Add infrastructure reliability monitoring

README.md

@@ -32,7 +32,7 @@ ssh root@192.168.0.10
 
 ```bash
 apt-get update
-apt-get install -y curl ca-certificates tar smartmontools
+apt-get install -y curl ca-certificates tar smartmontools iproute2 iputils-ping
 
 cd /tmp
 curl -fL \
@@ -120,6 +120,15 @@ ss -lntp | grep 9105
 
 Панель показывает:
 
+- изменение критичных SMART-счётчиков, результаты self-test и исчезновение дисков;
+- автоматический короткий SMART self-test раз в неделю для дисков без свежего теста;
+- datasets, vdev-классы, ошибки read/write/checksum, scrub и ARC ZFS;
+- доступность шлюза, DNS и интернета, packet loss, latency и внешний IP;
+- тип, состояние, скорость, duplex и master сетевых интерфейсов;
+- состояние связи с NUT/apcupsd, возраст батареи и расчётную автономность UPS;
+- настраиваемый сценарий корректного выключения VM/LXC и Proxmox с безопасным dry-run;
+- самоконтроль сборщика, записи истории, свободного места базы и systemd watchdog;
+
 - модель и текущую загрузку CPU;
 - количество сокетов, физических ядер и логических потоков;
 - среднюю текущую частоту;

alerts.go

@@ -75,6 +75,15 @@ func evaluateAlerts(metrics dashboardMetrics, thresholds AlertThresholds) []Aler
 				add("disk-temp-"+disk.Name, "warning", "Диски", disk.Model+": высокая температура", fmt.Sprintf("Температура %.1f °C.", *disk.Temperature))
 			}
 		}
+		if len(disk.CounterChanges) > 0 {
+			add("disk-growth-"+disk.Name, "critical", "Диски", disk.Model+": ухудшились SMART-счётчики", strings.Join(disk.CounterChanges, " · "))
+		}
+		if disk.SelfTestNeverRun {
+			add("disk-selftest-"+disk.Name, "warning", "Диски", disk.Model+": нет завершённого SMART self-test", "Короткий тест будет автоматически запущен; проверьте его результат после завершения.")
+		}
+	}
+	for _, missing := range metrics.MissingDisks {
+		add("disk-missing-"+missing, "critical", "Диски", "Физический диск исчез", missing+" больше не обнаруживается в /sys/block.")
 	}
 
 	for _, pool := range metrics.ZFS.Pools {
@@ -89,6 +98,25 @@ func evaluateAlerts(metrics dashboardMetrics, thresholds AlertThresholds) []Aler
 		if strings.Contains(strings.ToLower(pool.Scan), "in progress") {
 			add("zfs-scan-"+pool.Name, "info", "ZFS", pool.Name+": выполняется scrub/resilver", pool.Scan)
 		}
+		if pool.ReadErrors+pool.WriteErrors+pool.ChecksumErrors > 0 {
+			add("zfs-vdev-errors-"+pool.Name, "critical", "ZFS", pool.Name+": ошибки чтения/записи/checksum", fmt.Sprintf("read=%d, write=%d, checksum=%d.", pool.ReadErrors, pool.WriteErrors, pool.ChecksumErrors))
+		}
+		if len(pool.ErrorChanges) > 0 {
+			add("zfs-new-errors-"+pool.Name, "critical", "ZFS", pool.Name+": появились новые ошибки", strings.Join(pool.ErrorChanges, " · "))
+		}
+		if pool.ScrubTooOld {
+			add("zfs-scrub-old-"+pool.Name, "warning", "ZFS", pool.Name+": scrub просрочен", "Scrub не выполнялся более 35 дней либо ни разу не запускался.")
+		}
+	}
+	for _, dataset := range metrics.ZFS.Datasets {
+		if dataset.CapacityPercent >= 90 {
+			add("zfs-dataset-"+dataset.Name, "critical", "ZFS", dataset.Name+": dataset почти заполнен", fmt.Sprintf("Заполнение %.1f%%, доступно %.1f GB.", dataset.CapacityPercent, float64(dataset.AvailableBytes)/1073741824))
+		} else if dataset.CapacityPercent >= 80 {
+			add("zfs-dataset-"+dataset.Name, "warning", "ZFS", dataset.Name+": мало свободного места", fmt.Sprintf("Заполнение %.1f%%.", dataset.CapacityPercent))
+		}
+	}
+	if metrics.ZFS.ARC.Available && metrics.ZFS.ARC.Hits+metrics.ZFS.ARC.Misses > 10000 && metrics.ZFS.ARC.HitRatio < 70 {
+		add("zfs-arc-hit", "warning", "ZFS", "Низкий ARC hit ratio", fmt.Sprintf("Попадания ARC: %.1f%%.", metrics.ZFS.ARC.HitRatio))
 	}
 
 	if metrics.UPS.Available {
@@ -101,6 +129,14 @@ func evaluateAlerts(metrics dashboardMetrics, thresholds AlertThresholds) []Aler
 		if metrics.UPS.ChargePercent != nil && *metrics.UPS.ChargePercent < thresholds.UPSChargeCritical {
 			add("ups-charge", "critical", "UPS", "Низкий заряд батареи", fmt.Sprintf("Осталось %.1f%%.", *metrics.UPS.ChargePercent))
 		}
+		if metrics.UPS.ReplaceBattery {
+			add("ups-replace-battery", "critical", "UPS", "UPS требует заменить батарею", "NUT/APC сообщает флаг Replace Battery.")
+		}
+		if metrics.UPS.BatteryAgeDays > 3*365 {
+			add("ups-battery-age", "warning", "UPS", "Батарея UPS старше трёх лет", fmt.Sprintf("Возраст батареи примерно %d дней.", metrics.UPS.BatteryAgeDays))
+		}
+	} else if metrics.UPS.Source != "" {
+		add("ups-communication", "warning", "UPS", "Нет связи с UPS", "NUT и apcupsd не вернули данные. Проверьте USB/сетевое подключение и службу UPS.")
 	}
 
 	if metrics.Backups.Available && metrics.Guests.Available {
@@ -164,6 +200,44 @@ func evaluateAlerts(metrics dashboardMetrics, thresholds AlertThresholds) []Aler
 	if networkErrors > 0 {
 		add("network-errors", "warning", "Сеть", "Сетевые ошибки и потери пакетов", fmt.Sprintf("Суммарный счётчик интерфейсов: %d.", networkErrors))
 	}
+	if !metrics.Network.Health.CheckedAt.IsZero() && !metrics.Network.Health.GatewayUp {
+		add("network-gateway", "critical", "Сеть", "Недоступен локальный шлюз", firstNonEmpty(metrics.Network.Health.Gateway, "Default gateway не определён."))
+	}
+	if !metrics.Network.Health.CheckedAt.IsZero() && !metrics.Network.Health.DNSUp {
+		add("network-dns", "critical", "Сеть", "Не работает DNS", "Не удалось разрешить deb.debian.org.")
+	}
+	if !metrics.Network.Health.CheckedAt.IsZero() && !metrics.Network.Health.InternetUp {
+		add("network-internet", "warning", "Сеть", "Нет доступа в интернет", "Внешняя проверка IP не выполнена.")
+	}
+	if metrics.Network.Health.PacketLoss > 0 {
+		add("network-loss", "warning", "Сеть", "Потери пакетов до шлюза", fmt.Sprintf("Потери %.1f%%, задержка %.1f ms.", metrics.Network.Health.PacketLoss, metrics.Network.Health.LatencyMs))
+	}
+	if metrics.Network.Health.ExternalIPChanged {
+		add("network-public-ip", "info", "Сеть", "Изменился внешний IP", "Новый адрес: "+metrics.Network.Health.ExternalIP)
+	}
+	for _, name := range metrics.Network.Health.DisappearedInterfaces {
+		add("network-interface-missing-"+name, "warning", "Сеть", "Исчез сетевой интерфейс", name+" больше не присутствует в /proc/net/dev.")
+	}
+	for _, iface := range metrics.Network.Interfaces {
+		if iface.Kind == "Physical" && !iface.Carrier && (iface.Master != "" || iface.ReceiveBytes+iface.TransmitBytes > 0) {
+			add("network-link-"+iface.Name, "warning", "Сеть", iface.Name+": нет carrier", "Физический линк не поднят.")
+		}
+		if (iface.Kind == "Bridge" || iface.Kind == "Bond" || iface.Kind == "VLAN") && iface.State == "down" {
+			add("network-state-"+iface.Name, "warning", "Сеть", iface.Name+": интерфейс выключен", iface.Kind+" находится в состоянии down.")
+		}
+	}
+	if !metrics.Monitor.LastSuccessfulCollection.IsZero() && time.Since(metrics.Monitor.LastSuccessfulCollection) > 30*time.Second {
+		add("monitor-stale", "critical", "Dashboard", "Данные мониторинга устарели", "Последний успешный сбор был "+time.Since(metrics.Monitor.LastSuccessfulCollection).Round(time.Second).String()+" назад.")
+	}
+	if metrics.Monitor.SlowCollector {
+		add("monitor-slow", "warning", "Dashboard", "Сбор метрик выполняется слишком долго", fmt.Sprintf("Последний цикл занял %d ms.", metrics.Monitor.CollectionDurationMs))
+	}
+	if metrics.Monitor.HistoryWriteError != "" {
+		add("monitor-history", "critical", "Dashboard", "Ошибка записи истории", metrics.Monitor.HistoryWriteError)
+	}
+	if metrics.Monitor.DatabaseFreeBytes > 0 && metrics.Monitor.DatabaseFreeBytes < 1024*1024*1024 {
+		add("monitor-db-space", "warning", "Dashboard", "Мало места для базы истории", fmt.Sprintf("На разделе базы свободно %.1f GB.", float64(metrics.Monitor.DatabaseFreeBytes)/1073741824))
+	}
 	oldSnapshots := 0
 	for _, snapshot := range metrics.Activity.Snapshots {
 		if snapshot.AgeSeconds > 30*24*3600 {

disks.go

@@ -28,12 +28,25 @@ type DiskMetrics struct {
 	Temperature      *float64 `json:"temperatureCelsius"`
 	PowerOnHours     *uint64  `json:"powerOnHours"`
 	WearUsedPercent  *float64 `json:"wearUsedPercent"`
+	Reallocated      uint64   `json:"reallocated"`
+	Pending          uint64   `json:"pending"`
+	Uncorrectable    uint64   `json:"uncorrectable"`
+	CRCErrors        uint64   `json:"crcErrors"`
+	MediaErrors      uint64   `json:"mediaErrors"`
+	CounterChanges   []string `json:"counterChanges"`
+	LastSelfTest     string   `json:"lastSelfTest"`
+	LastSelfTestAt   string   `json:"lastSelfTestAt"`
+	SelfTestNeverRun bool     `json:"selfTestNeverRun"`
+	ShortTestStarted bool     `json:"shortTestStarted"`
 }
 
 type diskCollector struct {
 	mu        sync.Mutex
 	cached    []DiskMetrics
 	updatedAt time.Time
+	previous  map[string]DiskMetrics
+	missing   []string
+	lastTests map[string]time.Time
 }
 
 func (c *diskCollector) collect() []DiskMetrics {
@@ -42,11 +55,73 @@ func (c *diskCollector) collect() []DiskMetrics {
 	if time.Since(c.updatedAt) < time.Minute && c.cached != nil {
 		return c.cached
 	}
-	c.cached = readPhysicalDisks()
+	current := readPhysicalDisks()
+	if c.previous == nil {
+		c.previous = map[string]DiskMetrics{}
+		c.lastTests = map[string]time.Time{}
+	}
+	seen := map[string]bool{}
+	c.missing = nil
+	for i := range current {
+		disk := &current[i]
+		key := disk.Serial
+		if key == "" || key == "Не указан" {
+			key = disk.Path
+		}
+		seen[key] = true
+		if old, ok := c.previous[key]; ok {
+			disk.CounterChanges = diskCounterChanges(old, *disk)
+			disk.AttentionReasons = append(disk.AttentionReasons, disk.CounterChanges...)
+		}
+		if disk.SMARTAvailable && (disk.SelfTestNeverRun || selfTestOlderThanWeek(disk.LastSelfTestAt)) && time.Since(c.lastTests[key]) >= 7*24*time.Hour {
+			disk.ShortTestStarted = startShortSMARTTest(disk.Path)
+			if disk.ShortTestStarted {
+				c.lastTests[key] = time.Now()
+			}
+		}
+		c.previous[key] = *disk
+	}
+	for key, disk := range c.previous {
+		if !seen[key] {
+			c.missing = append(c.missing, disk.Model+" ("+disk.Serial+")")
+		}
+	}
+	c.cached = current
 	c.updatedAt = time.Now()
 	return c.cached
 }
 
+func (c *diskCollector) missingDisks() []string {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	return append([]string(nil), c.missing...)
