s instead of 1s to reduce flicker - Show ssh, bee-sshsetup in service list - Show failure reason for failed services - Show last journal line for activating services - Show IP addresses and web UI URL when network is up - Render frame to variable before printing to reduce flicker Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:20:07 +03:00
Michael Chus	4003cb7676	Lower kernel console loglevel to 3 to reduce boot noise loglevel=6 floods the screen with mpt3sas/scsi/sd informational messages, hiding systemd service status and bee-boot-status display. loglevel=3 shows only kernel errors; all messages still go to serial. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 19:19:09 +03:00
Michael Chus	2875313ba0	Improve boot UX: status display, faster GUI, loading spinner - Add bee-boot-status service: shows live service status on tty1 with ASCII logo before getty, exits when all bee services settle - Remove lightdm dependency on bee-preflight so GUI starts immediately without waiting for NVIDIA driver load - Replace Chromium blank-page problem with /loading spinner page that polls /api/services and auto-redirects when services are ready; add "Open app now" override button; use fresh --user-data-dir=/tmp/bee-chrome - Unify branding: add "Hardware Audit LiveCD" subtitle to GRUB menu, bee-boot-status (with yellow ASCII logo), and web spinner Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-05 18:58:24 +03:00
Michael Chus	f1621efee4	Mirror task lifecycle to serial console	2026-04-05 18:34:06 +03:00
Michael Chus	4461249cc3	Make memory stress size follow available RAM	2026-04-05 18:33:26 +03:00