+}
+
+func diskCounterChanges(old, current DiskMetrics) []string {
+	checks := []struct {
+		label    string
+		old, now uint64
+	}{{"Reallocated вырос", old.Reallocated, current.Reallocated}, {"Pending вырос", old.Pending, current.Pending}, {"Uncorrectable вырос", old.Uncorrectable, current.Uncorrectable}, {"CRC-ошибки выросли", old.CRCErrors, current.CRCErrors}, {"Media errors выросли", old.MediaErrors, current.MediaErrors}}
+	var changes []string
+	for _, check := range checks {
+		if check.now > check.old {
+			changes = append(changes, check.label+": "+strconv.FormatUint(check.old, 10)+" → "+strconv.FormatUint(check.now, 10))
+		}
+	}
+	return changes
+}
+
+func selfTestOlderThanWeek(value string) bool {
+	when, err := time.Parse(time.RFC3339, value)
+	return err == nil && time.Since(when) > 7*24*time.Hour
+}
+
+func startShortSMARTTest(path string) bool {
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+	return exec.CommandContext(ctx, "smartctl", "-t", "short", path).Run() == nil
+}
+
 func readPhysicalDisks() []DiskMetrics {
 	entries, _ := os.ReadDir("/sys/block")
 	usage := readDiskUsage()
@@ -202,6 +277,16 @@ type smartctlJSON struct {
 			Count int `json:"count"`
 		} `json:"summary"`
 	} `json:"ata_smart_error_log"`
+	ATASelfTestLog struct {
+		Standard struct {
+			Table []struct {
+				Status struct {
+					String string `json:"string"`
+				} `json:"status"`
+				LifetimeHours uint64 `json:"lifetime_hours"`
+			} `json:"table"`
+		} `json:"standard"`
+	} `json:"ata_smart_self_test_log"`
 	NVMeHealth *struct {
 		Temperature     float64 `json:"temperature"`
 		PercentageUsed  float64 `json:"percentage_used"`
@@ -254,6 +339,7 @@ func applySMART(disk *DiskMetrics) {
 	}
 	if smart.NVMeHealth != nil {
 		disk.WearUsedPercent = &smart.NVMeHealth.PercentageUsed
+		disk.MediaErrors = smart.NVMeHealth.MediaErrors
 		if smart.NVMeHealth.CriticalWarning != 0 {
 			disk.AttentionReasons = append(disk.AttentionReasons, "NVMe сообщает критическое предупреждение")
 		}
@@ -261,6 +347,17 @@ func applySMART(disk *DiskMetrics) {
 			disk.AttentionReasons = append(disk.AttentionReasons, "Ошибки целостности носителя: "+strconv.FormatUint(smart.NVMeHealth.MediaErrors, 10))
 		}
 	}
+	if len(smart.ATASelfTestLog.Standard.Table) > 0 {
+		last := smart.ATASelfTestLog.Standard.Table[0]
+		disk.LastSelfTest = firstNonEmpty(last.Status.String, "Результат неизвестен")
+		if disk.PowerOnHours != nil && *disk.PowerOnHours >= last.LifetimeHours {
+			disk.LastSelfTestAt = time.Now().Add(-time.Duration(*disk.PowerOnHours-last.LifetimeHours) * time.Hour).Format(time.RFC3339)
+		}
+	} else {
+		disk.SelfTestNeverRun = true
+		disk.LastSelfTest = "Никогда не запускался"
+		disk.AttentionReasons = append(disk.AttentionReasons, "SMART self-test ещё ни разу не запускался")
+	}
 	applyATAAttributes(disk, smart)
 	applySMARTExitStatus(disk, smart.Smartctl.ExitStatus)
 	if len(disk.AttentionReasons) > 0 && disk.SMARTStatus == "Исправен" {
@@ -275,6 +372,16 @@ func applyATAAttributes(disk *DiskMetrics, smart smartctlJSON) {
 		"Wear_Leveling_Count": true,
 	}
 	for _, attribute := range smart.ATAAttributes.Table {
+		switch attribute.Name {
+		case "Reallocated_Sector_Ct", "Reallocated_Event_Count":
+			disk.Reallocated += uint64(max(attribute.Raw.Value, 0))
+		case "Current_Pending_Sector":
+			disk.Pending = uint64(max(attribute.Raw.Value, 0))
+		case "Offline_Uncorrectable", "Reported_Uncorrect":
+			disk.Uncorrectable += uint64(max(attribute.Raw.Value, 0))
+		case "UDMA_CRC_Error_Count", "Interface_CRC_Error_Count":
+			disk.CRCErrors += uint64(max(attribute.Raw.Value, 0))
+		}
 		if wearNames[attribute.Name] && attribute.Value >= 0 && attribute.Value <= 100 {
 			wear := float64(100 - attribute.Value)
 			disk.WearUsedPercent = &wear

main.go

@@ -50,6 +50,7 @@ func main() {
 		Handler:           routes(collector, store),
 		ReadHeaderTimeout: 5 * time.Second,
 	}
+	runSystemdWatchdog()
 
 	go func() {
 		log.Printf("Proxmox Dashboard %s запущен на http://0.0.0.0%s", version, address)

network.go

@@ -2,7 +2,13 @@ package main
 
 import (
 	"bufio"
+	"context"
+	"io"
+	"net"
+	"net/http"
 	"os"
+	"os/exec"
+	"path/filepath"
 	"strconv"
 	"strings"
 	"sync"
@@ -16,6 +22,20 @@ type NetworkMetrics struct {
 	ReceivePerSec  float64            `json:"receivePerSec"`
 	TransmitPerSec float64            `json:"transmitPerSec"`
 	Interfaces     []NetworkInterface `json:"interfaces"`
+	Health         NetworkHealth      `json:"health"`
+}
+
+type NetworkHealth struct {
+	Gateway               string    `json:"gateway"`
+	GatewayUp             bool      `json:"gatewayUp"`
+	DNSUp                 bool      `json:"dnsUp"`
+	InternetUp            bool      `json:"internetUp"`
+	PacketLoss            float64   `json:"packetLoss"`
+	LatencyMs             float64   `json:"latencyMs"`
+	ExternalIP            string    `json:"externalIp"`
+	ExternalIPChanged     bool      `json:"externalIpChanged"`
+	DisappearedInterfaces []string  `json:"disappearedInterfaces"`
+	CheckedAt             time.Time `json:"checkedAt"`
 }
 
 type NetworkInterface struct {
@@ -31,6 +51,12 @@ type NetworkInterface struct {
 	VMID           int     `json:"vmid,omitempty"`
 	GuestName      string  `json:"guestName,omitempty"`
 	GuestType      string  `json:"guestType,omitempty"`
+	Kind           string  `json:"kind"`
+	State          string  `json:"state"`
+	Carrier        bool    `json:"carrier"`
+	SpeedMbps      int     `json:"speedMbps,omitempty"`
+	Duplex         string  `json:"duplex,omitempty"`
+	Master         string  `json:"master,omitempty"`
 }
 
 type networkSnapshot struct {
@@ -44,6 +70,10 @@ type networkCollector struct {
 	path               string
 	previous           networkSnapshot
 	previousInterfaces map[string]NetworkInterface
+	health             NetworkHealth
+	healthUpdatedAt    time.Time
+	externalIP         string
+	knownInterfaces    map[string]bool
 }
 
 func newNetworkCollector(path string) (*networkCollector, error) {
@@ -56,7 +86,11 @@ func newNetworkCollector(path string) (*networkCollector, error) {
 	for _, v := range details {
 		previous[v.Name] = v
 	}
-	return &networkCollector{path: path, previous: snapshot, previousInterfaces: previous}, nil
+	known := map[string]bool{}
+	for _, v := range details {
+		known[v.Name] = true
+	}
+	return &networkCollector{path: path, previous: snapshot, previousInterfaces: previous, knownInterfaces: known}, nil
 }
 
 func (c *networkCollector) collect() (NetworkMetrics, error) {
@@ -75,6 +109,7 @@ func (c *networkCollector) collect() (NetworkMetrics, error) {
 	}
 	for i := range details {
 		v := &details[i]
+		decorateInterface(v)
 		if old, ok := c.previousInterfaces[v.Name]; ok && elapsed > 0 && v.ReceiveBytes >= old.ReceiveBytes && v.TransmitBytes >= old.TransmitBytes {
 			v.ReceivePerSec = float64(v.ReceiveBytes-old.ReceiveBytes) / elapsed
 			v.TransmitPerSec = float64(v.TransmitBytes-old.TransmitBytes) / elapsed
@@ -85,12 +120,100 @@ func (c *networkCollector) collect() (NetworkMetrics, error) {
 		c.previousInterfaces[v.Name] = v
 	}
 	c.previous = current
+	if time.Since(c.healthUpdatedAt) > 30*time.Second {
+		c.health = checkNetworkHealth(c.externalIP)
+		if c.health.ExternalIP != "" {
+			c.externalIP = c.health.ExternalIP
+		}
+		c.healthUpdatedAt = time.Now()
+	}
+	currentNames := map[string]bool{}
+	for _, v := range details {
+		currentNames[v.Name] = true
+	}
+	c.health.DisappearedInterfaces = nil
+	for name := range c.knownInterfaces {
+		if !currentNames[name] {
+			c.health.DisappearedInterfaces = append(c.health.DisappearedInterfaces, name)
+		}
+	}
+	c.knownInterfaces = currentNames
 	return NetworkMetrics{
 		Interface: strings.Join(interfaceNames, ", "), Interfaces: details, ReceiveBytes: current.received,
-		TransmitBytes: current.sent, ReceivePerSec: receivedRate, TransmitPerSec: sentRate,
+		TransmitBytes: current.sent, ReceivePerSec: receivedRate, TransmitPerSec: sentRate, Health: c.health,
 	}, nil
 }
 
+func decorateInterface(v *NetworkInterface) {
+	base := filepath.Join("/sys/class/net", v.Name)
+	v.State = readTrimmed(filepath.Join(base, "operstate"))
+	v.Carrier = readTrimmed(filepath.Join(base, "carrier")) == "1"
+	v.SpeedMbps, _ = strconv.Atoi(readTrimmed(filepath.Join(base, "speed")))
+	v.Duplex = readTrimmed(filepath.Join(base, "duplex"))
+	if target, err := filepath.EvalSymlinks(filepath.Join(base, "master")); err == nil {
+		v.Master = filepath.Base(target)
+	}
+	switch {
+	case strings.HasPrefix(v.Name, "vmbr"):
+		v.Kind = "Bridge"
+	case strings.HasPrefix(v.Name, "bond"):
+		v.Kind = "Bond"
+	case strings.Contains(v.Name, "."):
+		v.Kind = "VLAN"
+	case strings.HasPrefix(v.Name, "tap") || strings.HasPrefix(v.Name, "veth"):
+		v.Kind = "Guest"
+	default:
+		v.Kind = "Physical"
+	}
+}
+
+func checkNetworkHealth(previousIP string) NetworkHealth {
+	result := NetworkHealth{CheckedAt: time.Now()}
+	ctx, cancel := context.WithTimeout(context.Background(), 6*time.Second)
+	defer cancel()
+	if out, err := exec.CommandContext(ctx, "ip", "route", "show", "default").Output(); err == nil {
+		f := strings.Fields(string(out))
+		for i := range f {
+			if f[i] == "via" && i+1 < len(f) {
+				result.Gateway = f[i+1]
+				break
+			}
+		}
+	}
+	if result.Gateway != "" {
+		out, _ := exec.CommandContext(ctx, "ping", "-c", "3", "-W", "1", result.Gateway).CombinedOutput()
+		text := string(out)
+		result.GatewayUp = strings.Contains(text, "0% packet loss")
+		if i := strings.Index(text, "packet loss"); i > 0 {
+			start := strings.LastIndex(text[:i], " ")
+			if start >= 0 {
+				result.PacketLoss, _ = strconv.ParseFloat(strings.TrimSuffix(strings.TrimSpace(text[start:i]), "%"), 64)
+			}
+		}
+		if i := strings.Index(text, "min/avg/max"); i >= 0 {
+			if eq := strings.Index(text[i:], "="); eq >= 0 {
+				parts := strings.Split(strings.Fields(text[i+eq+1:])[0], "/")
+				if len(parts) > 1 {
+					result.LatencyMs, _ = strconv.ParseFloat(parts[1], 64)
+				}
+			}
+		}
+	}
+	resolver := net.Resolver{}
+	_, err := resolver.LookupHost(ctx, "deb.debian.org")
+	result.DNSUp = err == nil
+	client := http.Client{Timeout: 4 * time.Second}
+	request, _ := http.NewRequestWithContext(ctx, http.MethodGet, "https://api.ipify.org", nil)
+	if response, err := client.Do(request); err == nil {
+		defer response.Body.Close()
+		body, _ := io.ReadAll(io.LimitReader(response.Body, 128))
+		result.ExternalIP = strings.TrimSpace(string(body))
+		result.InternetUp = response.StatusCode == 200 && net.ParseIP(result.ExternalIP) != nil
+	}
+	result.ExternalIPChanged = previousIP != "" && result.ExternalIP != "" && previousIP != result.ExternalIP
+	return result
+}
+
 func readNetworkSnapshot(path string) (networkSnapshot, []string, []NetworkInterface, error) {
 	file, err := os.Open(path)
 	if err != nil {

proxmox-cpu-dashboard.service

@@ -4,7 +4,7 @@ After=network-online.target
 Wants=network-online.target
 
 [Service]
-Type=simple
+Type=notify
 ExecStart=/usr/local/bin/proxmox-cpu-dashboard
 Environment=DASHBOARD_ADDR=:9105
 Environment=DASHBOARD_DB_PATH=/var/lib/proxmox-dashboard/dashboard.db
@@ -12,6 +12,8 @@ StateDirectory=proxmox-dashboard
 StateDirectoryMode=0750
 Restart=on-failure
 RestartSec=3
+WatchdogSec=30
+TimeoutStartSec=20
 NoNewPrivileges=true
 ProtectHome=true
 PrivateTmp=true

reliability.go

@@ -0,0 +1,226 @@
+package main
+
+import (
+	"context"
+	"fmt"
+	"log"
+	"net"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"sort"
+	"strings"
+	"sync"
+	"syscall"
+	"time"
+)
+
+func notifySystemd(message string) bool {
+	address := os.Getenv("NOTIFY_SOCKET")
+	if address == "" {
+		return false
+	}
+	if strings.HasPrefix(address, "@") {
+		address = "\x00" + address[1:]
+	}
+	connection, err := net.DialUnix("unixgram", nil, &net.UnixAddr{Name: address, Net: "unixgram"})
+	if err != nil {
+		return false
+	}
+	defer connection.Close()
+	_, err = connection.Write([]byte(message))
+	return err == nil
+}
+func runSystemdWatchdog() {
+	if !notifySystemd("READY=1\nSTATUS=Dashboard collects Proxmox metrics") {
+		return
+	}
+	go func() {
+		ticker := time.NewTicker(10 * time.Second)
+		defer ticker.Stop()
+		for range ticker.C {
+			notifySystemd("WATCHDOG=1")
+		}
+	}()
+}
+
+type MonitorHealth struct {
+	LastSuccessfulCollection time.Time `json:"lastSuccessfulCollection"`
+	LastCollectionError      string    `json:"lastCollectionError"`
+	CollectionDurationMs     int64     `json:"collectionDurationMs"`
+	SlowCollector            bool      `json:"slowCollector"`
+	LastHistoryWrite         time.Time `json:"lastHistoryWrite"`
+	HistoryWriteError        string    `json:"historyWriteError"`
+	DatabaseFreeBytes        uint64    `json:"databaseFreeBytes"`
+	DatabasePath             string    `json:"databasePath"`
+	WatchdogEnabled          bool      `json:"watchdogEnabled"`
+}
+
+type monitorTracker struct {
+	mu          sync.RWMutex
+	lastSuccess time.Time
+	lastError   string
+	duration    time.Duration
+}
+
+func (m *monitorTracker) record(start time.Time, err error) {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	m.duration = time.Since(start)
+	if err != nil {
+		m.lastError = err.Error()
+		return
+	}
+	m.lastSuccess = time.Now()
+	m.lastError = ""
+}
+
+func (m *monitorTracker) snapshot(store *Store) MonitorHealth {
+	m.mu.RLock()
+	result := MonitorHealth{LastSuccessfulCollection: m.lastSuccess, LastCollectionError: m.lastError, CollectionDurationMs: m.duration.Milliseconds(), SlowCollector: m.duration > 10*time.Second, DatabasePath: store.path, WatchdogEnabled: os.Getenv("NOTIFY_SOCKET") != ""}
+	m.mu.RUnlock()
+	result.LastHistoryWrite, result.HistoryWriteError = store.historyHealth()
+	path := store.path
+	if path == ":memory:" {
+		path = "/tmp"
+	} else {
+		path = filepath.Dir(path)
+	}
+	var stat syscall.Statfs_t
+	if syscall.Statfs(path, &stat) == nil {
+		result.DatabaseFreeBytes = stat.Bavail * uint64(stat.Bsize)
+	}
+	return result
+}
+
+type UPSShutdownSettings struct {
+	Enabled        bool    `json:"enabled"`
+	ChargePercent  float64 `json:"chargePercent"`
+	RuntimeSeconds float64 `json:"runtimeSeconds"`
+	GraceSeconds   int     `json:"graceSeconds"`
+	GuestOrder     []int   `json:"guestOrder"`
+}
+
+func defaultUPSShutdownSettings() UPSShutdownSettings {
+	return UPSShutdownSettings{ChargePercent: 15, RuntimeSeconds: 600, GraceSeconds: 120}
+}
+func (s *Store) UPSShutdownSettings() UPSShutdownSettings {
+	value := defaultUPSShutdownSettings()
+	s.loadJSONSetting("ups_shutdown", &value)
+	return value
+}
+func (s *Store) SaveUPSShutdownSettings(value UPSShutdownSettings) error {
+	if value.ChargePercent < 1 || value.ChargePercent > 80 || value.RuntimeSeconds < 60 || value.RuntimeSeconds > 7200 || value.GraceSeconds < 30 || value.GraceSeconds > 1800 {
+		return fmt.Errorf("проверьте пороги UPS и задержку")
+	}
+	return s.saveJSONSetting("ups_shutdown", value)
+}
+
+type ShutdownStep struct {
+	Order  int    `json:"order"`
+	Target string `json:"target"`
+	Action string `json:"action"`
+	Detail string `json:"detail"`
+}
+
+func buildShutdownDryRun(settings UPSShutdownSettings, guests GuestsMetrics) []ShutdownStep {
+	byID := map[int]GuestMetrics{}
+	for _, g := range guests.Guests {
+		if g.Template == 0 && g.Status == "running" {
+			byID[g.VMID] = g
+		}
+	}
+	var ordered []GuestMetrics
+	for _, id := range settings.GuestOrder {
+		if g, ok := byID[id]; ok {
+			ordered = append(ordered, g)
+			delete(byID, id)
+		}
+	}
+	var rest []GuestMetrics
+	for _, g := range byID {
+		rest = append(rest, g)
+	}
+	sort.Slice(rest, func(i, j int) bool { return rest[i].VMID < rest[j].VMID })
+	ordered = append(ordered, rest...)
+	steps := []ShutdownStep{{Order: 1, Target: "UPS", Action: "Ожидание порога", Detail: fmt.Sprintf("заряд ≤ %.0f%% или автономность ≤ %.0f сек.", settings.ChargePercent, settings.RuntimeSeconds)}}
+	for _, g := range ordered {
+		kind := "LXC"
+		if g.Type == "qemu" {
+			kind = "VM"
+		}
+		steps = append(steps, ShutdownStep{Order: len(steps) + 1, Target: fmt.Sprintf("%s %d · %s", kind, g.VMID, g.Name), Action: "Корректное завершение", Detail: fmt.Sprintf("ожидание до %d сек.", settings.GraceSeconds)})
+	}
+	steps = append(steps, ShutdownStep{Order: len(steps) + 1, Target: "Proxmox host", Action: "Завершение работы", Detail: "только после остановки VM/LXC"})
+	return steps
+}
+
+func runUPSShutdownController(store *Store, ups *upsCollector, guests *guestCollector) {
+	ticker := time.NewTicker(10 * time.Second)
+	defer ticker.Stop()
+	triggered := false
+	for range ticker.C {
+		settings := store.UPSShutdownSettings()
+		metrics := ups.collect()
+		onBattery := strings.Contains(metrics.Status, "OB") || metrics.Status == "ONBATT"
+		if !settings.Enabled || !metrics.Available || !onBattery {
+			if metrics.Available && !onBattery {
+				triggered = false
+			}
+			continue
+		}
+		lowCharge := metrics.ChargePercent != nil && *metrics.ChargePercent <= settings.ChargePercent
+		lowRuntime := metrics.RuntimeSeconds != nil && *metrics.RuntimeSeconds <= settings.RuntimeSeconds
+		if triggered || (!lowCharge && !lowRuntime) {
+			continue
+		}
+		triggered = true
+		log.Printf("UPS достиг порога безопасного выключения; запускаем остановку гостей")
+		go executeUPSShutdown(settings, guests.collect())
+	}
+}
+
+func executeUPSShutdown(settings UPSShutdownSettings, guests GuestsMetrics) {
+	steps := buildShutdownDryRun(settings, guests)
+	_ = steps
+	byID := map[int]GuestMetrics{}
+	for _, g := range guests.Guests {
+		if g.Template == 0 && g.Status == "running" {
+			byID[g.VMID] = g
+		}
+	}
+	var order []int
+	for _, id := range settings.GuestOrder {
+		if _, ok := byID[id]; ok {
+			order = append(order, id)
+			delete(byID, id)
+		}
+	}
+	var remaining []int
+	for id := range byID {
+		remaining = append(remaining, id)
+	}
+	sort.Ints(remaining)
+	order = append(order, remaining...)
+	for _, id := range order {
+		g := GuestMetrics{}
+		for _, candidate := range guests.Guests {
+			if candidate.VMID == id {
+				g = candidate
+				break
+			}
+		}
+		command := "pct"
+		if g.Type == "qemu" {
+			command = "qm"
+		}
+		ctx, cancel := context.WithTimeout(context.Background(), time.Duration(settings.GraceSeconds+30)*time.Second)
+		err := exec.CommandContext(ctx, command, "shutdown", fmt.Sprint(id), "--timeout", fmt.Sprint(settings.GraceSeconds)).Run()
+		cancel()
+		if err != nil {
+			log.Printf("UPS shutdown: не удалось остановить VMID %d: %v", id, err)
+		}
+	}
+	log.Printf("UPS shutdown: гости обработаны, выключаем Proxmox host")
+	_ = exec.Command("shutdown", "-h", "now").Run()
+}

reliability_test.go

@@ -0,0 +1,38 @@
+package main
+
+import (
+	"strings"
+	"testing"
+	"time"
+)
+
+func TestDiskCounterChanges(t *testing.T) {
+	old := DiskMetrics{Reallocated: 1, Pending: 2, CRCErrors: 4}
+	current := DiskMetrics{Reallocated: 2, Pending: 2, CRCErrors: 7}
+	changes := strings.Join(diskCounterChanges(old, current), " ")
+	if !strings.Contains(changes, "Reallocated") || !strings.Contains(changes, "CRC") || strings.Contains(changes, "Pending") {
+		t.Fatalf("unexpected counter changes: %s", changes)
+	}
+}
+
+func TestParseScrub(t *testing.T) {
+	status, at, duration, old := parseScrub("scrub repaired 0B in 01:12:03 with 0 errors on Fri Jan  2 03:04:05 2026")
+	if status != "Завершён" || at == "" || duration != "01:12:03" || !old {
+		t.Fatalf("unexpected scrub: %q %q %q %v", status, at, duration, old)
+	}
+}
+
+func TestShutdownDryRunUsesConfiguredOrder(t *testing.T) {
+	settings := UPSShutdownSettings{ChargePercent: 15, RuntimeSeconds: 600, GraceSeconds: 120, GuestOrder: []int{105}}
+	guests := GuestsMetrics{Guests: []GuestMetrics{{VMID: 101, Name: "one", Type: "qemu", Status: "running"}, {VMID: 105, Name: "five", Type: "lxc", Status: "running"}}}
+	steps := buildShutdownDryRun(settings, guests)
+	if len(steps) != 4 || !strings.Contains(steps[1].Target, "105") || !strings.Contains(steps[2].Target, "101") || steps[3].Target != "Proxmox host" {
+		t.Fatalf("unexpected plan: %+v", steps)
+	}
+}
+
+func TestSelfTestOlderThanWeek(t *testing.T) {
+	if !selfTestOlderThanWeek(time.Now().Add(-8*24*time.Hour).Format(time.RFC3339)) || selfTestOlderThanWeek(time.Now().Format(time.RFC3339)) {
+		t.Fatal("self-test age evaluation is incorrect")
+	}
+}

store.go

@@ -7,6 +7,7 @@ import (
 	"os"
 	"path/filepath"
 	"strings"
+	"sync"
 	"time"
 
 	_ "modernc.org/sqlite"
@@ -58,7 +59,13 @@ type HistoryPoint struct {
 	GuestsMemory    float64  `json:"guestsMemory"`
 }
 
-type Store struct{ db *sql.DB }
+type Store struct {
+	db               *sql.DB
+	path             string
+	healthMu         sync.RWMutex
+	lastHistoryAt    time.Time
+	lastHistoryError string
+}
 
 type EmailSettings struct {
 	Enabled         bool     `json:"enabled"`
@@ -130,7 +137,7 @@ func openStore(path string) (*Store, error) {
 	_, _ = db.Exec(`ALTER TABLE alert_events ADD COLUMN recovery_sent_at INTEGER`)
 	_, _ = db.Exec(`ALTER TABLE service_checks ADD COLUMN endpoint TEXT NOT NULL DEFAULT 'public'`)
 	_, _ = db.Exec(`PRAGMA foreign_keys=ON`)
-	return &Store{db: db}, nil
+	return &Store{db: db, path: path}, nil
 }
 
 type EntityHistoryPoint struct {
@@ -285,9 +292,45 @@ func (s *Store) AddHistory(point HistoryPoint) error {
 	_, err := s.db.Exec(`INSERT OR REPLACE INTO history(ts,cpu_usage,cpu_temp,memory_usage,swap_usage,root_usage,network_rx,network_tx,ups_charge,ups_load,zfs_usage,guests_cpu,guests_memory) VALUES(?,?,?,?,?,?,?,?,?,?,?,?,?)`,
 		point.Timestamp, point.CPUUsage, point.CPUTemperature, point.MemoryUsage, point.SwapUsage, point.RootUsage, point.NetworkReceive, point.NetworkTransmit, point.UPSCharge, point.UPSLoad, point.ZFSUsage, point.GuestsCPU, point.GuestsMemory)
 	if err != nil {
+		s.setHistoryHealth(time.Time{}, err)
 		return err
 	}
 	_, err = s.db.Exec(`DELETE FROM history WHERE ts < ?`, time.Now().Add(-historyRetention).Unix())
+	s.setHistoryHealth(time.Now(), err)
+	return err
+}
+
+func (s *Store) setHistoryHealth(at time.Time, err error) {
+	s.healthMu.Lock()
+	defer s.healthMu.Unlock()
+	if !at.IsZero() {
+		s.lastHistoryAt = at
+	}
+	if err != nil {
+		s.lastHistoryError = err.Error()
+	} else {
+		s.lastHistoryError = ""
+	}
+}
+func (s *Store) historyHealth() (time.Time, string) {
+	s.healthMu.RLock()
+	defer s.healthMu.RUnlock()
+	return s.lastHistoryAt, s.lastHistoryError
+}
+
+func (s *Store) loadJSONSetting(key string, target any) bool {
+	var value string
+	if s.db.QueryRow(`SELECT value FROM settings WHERE key=?`, key).Scan(&value) != nil {
+		return false
+	}
+	return json.Unmarshal([]byte(value), target) == nil
+}
+func (s *Store) saveJSONSetting(key string, value any) error {
+	data, err := json.Marshal(value)
+	if err != nil {
+		return err
+	}
+	_, err = s.db.Exec(`INSERT INTO settings(key,value) VALUES(?,?) ON CONFLICT(key) DO UPDATE SET value=excluded.value`, key, string(data))
 	return err
 }
 

ups.go

@@ -10,20 +10,26 @@ import (
 )
 
 type UPSMetrics struct {
-	Available      bool     `json:"available"`
-	Source         string   `json:"source"`
-	Name           string   `json:"name"`
-	Model          string   `json:"model"`
-	Status         string   `json:"status"`
-	StatusLabel    string   `json:"statusLabel"`
-	ChargePercent  *float64 `json:"chargePercent"`
-	RuntimeSeconds *float64 `json:"runtimeSeconds"`
-	LoadPercent    *float64 `json:"loadPercent"`
-	InputVoltage   *float64 `json:"inputVoltage"`
-	OutputVoltage  *float64 `json:"outputVoltage"`
-	InputFrequency *float64 `json:"inputFrequency"`
-	BatteryVoltage *float64 `json:"batteryVoltage"`
-	Temperature    *float64 `json:"temperatureCelsius"`
+	Available         bool     `json:"available"`
+	Source            string   `json:"source"`
+	Name              string   `json:"name"`
+	Model             string   `json:"model"`
+	Status            string   `json:"status"`
+	StatusLabel       string   `json:"statusLabel"`
+	ChargePercent     *float64 `json:"chargePercent"`
+	RuntimeSeconds    *float64 `json:"runtimeSeconds"`
+	LoadPercent       *float64 `json:"loadPercent"`
+	InputVoltage      *float64 `json:"inputVoltage"`
+	OutputVoltage     *float64 `json:"outputVoltage"`
+	InputFrequency    *float64 `json:"inputFrequency"`
+	BatteryVoltage    *float64 `json:"batteryVoltage"`
+	Temperature       *float64 `json:"temperatureCelsius"`
+	CommunicationOK   bool     `json:"communicationOk"`
+	BatteryDate       string   `json:"batteryDate"`
+	BatteryAgeDays    int      `json:"batteryAgeDays,omitempty"`
+	ReplaceBattery    bool     `json:"replaceBattery"`
+	LowBatteryCharge  *float64 `json:"lowBatteryCharge,omitempty"`
+	LowBatteryRuntime *float64 `json:"lowBatteryRuntime,omitempty"`
 }
 
 type upsCollector struct {
@@ -43,7 +49,13 @@ func (c *upsCollector) collect() UPSMetrics {
 	} else if metrics, ok := readAPCUPSD(); ok {
 		c.cached = metrics
 	} else {
-		c.cached = UPSMetrics{}
+		if c.cached.Available || c.cached.Source != "" {
+			c.cached.Available = false
+			c.cached.CommunicationOK = false
+			c.cached.StatusLabel = "Нет связи"
+		} else {
+			c.cached = UPSMetrics{}
+		}
 	}
 	c.updatedAt = time.Now()
 	return c.cached
@@ -85,6 +97,7 @@ func readAPCUPSD() (UPSMetrics, bool) {
 		"ups.load": raw["LOADPCT"], "input.voltage": raw["LINEV"],
 		"output.voltage": raw["OUTPUTV"], "input.frequency": raw["LINEFREQ"],
 		"battery.voltage": raw["BATTV"], "ups.temperature": raw["ITEMP"],
+		"battery.date": raw["BATTDATE"], "battery.charge.low": raw["MBATTCHG"], "battery.runtime.low": minutesToSeconds(raw["MINTIMEL"]),
 	}
 	return upsFromValues("apcupsd", raw["UPSNAME"], values), true
 }
@@ -102,14 +115,24 @@ func parseColonValues(output string) map[string]string {
 
 func upsFromValues(source, name string, values map[string]string) UPSMetrics {
 	status := values["ups.status"]
-	return UPSMetrics{
-		Available: true, Source: source, Name: name, Model: values["device.model"],
+	metrics := UPSMetrics{
+		Available: true, CommunicationOK: true, Source: source, Name: name, Model: values["device.model"],
 		Status: status, StatusLabel: upsStatusLabel(status),
 		ChargePercent: numberPointer(values["battery.charge"]), RuntimeSeconds: numberPointer(values["battery.runtime"]),
 		LoadPercent: numberPointer(values["ups.load"]), InputVoltage: numberPointer(values["input.voltage"]),
 		OutputVoltage: numberPointer(values["output.voltage"]), InputFrequency: numberPointer(values["input.frequency"]),
 		BatteryVoltage: numberPointer(values["battery.voltage"]), Temperature: numberPointer(firstNonEmpty(values["ups.temperature"], values["battery.temperature"])),
+		BatteryDate: firstNonEmpty(values["battery.date"], values["battery.mfr.date"]), ReplaceBattery: strings.Contains(status, "RB"), LowBatteryCharge: numberPointer(values["battery.charge.low"]), LowBatteryRuntime: numberPointer(values["battery.runtime.low"]),
 	}
+	if metrics.BatteryDate != "" {
+		for _, layout := range []string{"2006/01/02", "2006-01-02", "01/02/2006", "02/01/2006"} {
+			if date, err := time.Parse(layout, metrics.BatteryDate); err == nil {
+				metrics.BatteryAgeDays = int(time.Since(date).Hours() / 24)
+				break
+			}
+		}
+	}
+	return metrics
 }
 
 func numberPointer(value string) *float64 {

web.go

@@ -15,6 +15,7 @@ type dashboardMetrics struct {
 	Memory       MemoryMetrics       `json:"memory"`
 	Storage      StorageMetrics      `json:"storage"`
 	Disks        []DiskMetrics       `json:"disks"`
+	MissingDisks []string            `json:"missingDisks"`
 	Network      NetworkMetrics      `json:"network"`
 	ZFS          ZFSMetrics          `json:"zfs"`
 	UPS          UPSMetrics          `json:"ups"`
@@ -26,10 +27,12 @@ type dashboardMetrics struct {
 	Services     ServicesMetrics     `json:"services"`
 	SystemHealth SystemHealthMetrics `json:"systemHealth"`
 	Activity     ProxmoxActivity     `json:"activity"`
+	Monitor      MonitorHealth       `json:"monitor"`
+	UPSShutdown  UPSShutdownSettings `json:"upsShutdown"`
 	Timestamp    time.Time           `json:"timestamp"`
 }
 
-func collectDashboard(cpuCollector *cpuCollector, networkCollector *networkCollector, diskCollector *diskCollector, zfsCollector *zfsCollector, upsCollector *upsCollector, guestCollector *guestCollector, backupCollector *backupCollector, maintenanceCollector *maintenanceCollector, serviceCollector *serviceCollector, systemHealth *systemHealthCollector, activity *activityCollector, store *Store) (dashboardMetrics, error) {
+func collectDashboard(cpuCollector *cpuCollector, networkCollector *networkCollector, diskCollector *diskCollector, zfsCollector *zfsCollector, upsCollector *upsCollector, guestCollector *guestCollector, backupCollector *backupCollector, maintenanceCollector *maintenanceCollector, serviceCollector *serviceCollector, systemHealth *systemHealthCollector, activity *activityCollector, store *Store, monitor *monitorTracker) (dashboardMetrics, error) {
 	cpu, err := cpuCollector.collect()
 	if err != nil {
 		return dashboardMetrics{}, err
@@ -52,6 +55,9 @@ func collectDashboard(cpuCollector *cpuCollector, networkCollector *networkColle
 		ZFS: zfsCollector.collect(), UPS: upsCollector.collect(),
 		Guests: guestCollector.collect(), Backups: backupCollector.collect(), Maintenance: maintenanceCollector.collect(), Services: serviceCollector.collect(), SystemHealth: systemHealth.collect(), Update: readUpdateStatus(), Timestamp: time.Now(),
 	}
+	metrics.MissingDisks = diskCollector.missingDisks()
+	metrics.Monitor = monitor.snapshot(store)
+	metrics.UPSShutdown = store.UPSShutdownSettings()
 	metrics.Activity = activity.collect(metrics.Guests)
 	decorateNetworkInterfaces(&metrics.Network, metrics.Guests)
 	metrics.Alerts = evaluateAlerts(metrics, store.Thresholds())
@@ -115,8 +121,13 @@ func routes(collector *cpuCollector, store *Store) http.Handler {
 	services := newServiceCollector(store)
 	systemHealth := newSystemHealthCollector()
 	activity := newActivityCollector()
+	monitor := &monitorTracker{}
+	go runUPSShutdownController(store, ups, guests)
 	collect := func() (dashboardMetrics, error) {
-		return collectDashboard(collector, network, disks, zfs, ups, guests, backups, maintenance, services, systemHealth, activity, store)
+		started := time.Now()
+		metrics, err := collectDashboard(collector, network, disks, zfs, ups, guests, backups, maintenance, services, systemHealth, activity, store, monitor)
+		monitor.record(started, err)
+		return metrics, err
 	}
 	go runHistoryCollector(collect, store)
 	go runEmailNotifier(collect, store)
@@ -219,6 +230,32 @@ func routes(collector *cpuCollector, store *Store) http.Handler {
 		settings := store.EmailSettings(true)
 		writeJSON(w, map[string]bool{"sent": true}, sendAlertEmail(settings, nil, true))
 	})
+	mux.HandleFunc("/api/settings/ups-shutdown", func(w http.ResponseWriter, r *http.Request) {
+		switch r.Method {
+		case http.MethodGet:
+			writeJSON(w, store.UPSShutdownSettings(), nil)
+		case http.MethodPut:
+			var settings UPSShutdownSettings
+			if json.NewDecoder(io.LimitReader(r.Body, 16<<10)).Decode(&settings) != nil {
+				http.Error(w, "Некорректные настройки", http.StatusBadRequest)
+				return
+			}
+			if err := store.SaveUPSShutdownSettings(settings); err != nil {
+				http.Error(w, err.Error(), http.StatusBadRequest)
+				return
+			}
+			writeJSON(w, settings, nil)
+		default:
+			w.WriteHeader(http.StatusMethodNotAllowed)
+		}
+	})
+	mux.HandleFunc("/api/ups/shutdown-dry-run", func(w http.ResponseWriter, r *http.Request) {
+		if r.Method != http.MethodPost {
+			w.WriteHeader(http.StatusMethodNotAllowed)
+			return
+		}
+		writeJSON(w, buildShutdownDryRun(store.UPSShutdownSettings(), guests.collect()), nil)
+	})
 	mux.HandleFunc("/api/history", func(w http.ResponseWriter, r *http.Request) {
 		hours, _ := strconv.Atoi(r.URL.Query().Get("hours"))
 		points, err := store.History(hours)
@@ -347,6 +384,7 @@ footer{color:#657581;font-size:12px;margin-top:17px}
 .services .fill{background:linear-gradient(90deg,#55d58a,#43c6d9)}.system-health .fill{background:linear-gradient(90deg,#6ba8ff,#9d8cff)}.service-form{display:grid;grid-template-columns:1fr 1fr;gap:12px}.field{display:grid;gap:6px;min-width:0}.field>span{color:var(--muted);font-size:11px;text-transform:uppercase;letter-spacing:.07em}.field small{color:var(--muted);font-size:11px}.field input,.field select{width:100%;min-width:0;padding:11px 12px;border:1px solid var(--line);border-radius:9px;background:var(--bg);color:var(--text)}.check-field{display:flex;align-items:center;gap:10px;padding:12px;border:1px solid var(--line);border-radius:9px}.check-field input{width:18px;height:18px;accent-color:var(--green)}.service-actions{display:flex;flex-wrap:wrap;gap:7px;margin-top:10px}.service-actions button,.open-address{padding:7px 10px;border:1px solid var(--line);border-radius:8px;background:var(--panel);color:var(--text);cursor:pointer;text-decoration:none}.status-up{color:var(--green)}.status-down{color:#ff665c}.issue-message{overflow-wrap:anywhere}.full-span{grid-column:1/-1}.endpoint-list{display:grid;gap:10px;margin-top:16px}.endpoint{padding:14px;border:1px solid var(--line);border-radius:12px;background:rgba(100,130,150,.04)}.endpoint-head{display:flex;align-items:flex-start;justify-content:space-between;gap:12px}.endpoint-address{color:var(--muted);font:12px ui-monospace,monospace;overflow-wrap:anywhere}.certificate{margin-top:11px;padding-top:11px;border-top:1px solid var(--line);color:var(--muted);font-size:12px}.certificate b{color:var(--text)}.maintenance-grid .detail{min-width:0}.maintenance-grid .detail strong{font-size:clamp(15px,2.2vw,21px);overflow-wrap:anywhere;word-break:break-word}.system-tabs{display:flex;gap:6px;padding:12px 26px;border-bottom:1px solid var(--line)}.system-panel[hidden]{display:none}
 .snapshots .fill{background:linear-gradient(90deg,#9d8cff,#ff78b8)}.tasks .fill{background:linear-gradient(90deg,#43c6d9,#6ba8ff)}.homelab-map .fill{background:linear-gradient(90deg,#55d58a,#9d8cff)}.map-board{display:grid;grid-template-columns:170px 1fr 1fr 1fr;gap:14px;padding:18px;min-width:680px}.map-column{display:grid;align-content:start;gap:9px}.map-column-title{color:var(--muted);font-size:10px;text-transform:uppercase;letter-spacing:.1em}.map-node{padding:12px;border:1px solid var(--line);border-radius:12px;background:var(--panel)}.map-node.good{border-color:#2d6245}.map-node.bad{border-color:#71362f}.map-node.warn{border-color:#6d4c2c}.map-node strong{display:block}.map-node small{display:block;color:var(--muted);margin-top:3px}.map-scroll{overflow-x:auto}.backup-check{margin-top:9px;color:var(--muted);font-size:12px}.task-filter{display:flex;gap:6px;padding:12px 16px;overflow-x:auto}
 .topology-board{padding:18px;min-width:620px}.topology-host{margin-bottom:14px}.topology-grid{display:grid;grid-template-columns:repeat(2,minmax(270px,1fr));gap:12px}.topology-group{padding:14px;border:1px solid var(--line);border-radius:15px;background:rgba(100,130,150,.04)}.topology-group-head{display:flex;justify-content:space-between;gap:10px;padding-bottom:10px;border-bottom:1px solid var(--line)}.topology-section{margin-top:11px}.topology-section-title{color:var(--muted);font-size:10px;text-transform:uppercase;letter-spacing:.08em;margin-bottom:6px}.topology-items{display:flex;flex-wrap:wrap;gap:6px}.topology-chip{padding:6px 9px;border:1px solid var(--line);border-radius:9px;background:var(--panel);font-size:12px}.topology-chip.good{border-color:#2d6245}.topology-chip.bad{border-color:#71362f}.topology-unassigned{margin-top:14px}@media(max-width:700px){.topology-grid{grid-template-columns:1fr}}
+.monitor .fill{background:linear-gradient(90deg,#55d58a,#43c6d9)}.shutdown-plan{display:grid;gap:8px;margin-top:12px}.shutdown-step{padding:10px 12px;border:1px solid var(--line);border-radius:10px}.infra-status{display:grid;grid-template-columns:repeat(4,1fr);gap:8px;margin:14px 0}.infra-status>div{padding:11px;border:1px solid var(--line);border-radius:10px}.infra-status b{display:block;margin-top:4px}@media(max-width:700px){.infra-status{grid-template-columns:1fr 1fr}}
 </style>
 </head>
 <body><main>
@@ -394,6 +432,10 @@ footer{color:#657581;font-size:12px;margin-top:17px}
  <div class="quick"><div><span class="label">Состояние</span><strong id="upsStatus">—</strong></div><div><span class="label">Автономность</span><strong id="upsRuntime">—</strong></div></div>
  <div class="hint">Нажмите, чтобы посмотреть подробности →</div>
 </button>
+<button class="card monitor" data-card="monitor" data-open="monitorDialog" aria-label="Открыть самоконтроль дашборда">
+ <div class="card-head"><div><div class="label">Monitoring health</div><div class="name">Самоконтроль</div></div><div class="percent"><span id="monitorState">—</span></div></div><div class="bar"><div class="fill" id="monitorBar"></div></div>
+ <div class="quick"><div><span class="label">Последний сбор</span><strong id="monitorLast">—</strong></div><div><span class="label">База</span><strong id="monitorDB">—</strong></div></div><div class="hint">Watchdog, история и свежесть данных →</div>
+</button>
 <button class="card guests" data-card="guests" data-open="guestsDialog" aria-label="Открыть подробности VM и LXC">
  <div class="card-head"><div><div class="label">Virtualization</div><div class="name">VM и LXC</div></div><div class="percent"><span id="guestRunning">—</span><small id="guestTotal"> / —</small></div></div>
  <div class="bar"><div class="fill" id="guestBar"></div></div>
@@ -484,13 +526,14 @@ footer{color:#657581;font-size:12px;margin-top:17px}
   <div class="detail"><span class="label">Получение сейчас</span><strong id="dNetDown">—</strong></div><div class="detail"><span class="label">Передача сейчас</span><strong id="dNetUp">—</strong></div><div class="detail"><span class="label">Суммарная скорость</span><strong id="dNetTotal">—</strong></div>
   <div class="detail"><span class="label">Получено с запуска</span><strong id="dNetReceived">—</strong></div><div class="detail"><span class="label">Передано с запуска</span><strong id="dNetSent">—</strong></div><div class="detail"><span class="label">Интерфейсы</span><strong id="dNetInterfaces">—</strong></div>
  </div>
+ <div class="wide"><span class="label">Доступность инфраструктуры</span><div class="infra-status"><div><span>Шлюз</span><b id="dNetGateway">—</b></div><div><span>DNS</span><b id="dNetDNS">—</b></div><div><span>Интернет</span><b id="dNetInternet">—</b></div><div><span>Внешний IP</span><b id="dNetExternalIP">—</b></div></div></div>
  <div class="disk-list" id="networkInterfacesList"></div>
  <div class="chart-section"><div class="chart-head"><strong>Входящий и исходящий трафик</strong><span>Последние 7 дней</span></div><canvas class="metric-chart" id="networkChart"></canvas></div>
 </dialog>
 
 <dialog id="zfsDialog">
  <div class="modal-head"><div><div class="label">ZFS details</div><h2>Пулы хранения</h2></div><button class="close" aria-label="Закрыть">×</button></div>
- <div class="disk-list" id="zfsList"><span class="empty">Поиск ZFS-пулов…</span></div>
+ <div class="wide"><span class="label">ARC</span><div class="infra-status"><div><span>Размер</span><b id="zfsArcSize">—</b></div><div><span>Цель</span><b id="zfsArcTarget">—</b></div><div><span>Hit ratio</span><b id="zfsArcHit">—</b></div><div><span>Datasets</span><b id="zfsDatasetCount">—</b></div></div></div><div class="disk-list" id="zfsList"><span class="empty">Поиск ZFS-пулов…</span></div>
  <div class="chart-section"><div class="chart-head"><strong>Заполнение каждого ZFS-пула</strong><span>Отдельно по каждому пулу</span></div><canvas class="metric-chart" id="zfsChart"></canvas></div>
 </dialog>
 
@@ -500,11 +543,14 @@ footer{color:#657581;font-size:12px;margin-top:17px}
   <div class="detail"><span class="label">Состояние</span><strong id="dUpsStatus">—</strong></div><div class="detail"><span class="label">Заряд</span><strong id="dUpsCharge">—</strong></div><div class="detail"><span class="label">Автономность</span><strong id="dUpsRuntime">—</strong></div>
   <div class="detail"><span class="label">Нагрузка</span><strong id="dUpsLoad">—</strong></div><div class="detail"><span class="label">Входное напряжение</span><strong id="dUpsInput">—</strong></div><div class="detail"><span class="label">Выходное напряжение</span><strong id="dUpsOutput">—</strong></div>
   <div class="detail"><span class="label">Частота</span><strong id="dUpsFrequency">—</strong></div><div class="detail"><span class="label">Батарея</span><strong id="dUpsBattery">—</strong></div><div class="detail"><span class="label">Температура</span><strong id="dUpsTemp">—</strong></div>
+  <div class="detail"><span class="label">Связь с UPS</span><strong id="dUpsCommunication">—</strong></div><div class="detail"><span class="label">Дата батареи</span><strong id="dUpsBatteryDate">—</strong></div><div class="detail"><span class="label">Возраст батареи</span><strong id="dUpsBatteryAge">—</strong></div>
  </div>
  <div class="wide"><span class="label">Источник данных</span><div class="temps"><span class="temp" id="dUpsSource">Не настроен</span></div></div>
  <div class="chart-section"><div class="chart-head"><strong>Заряд и нагрузка UPS</strong><span>Последние 7 дней</span></div><canvas class="metric-chart" id="upsChart"></canvas></div>
 </dialog>
 
+<dialog id="monitorDialog"><div class="modal-head"><div><div class="label">Dashboard health</div><h2>Самоконтроль мониторинга</h2></div><button class="close" aria-label="Закрыть">×</button></div><div class="detail-grid"><div class="detail"><span class="label">Последний успешный сбор</span><strong id="dMonitorLast">—</strong></div><div class="detail"><span class="label">Длительность цикла</span><strong id="dMonitorDuration">—</strong></div><div class="detail"><span class="label">Последняя запись истории</span><strong id="dMonitorHistory">—</strong></div><div class="detail"><span class="label">Свободно для базы</span><strong id="dMonitorDB">—</strong></div><div class="detail"><span class="label">Systemd watchdog</span><strong id="dMonitorWatchdog">—</strong></div><div class="detail"><span class="label">Ошибка</span><strong id="dMonitorError">—</strong></div></div></dialog>
+
 <dialog id="guestsDialog">
  <div class="modal-head"><div><div class="label">Virtualization details</div><h2>VM и LXC</h2></div><button class="close" aria-label="Закрыть">×</button></div>
  <div class="disk-list" id="guestList"><span class="empty">Получение списка гостей…</span></div>
@@ -547,18 +593,19 @@ footer{color:#657581;font-size:12px;margin-top:17px}
 
 <dialog id="settingsDialog">
  <div class="modal-head"><div><div class="label">Dashboard</div><h2>Настройки</h2></div><button class="close" aria-label="Закрыть">×</button></div>
- <div class="settings-tabs"><button class="settings-tab active" data-settings-tab="appearance">Интерфейс</button><button class="settings-tab" data-settings-tab="alerts">Пороги</button><button class="settings-tab" data-settings-tab="services">Сервисы</button><button class="settings-tab" data-settings-tab="email">Почта</button><button class="settings-tab" data-settings-tab="updates">Обновления</button></div>
+ <div class="settings-tabs"><button class="settings-tab active" data-settings-tab="appearance">Интерфейс</button><button class="settings-tab" data-settings-tab="alerts">Пороги</button><button class="settings-tab" data-settings-tab="services">Сервисы</button><button class="settings-tab" data-settings-tab="ups">UPS shutdown</button><button class="settings-tab" data-settings-tab="email">Почта</button><button class="settings-tab" data-settings-tab="updates">Обновления</button></div>
  <div class="settings-body settings-panel" data-settings-panel="appearance">
    <div class="settings-group"><div class="label">Тема оформления</div><div class="theme-options"><button class="theme-option" data-theme-choice="dark">Тёмная</button><button class="theme-option" data-theme-choice="light">Светлая</button></div></div>
    <div class="settings-group"><div class="label">Видимые карточки</div><div class="setting-list">
     <label class="setting-item"><input type="checkbox" data-card-toggle="alerts"> Центр уведомлений</label>
     <label class="setting-item"><input type="checkbox" data-card-toggle="cpu"> CPU</label><label class="setting-item"><input type="checkbox" data-card-toggle="memory"> Память</label>
     <label class="setting-item"><input type="checkbox" data-card-toggle="storage"> Диски</label><label class="setting-item"><input type="checkbox" data-card-toggle="network"> Сеть</label>
-    <label class="setting-item"><input type="checkbox" data-card-toggle="zfs"> ZFS</label><label class="setting-item"><input type="checkbox" data-card-toggle="ups"> UPS</label>
+    <label class="setting-item"><input type="checkbox" data-card-toggle="zfs"> ZFS</label><label class="setting-item"><input type="checkbox" data-card-toggle="ups"> UPS</label><label class="setting-item"><input type="checkbox" data-card-toggle="monitor"> Самоконтроль</label>
     <label class="setting-item"><input type="checkbox" data-card-toggle="guests"> VM и LXC</label><label class="setting-item"><input type="checkbox" data-card-toggle="maintenance"> Proxmox</label><label class="setting-item"><input type="checkbox" data-card-toggle="backups"> Резервные копии</label><label class="setting-item"><input type="checkbox" data-card-toggle="services"> Сервисы</label><label class="setting-item"><input type="checkbox" data-card-toggle="system"> Система</label><label class="setting-item"><input type="checkbox" data-card-toggle="snapshots"> Snapshots</label><label class="setting-item"><input type="checkbox" data-card-toggle="tasks"> Задачи Proxmox</label><label class="setting-item"><input type="checkbox" data-card-toggle="map"> Карта homelab</label>
    </div></div>
    <div class="settings-group"><div class="label">Порядок карточек</div><div class="settings-note">Перетаскивайте строки или используйте стрелки.</div><div class="card-order" id="cardOrder"></div></div>
  </div>
+ <div class="settings-body settings-panel" data-settings-panel="ups" hidden><div class="settings-group"><div class="label">Безопасное выключение при работе от батареи</div><div class="settings-note">Автоматическое выключение по умолчанию отключено. Dry-run ничего не останавливает и показывает будущий порядок.</div><div class="threshold-grid"><label class="setting-item"><input type="checkbox" id="upsShutdownEnabled"> Разрешить сценарий</label><label class="threshold-item"><span>Заряд, %</span><input type="number" id="upsShutdownCharge"></label><label class="threshold-item"><span>Автономность, сек.</span><input type="number" id="upsShutdownRuntime"></label><label class="threshold-item"><span>Ожидание гостя, сек.</span><input type="number" id="upsShutdownGrace"></label><label class="field full-span"><span>Порядок VMID через запятую</span><input id="upsShutdownOrder" placeholder="101, 105, 220"><small>Неуказанные запущенные гости добавятся после них по VMID.</small></label></div><button class="save-settings" id="saveUPSShutdown">Сохранить сценарий</button><button class="theme-option" id="dryRunUPSShutdown" style="margin-top:10px;width:100%">Показать dry-run</button><div class="shutdown-plan" id="upsShutdownPlan"></div><div class="settings-note" id="upsShutdownStatus"></div></div></div>
  <div class="settings-body settings-panel" data-settings-panel="alerts" hidden>
   <div class="threshold-grid">
    <label class="threshold-item"><span>CPU · предупреждение, %</span><input type="number" data-threshold="cpuWarning"></label><label class="threshold-item"><span>CPU · критично, %</span><input type="number" data-threshold="cpuCritical"></label>
@@ -610,10 +657,11 @@ const rate=n=>size(n)+'/s';
 const safe=v=>String(v??'').replace(/[&<>"']/g,c=>({'&':'&amp;','<':'&lt;','>':'&gt;','"':'&quot;',"'":'&#39;'}[c]));
 const uptime=s=>{const d=Math.floor(s/86400),h=Math.floor(s%86400/3600),m=Math.floor(s%3600/60);return d?d+'д '+h+'ч':h?h+'ч '+m+'м':m+'м'};
 let alertSettings={backupMaxAgeHours:48};
+let lastDashboardEvent=0;
 let currentServiceStatuses=[];
 function endpointHTML(v){const tls=v.tlsDays==null?'Нет TLS':v.tlsDays<0?'Сертификат истёк':v.tlsDays+' дн. до окончания',cert=v.tlsSubject?'<div class="certificate"><div>Сертификат: <b>'+safe(v.tlsSubject)+'</b></div><div>Домены: <b>'+safe((v.tlsDnsNames||[]).join(', ')||'не указаны')+'</b></div><div>Издатель: <b>'+safe(v.tlsIssuer||'—')+'</b></div><div>Для этого адреса: <b class="'+(v.tlsHostnameValid?'status-up':'status-down')+'">'+(v.tlsHostnameValid?'подходит':'не подходит')+'</b></div></div>':'';return '<div class="endpoint"><div class="endpoint-head"><div><b>'+safe(v.label)+'</b><div class="endpoint-address">'+safe(v.address)+'</div></div><span class="badge '+(v.up?'':'bad')+'">'+(v.up?'Доступен':'Недоступен')+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Отклик</span><b>'+fixed(v.latencyMs)+' ms</b></div><div class="disk-stat"><span>HTTP</span><b>'+(v.statusCode||'—')+'</b></div><div class="disk-stat"><span>TLS</span><b>'+tls+'</b></div><div class="disk-stat"><span>Uptime · сутки</span><b>'+fixed(v.uptimeDay)+'%</b></div><div class="disk-stat"><span>Uptime · неделя</span><b>'+fixed(v.uptimeWeek)+'%</b></div></div>'+cert+(v.error?'<div class="settings-note status-down">'+safe(v.error)+'</div>':'')+'</div>'}
 function renderServices(svc){const items=svc.services||[];currentServiceStatuses=items;const avg=items.length?items.reduce((a,v)=>a+v.uptimeDay,0)/items.length:0,live=items.filter(v=>v.up),latency=live.length?live.reduce((a,v)=>a+v.latencyMs,0)/live.length:0;el('serviceUp').textContent=svc.up||0;el('serviceTotal').textContent=' / '+(svc.total||0);el('serviceBar').style.width=(svc.total?svc.up/svc.total*100:0)+'%';el('serviceUptime').textContent=svc.total?fixed(avg)+'%':'—';el('serviceLatency').textContent=svc.total?fixed(latency)+' ms':'—';el('serviceList').innerHTML=items.length?items.map(v=>'<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(v.name)+'</div><div class="disk-meta">'+(v.endpoints||[]).length+' '+((v.endpoints||[]).length===1?'адрес':'адреса')+' проверяются независимо</div></div><span class="badge '+(v.up?'':'bad')+'">'+(v.up?'Всё доступно':'Есть проблема')+'</span></div><div class="endpoint-list">'+(v.endpoints||[]).map(endpointHTML).join('')+'</div><div class="service-actions"><button data-service-open="'+v.id+'">Открыть сервис ↗</button></div></article>').join(''):'<span class="empty">Добавьте первую проверку в настройках.</span>'}
-function renderNetworkGroups(interfaces){const groups=new Map();interfaces.forEach(v=>{const key=v.vmid?'guest-'+v.vmid:'host',group=groups.get(key)||{vmid:v.vmid,name:v.vmid?(v.guestName||('VMID '+v.vmid)):'Интерфейсы хоста',type:v.guestType||'Host',items:[]};group.items.push(v);groups.set(key,group)});el('networkInterfacesList').innerHTML=[...groups.values()].map(g=>{const primary=g.vmid?g.items.filter(v=>v.name.startsWith('tap')||v.name.startsWith('veth')):g.items,rx=primary.reduce((a,v)=>a+v.receivePerSec,0),tx=primary.reduce((a,v)=>a+v.transmitPerSec,0),errors=g.items.reduce((a,v)=>a+v.receiveErrors+v.transmitErrors+v.receiveDrops+v.transmitDrops,0);return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(g.name)+'</div><div class="disk-meta">'+(g.vmid?'VMID '+g.vmid+' · ':'')+safe(g.type)+' · '+safe(g.items.map(v=>v.name).join(', '))+'</div></div><span class="badge '+(errors?'warn':'')+'">'+(errors?'Есть потери':'Без ошибок')+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Сейчас ↓</span><b>'+rate(rx)+'</b></div><div class="disk-stat"><span>Сейчас ↑</span><b>'+rate(tx)+'</b></div><div class="disk-stat"><span>Получено</span><b>'+size(primary.reduce((a,v)=>a+v.receiveBytes,0))+'</b></div><div class="disk-stat"><span>Передано</span><b>'+size(primary.reduce((a,v)=>a+v.transmitBytes,0))+'</b></div><div class="disk-stat"><span>Ошибки / drops</span><b>'+errors+'</b></div></div></article>'}).join('')}
+function renderNetworkGroups(interfaces){const groups=new Map();interfaces.forEach(v=>{const key=v.vmid?'guest-'+v.vmid:'host',group=groups.get(key)||{vmid:v.vmid,name:v.vmid?(v.guestName||('VMID '+v.vmid)):'Интерфейсы хоста',type:v.guestType||'Host',items:[]};group.items.push(v);groups.set(key,group)});el('networkInterfacesList').innerHTML=[...groups.values()].map(g=>{const primary=g.vmid?g.items.filter(v=>v.name.startsWith('tap')||v.name.startsWith('veth')):g.items,rx=primary.reduce((a,v)=>a+v.receivePerSec,0),tx=primary.reduce((a,v)=>a+v.transmitPerSec,0),errors=g.items.reduce((a,v)=>a+v.receiveErrors+v.transmitErrors+v.receiveDrops+v.transmitDrops,0),down=primary.some(v=>v.kind==='Physical'&&!v.carrier&&(v.master||v.receiveBytes+v.transmitBytes>0));return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(g.name)+'</div><div class="disk-meta">'+(g.vmid?'VMID '+g.vmid+' · ':'')+safe(g.type)+' · '+safe(g.items.map(v=>v.name+' ['+(v.kind||'Interface')+(v.speedMbps?' · '+v.speedMbps+'Mb/s':'')+(v.duplex?' · '+v.duplex:'')+(v.master?' → '+v.master:'')+']').join(', '))+'</div></div><span class="badge '+(errors||down?'warn':'')+'"'+(errors||down?' tabindex="0" data-reason="'+safe(down?'Нет carrier на используемом физическом интерфейсе':'Счётчики errors/drops больше нуля')+'"':'')+'>'+(down?'Нет линка':errors?'Есть потери':'Без ошибок')+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Сейчас ↓</span><b>'+rate(rx)+'</b></div><div class="disk-stat"><span>Сейчас ↑</span><b>'+rate(tx)+'</b></div><div class="disk-stat"><span>Получено</span><b>'+size(primary.reduce((a,v)=>a+v.receiveBytes,0))+'</b></div><div class="disk-stat"><span>Передано</span><b>'+size(primary.reduce((a,v)=>a+v.transmitBytes,0))+'</b></div><div class="disk-stat"><span>Ошибки / drops</span><b>'+errors+'</b></div></div></article>'}).join('')}
 const taskLabel=type=>({vzdump:'Backup',qmstart:'Запуск VM',qmstop:'Остановка VM',qmshutdown:'Выключение VM',qmsnapshot:'Snapshot VM',vzstart:'Запуск LXC',vzstop:'Остановка LXC',vzsnapshot:'Snapshot LXC',qmigrate:'Миграция',aptupdate:'Обновление пакетов'})[type]||type;
 function explainWarningBadges(){el('snapshotList').querySelectorAll('.badge.warn').forEach(v=>{v.tabIndex=0;v.dataset.reason='Snapshot старше 30 дней. Проверьте, нужен ли он: старые снимки могут занимать место в storage.'});el('taskList').querySelectorAll('.badge.bad').forEach(v=>{v.tabIndex=0;v.dataset.reason='Задача Proxmox завершилась не со статусом OK. Откройте её журнал в интерфейсе Proxmox для подробностей.'})}
 let currentTasks=[],currentTaskFilter='all';
@@ -622,6 +670,7 @@ function renderActivity(activity,guests,services,zfs){const snapshots=activity.s
 function topologyHost(address){if(!address)return '';try{return new URL(address.includes('://')?address:'http://'+address).hostname.replace(/^\[|\]$/g,'')}catch(e){return address.replace(/^[a-z]+:\/\//i,'').split('/')[0].replace(/:\d+$/,'')}}
 function renderHomelabMap(guests,services,zfs){const guestItems=(guests.guests||[]).filter(v=>!v.template),serviceItems=services.services||[],pools=zfs.pools||[],serviceOwners=new Map();serviceItems.forEach(s=>{const host=topologyHost(s.localAddress||'');const owner=guestItems.find(g=>(g.ipAddresses||[]).some(ip=>ip===host));serviceOwners.set(s.id,owner?.vmid||0)});const serviceChip=s=>{const healthy=(s.endpoints||[]).every(v=>v.up);return '<span class="topology-chip '+(healthy?'good':'bad')+'"'+(!healthy?' tabindex="0" data-reason="Один или несколько адресов сервиса недоступны"':'')+'>'+safe(s.name)+'</span>'},storageChip=name=>{const pool=pools.find(p=>p.name===name),healthy=!pool||pool.health==='ONLINE';return '<span class="topology-chip '+(healthy?'good':'bad')+'">'+safe(name)+(pool?' · '+fixed(pool.capacityPercent)+'%':'')+'</span>'};const groups=guestItems.map(g=>{const owned=serviceItems.filter(s=>serviceOwners.get(s.id)===g.vmid),storage=[...new Set(g.storage||[])],running=g.status==='running';return '<section class="topology-group"><div class="topology-group-head"><div><strong>'+safe(g.name||('VMID '+g.vmid))+'</strong><div class="disk-meta">VMID '+g.vmid+' · '+safe(g.type==='qemu'?'VM':'LXC')+' · '+safe((g.ipAddresses||[]).join(', ')||'IP не получен')+'</div></div><span class="badge '+(running?'':'warn')+'"'+(!running?' tabindex="0" data-reason="VM/LXC сейчас остановлена"':'')+'>'+(running?'Работает':'Остановлена')+'</span></div><div class="topology-section"><div class="topology-section-title">Сервисы</div><div class="topology-items">'+(owned.length?owned.map(serviceChip).join(''):'<span class="empty">Не сопоставлены</span>')+'</div></div><div class="topology-section"><div class="topology-section-title">Storage</div><div class="topology-items">'+(storage.length?storage.map(storageChip).join(''):'<span class="empty">Не определён</span>')+'</div></div></section>'}).join(''),unassigned=serviceItems.filter(s=>!serviceOwners.get(s.id)),usedStorage=new Set(guestItems.flatMap(g=>g.storage||[])),unassignedPools=pools.filter(p=>!usedStorage.has(p.name));el('mapBoard').innerHTML='<div class="topology-host map-node good"><strong>Proxmox host</strong><small>'+safe(location.hostname)+' · '+guestItems.length+' VM/LXC</small></div><div class="topology-grid">'+groups+'</div>'+((unassigned.length||unassignedPools.length)?'<section class="topology-group topology-unassigned"><div class="topology-group-head"><strong>Без автоматической привязки</strong><span class="badge warn" tabindex="0" data-reason="Для сервиса не найден совпадающий IP гостя либо storage не используется в конфигурации VM/LXC">Проверьте связи</span></div><div class="topology-section"><div class="topology-section-title">Сервисы</div><div class="topology-items">'+unassigned.map(serviceChip).join('')+'</div></div><div class="topology-section"><div class="topology-section-title">Остальные ZFS-пулы</div><div class="topology-items">'+unassignedPools.map(p=>storageChip(p.name)).join('')+'</div></div></section>':'');const storage=new Set([...usedStorage,...pools.map(p=>p.name)]);el('mapGuests').textContent=guestItems.length;el('mapResources').textContent=serviceItems.length+' / '+storage.size;el('mapNodes').textContent=1+guestItems.length+serviceItems.length+storage.size;el('mapBar').style.width='100%'}
 function render(x){const c=x.cpu,m=x.memory,s=x.storage,n=x.network,z=x.zfs||{available:false,pools:[]},u=x.ups||{available:false},g=x.guests||{available:false,guests:[]},b=x.backups||{available:false,items:[]},mt=x.maintenance||{},alerts=x.alerts||[],disks=x.disks||[],avg=c.temperatures.length?c.temperatures.reduce((a,t)=>a+t.celsius,0)/c.temperatures.length:null;
+ lastDashboardEvent=Date.now();
  const update=x.update||{},updating=['queued','installing'].includes(update.state);el('updateCurrent').textContent=update.currentVersion||'—';el('updateAvailable').textContent=update.availableVersion||'—';el('updateMode').textContent=update.mode==='auto'?'Автоматически':'Уведомлять';el('updateMessage').textContent=update.message||'Проверка ещё не выполнялась';el('updateChecked').textContent=update.checkedAt?'Последняя проверка: '+new Date(update.checkedAt).toLocaleString('ru-RU'):'Проверка запускается автоматически раз в сутки.';el('installUpdate').disabled=updating;el('installUpdate').textContent=updating?'Обновление выполняется…':'Проверить и установить обновление';
  const critical=alerts.filter(a=>a.severity==='critical').length,warnings=alerts.filter(a=>a.severity==='warning').length,infos=alerts.filter(a=>a.severity==='info').length;el('alertCount').textContent=alerts.length;el('alertCritical').textContent=critical;el('alertWarnings').textContent=warnings;el('alertHeadline').textContent=critical?'Требуется срочное внимание':warnings?'Есть предупреждения':infos?'Есть информационные события':'Система в норме';el('alertBar').style.width=(critical?100:warnings?65:infos?30:0)+'%';
  el('alertList').innerHTML=alerts.length?alerts.map(a=>'<article class="alert-item '+safe(a.severity)+'"><i class="alert-icon"></i><div><div class="alert-source">'+safe(a.source)+' · '+(a.severity==='critical'?'Критическое':a.severity==='warning'?'Предупреждение':'Информация')+'</div><div class="alert-title">'+safe(a.title)+'</div><div class="alert-message">'+safe(a.message)+'</div><button class="ack-alert" data-ack="'+a.eventId+'">Принято</button></div></article>').join(''):'<div class="all-good"><strong>Всё в порядке</strong><span>Новых неподтверждённых событий нет.</span></div>';
@@ -630,11 +679,13 @@ function render(x){const c=x.cpu,m=x.memory,s=x.storage,n=x.network,z=x.zfs||{av
  el('memUsage').textContent=fixed(m.usagePercent);el('memBar').style.width=Math.min(m.usagePercent,100)+'%';el('memUsed').textContent=size(m.usedBytes);el('memAvailable').textContent=size(m.availableBytes);el('dMemUsed').textContent=size(m.usedBytes);el('dMemAvailable').textContent=size(m.availableBytes);el('dMemTotal').textContent=size(m.totalBytes);el('dMemCache').textContent=size(m.cachedBytes);el('dMemBuffers').textContent=size(m.buffersBytes);el('dMemPercent').textContent=fixed(m.usagePercent)+'%';el('dSwapUsed').textContent=size(m.swapUsedBytes);el('dSwapTotal').textContent=size(m.swapTotalBytes);el('dSwapPercent').textContent=fixed(m.swapPercent)+'%';
  const failed=disks.filter(d=>d.smartStatus==='Ошибка').length,attention=disks.filter(d=>d.smartStatus==='Требует внимания').length,available=disks.filter(d=>d.smartAvailable).length;let health=failed?'Ошибка':attention?'Внимание':available===disks.length&&disks.length?'Исправны':available?'Частично':'Нет данных';
  el('diskCount').textContent=disks.length;el('diskCountLabel').textContent=disks.length===1?' диск':' дисков';el('diskHealth').textContent=health;el('diskUsage').textContent=fixed(s.usagePercent)+'%';el('diskBar').style.width=Math.min(s.usagePercent,100)+'%';el('dDiskUsed').textContent=size(s.usedBytes);el('dDiskFree').textContent=size(s.freeBytes);el('dDiskTotal').textContent=size(s.totalBytes);el('dDiskPercent').textContent=fixed(s.usagePercent)+'%';
- el('diskList').innerHTML=disks.length?disks.map(d=>{const badge=d.smartStatus==='Ошибка'?'bad':d.smartStatus==='Исправен'?'':'warn',reason=(d.attentionReasons||[]).join(' · '),tip=reason?' tabindex="0" data-reason="'+safe(reason)+'"':'';return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(d.model)+'</div><div class="disk-meta">'+safe(d.path)+' · SN: '+safe(d.serial)+'</div></div><span class="badge '+badge+'"'+tip+'>'+safe(d.smartStatus)+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Тип</span><b>'+safe(d.type)+'</b></div><div class="disk-stat"><span>Объём</span><b>'+size(d.sizeBytes)+'</b></div><div class="disk-stat"><span>Занято</span><b>'+(d.usedBytes==null?'—':size(d.usedBytes)+' · '+fixed(d.usagePercent)+'%')+'</b></div><div class="disk-stat"><span>Температура</span><b>'+(d.temperatureCelsius==null?'—':fixed(d.temperatureCelsius)+' °C')+'</b></div><div class="disk-stat"><span>Наработка</span><b>'+(d.powerOnHours==null?'—':d.powerOnHours+' ч')+'</b></div><div class="disk-stat"><span>Износ</span><b>'+(d.wearUsedPercent==null?'—':fixed(d.wearUsedPercent)+'%')+'</b></div></div></article>'}).join(''):'<span class="empty">Физические накопители не найдены</span>';
- const totalRate=n.receivePerSec+n.transmitPerSec;el('netInterfaces').textContent=n.interface||'Сетевые интерфейсы';el('netTotal').textContent=(totalRate/1048576).toFixed(2);el('netBar').style.width=Math.min(totalRate/125000000*100,100)+'%';el('netDown').textContent=rate(n.receivePerSec);el('netUp').textContent=rate(n.transmitPerSec);el('dNetDown').textContent=rate(n.receivePerSec);el('dNetUp').textContent=rate(n.transmitPerSec);el('dNetTotal').textContent=rate(totalRate);el('dNetReceived').textContent=size(n.receiveBytes);el('dNetSent').textContent=size(n.transmitBytes);el('dNetInterfaces').textContent=n.interface||'—';
+ el('diskList').innerHTML=disks.length?disks.map(d=>{const badge=d.smartStatus==='Ошибка'?'bad':d.smartStatus==='Исправен'?'':'warn',reason=(d.attentionReasons||[]).join(' · '),tip=reason?' tabindex="0" data-reason="'+safe(reason)+'"':'';return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(d.model)+'</div><div class="disk-meta">'+safe(d.path)+' · SN: '+safe(d.serial)+'</div></div><span class="badge '+badge+'"'+tip+'>'+safe(d.smartStatus)+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Тип / объём</span><b>'+safe(d.type)+' · '+size(d.sizeBytes)+'</b></div><div class="disk-stat"><span>Занято</span><b>'+(d.usedBytes==null?'—':size(d.usedBytes)+' · '+fixed(d.usagePercent)+'%')+'</b></div><div class="disk-stat"><span>Температура / износ</span><b>'+(d.temperatureCelsius==null?'—':fixed(d.temperatureCelsius)+' °C')+' · '+(d.wearUsedPercent==null?'—':fixed(d.wearUsedPercent)+'%')+'</b></div><div class="disk-stat"><span>Reallocated / Pending</span><b>'+d.reallocated+' / '+d.pending+'</b></div><div class="disk-stat"><span>Uncorrectable / CRC</span><b>'+d.uncorrectable+' / '+d.crcErrors+'</b></div><div class="disk-stat"><span>Последний self-test</span><b>'+safe(d.lastSelfTest||'—')+(d.lastSelfTestAt?' · '+new Date(d.lastSelfTestAt).toLocaleDateString('ru-RU'):'')+(d.shortTestStarted?' · запущен новый':'')+'</b></div></div></article>'}).join(''):'<span class="empty">Физические накопители не найдены</span>';
+ const totalRate=n.receivePerSec+n.transmitPerSec,nh=n.health||{};el('netInterfaces').textContent=n.interface||'Сетевые интерфейсы';el('netTotal').textContent=(totalRate/1048576).toFixed(2);el('netBar').style.width=Math.min(totalRate/125000000*100,100)+'%';el('netDown').textContent=rate(n.receivePerSec);el('netUp').textContent=rate(n.transmitPerSec);el('dNetDown').textContent=rate(n.receivePerSec);el('dNetUp').textContent=rate(n.transmitPerSec);el('dNetTotal').textContent=rate(totalRate);el('dNetReceived').textContent=size(n.receiveBytes);el('dNetSent').textContent=size(n.transmitBytes);el('dNetInterfaces').textContent=n.interface||'—';el('dNetGateway').textContent=(nh.gateway||'Не определён')+' · '+(nh.gatewayUp?'доступен':'нет ответа')+(nh.latencyMs?' · '+fixed(nh.latencyMs)+' ms':'');el('dNetDNS').textContent=nh.dnsUp?'Работает':'Ошибка';el('dNetInternet').textContent=nh.internetUp?'Доступен':'Недоступен';el('dNetExternalIP').textContent=(nh.externalIp||'—')+(nh.externalIpChanged?' · изменился':'');
  const pools=z.pools||[],zTotal=pools.reduce((a,p)=>a+p.sizeBytes,0),zUsed=pools.reduce((a,p)=>a+p.allocatedBytes,0),zPercent=zTotal?zUsed/zTotal*100:0,zBad=pools.filter(p=>p.health!=='ONLINE').length;el('zfsCount').textContent=pools.length;el('zfsHealth').textContent=!z.available?'Не найден':zBad?zBad+' требуют внимания':pools.length+' ONLINE';el('zfsUsed').textContent=zTotal?size(zUsed)+' / '+size(zTotal)+' · '+fixed(zPercent)+'%':'—';el('zfsBar').style.width=Math.min(zPercent,100)+'%';
- el('zfsList').innerHTML=pools.length?pools.map(p=>{const bad=p.health!=='ONLINE',reason=(p.attentionReasons||[]).join(' · '),tip=reason?' tabindex="0" data-reason="'+safe(reason)+'"':'';return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(p.name)+'</div><div class="disk-meta">'+safe(p.scan||'Данные scrub отсутствуют')+'</div></div><span class="badge '+(bad?'bad':'')+'"'+tip+'>'+safe(p.health)+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Объём</span><b>'+size(p.sizeBytes)+'</b></div><div class="disk-stat"><span>Занято</span><b>'+size(p.allocatedBytes)+'</b></div><div class="disk-stat"><span>Свободно</span><b>'+size(p.freeBytes)+'</b></div><div class="disk-stat"><span>Заполнение</span><b>'+fixed(p.capacityPercent)+'%</b></div><div class="disk-stat"><span>Фрагментация</span><b>'+fixed(p.fragmentationPercent)+'%</b></div><div class="disk-stat"><span>Dedup</span><b>'+safe(p.dedupRatio)+'</b></div></div></article>'}).join(''):'<span class="empty">'+(z.available?'ZFS-пулы не найдены':'Команда zpool недоступна')+'</span>';
- const val=(v,suffix='')=>v==null?'—':fixed(v)+suffix;el('upsName').textContent=u.available?(u.model||u.name||'Источник бесперебойного питания'):'UPS не настроен';el('upsCharge').textContent=u.chargePercent==null?'—':fixed(u.chargePercent);el('upsBar').style.width=u.chargePercent==null?'0%':Math.min(u.chargePercent,100)+'%';el('upsStatus').textContent=u.available?u.statusLabel:'Не найден';el('upsRuntime').textContent=u.runtimeSeconds==null?'—':uptime(u.runtimeSeconds);el('dUpsName').textContent=u.model||u.name||'Источник бесперебойного питания';el('dUpsStatus').textContent=u.available?u.statusLabel:'Не настроен';el('dUpsCharge').textContent=val(u.chargePercent,'%');el('dUpsRuntime').textContent=u.runtimeSeconds==null?'—':uptime(u.runtimeSeconds);el('dUpsLoad').textContent=val(u.loadPercent,'%');el('dUpsInput').textContent=val(u.inputVoltage,' V');el('dUpsOutput').textContent=val(u.outputVoltage,' V');el('dUpsFrequency').textContent=val(u.inputFrequency,' Hz');el('dUpsBattery').textContent=val(u.batteryVoltage,' V');el('dUpsTemp').textContent=val(u.temperatureCelsius,' °C');el('dUpsSource').textContent=u.available?(u.source+' · '+(u.name||'UPS')):'Установите и настройте NUT или apcupsd';
+ const datasets=z.datasets||[],arc=z.arc||{};el('zfsArcSize').textContent=arc.available?size(arc.sizeBytes):'—';el('zfsArcTarget').textContent=arc.available?size(arc.targetBytes):'—';el('zfsArcHit').textContent=arc.available?fixed(arc.hitRatio)+'%':'—';el('zfsDatasetCount').textContent=datasets.length;el('zfsList').innerHTML=pools.length?pools.map(p=>{const bad=p.health!=='ONLINE'||p.readErrors+p.writeErrors+p.checksumErrors>0,reason=(p.attentionReasons||[]).join(' · '),tip=reason?' tabindex="0" data-reason="'+safe(reason)+'"':'',poolDatasets=datasets.filter(d=>d.pool===p.name),vdevs=p.vdevs||[];return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(p.name)+'</div><div class="disk-meta">Scrub: '+safe(p.scrubStatus||p.scan||'нет данных')+(p.lastScrubAt?' · '+new Date(p.lastScrubAt).toLocaleString('ru-RU'):'')+(p.scrubDuration?' · '+safe(p.scrubDuration):'')+'</div></div><span class="badge '+(bad?'bad':p.scrubTooOld?'warn':'')+'"'+tip+'>'+safe(p.health)+'</span></div><div class="disk-stats"><div class="disk-stat"><span>Объём / свободно</span><b>'+size(p.sizeBytes)+' / '+size(p.freeBytes)+'</b></div><div class="disk-stat"><span>Заполнение</span><b>'+fixed(p.capacityPercent)+'%</b></div><div class="disk-stat"><span>Read / write / checksum</span><b>'+p.readErrors+' / '+p.writeErrors+' / '+p.checksumErrors+'</b></div></div><div class="topology-section"><div class="topology-section-title">Vdev · data / special / cache / log</div><div class="topology-items">'+(vdevs.length?vdevs.map(v=>'<span class="topology-chip '+(v.state==='ONLINE'?'good':'bad')+'">'+safe(v.class)+' · '+safe(v.name)+' · '+safe(v.state)+' · '+v.read+'/'+v.write+'/'+v.checksum+'</span>').join(''):'<span class="empty">Нет данных</span>')+'</div></div><div class="topology-section"><div class="topology-section-title">Datasets</div><div class="topology-items">'+(poolDatasets.length?poolDatasets.map(d=>'<span class="topology-chip '+(d.capacityPercent>=80?'bad':'good')+'">'+safe(d.name)+' · '+fixed(d.capacityPercent)+'% · '+size(d.usedBytes)+'</span>').join(''):'<span class="empty">Нет datasets</span>')+'</div></div></article>'}).join(''):'<span class="empty">'+(z.available?'ZFS-пулы не найдены':'Команда zpool недоступна')+'</span>';
+ const val=(v,suffix='')=>v==null?'—':fixed(v)+suffix;el('upsName').textContent=u.available?(u.model||u.name||'Источник бесперебойного питания'):'UPS не настроен';el('upsCharge').textContent=u.chargePercent==null?'—':fixed(u.chargePercent);el('upsBar').style.width=u.chargePercent==null?'0%':Math.min(u.chargePercent,100)+'%';el('upsStatus').textContent=u.available?u.statusLabel:'Не найден';el('upsRuntime').textContent=u.runtimeSeconds==null?'—':uptime(u.runtimeSeconds);el('dUpsName').textContent=u.model||u.name||'Источник бесперебойного питания';el('dUpsStatus').textContent=u.available?u.statusLabel:'Не настроен';el('dUpsCharge').textContent=val(u.chargePercent,'%');el('dUpsRuntime').textContent=u.runtimeSeconds==null?'—':uptime(u.runtimeSeconds);el('dUpsLoad').textContent=val(u.loadPercent,'%');el('dUpsInput').textContent=val(u.inputVoltage,' V');el('dUpsOutput').textContent=val(u.outputVoltage,' V');el('dUpsFrequency').textContent=val(u.inputFrequency,' Hz');el('dUpsBattery').textContent=val(u.batteryVoltage,' V');el('dUpsTemp').textContent=val(u.temperatureCelsius,' °C');el('dUpsCommunication').textContent=u.communicationOk?'Связь установлена':'Нет связи';el('dUpsBatteryDate').textContent=u.batteryDate||'Не сообщает';el('dUpsBatteryAge').textContent=u.batteryAgeDays?u.batteryAgeDays+' дней':'—';el('dUpsSource').textContent=u.available?(u.source+' · '+(u.name||'UPS')):'Установите и настройте NUT или apcupsd';
+
+ const mh=x.monitor||{},last=mh.lastSuccessfulCollection?new Date(mh.lastSuccessfulCollection):null,stale=!last||Date.now()-last.getTime()>30000,monitorProblem=stale||mh.lastCollectionError||mh.historyWriteError||mh.slowCollector;el('monitorState').textContent=monitorProblem?'Внимание':'OK';el('monitorBar').style.width=monitorProblem?'100%':'12%';el('monitorLast').textContent=last?Math.max(0,Math.round((Date.now()-last.getTime())/1000))+' сек. назад':'—';el('monitorDB').textContent=size(mh.databaseFreeBytes);el('dMonitorLast').textContent=last?last.toLocaleString('ru-RU'):'Ещё не завершён';el('dMonitorDuration').textContent=(mh.collectionDurationMs||0)+' ms';el('dMonitorHistory').textContent=mh.lastHistoryWrite?new Date(mh.lastHistoryWrite).toLocaleString('ru-RU'):'Ещё не записывалась';el('dMonitorDB').textContent=size(mh.databaseFreeBytes);el('dMonitorWatchdog').textContent=mh.watchdogEnabled?'Включён · 30 сек.':'Не активен';el('dMonitorError').textContent=mh.lastCollectionError||mh.historyWriteError||'Нет';
  const guests=(g.guests||[]).filter(v=>!v.template),running=guests.filter(v=>v.status==='running'),guestCores=running.reduce((a,v)=>a+v.maxcpu,0),guestCPU=guestCores?running.reduce((a,v)=>a+v.cpu*v.maxcpu,0)/guestCores*100:0,guestMem=running.reduce((a,v)=>a+v.mem,0),guestMaxMem=running.reduce((a,v)=>a+v.maxmem,0);el('guestRunning').textContent=running.length;el('guestTotal').textContent=' / '+guests.length;el('guestBar').style.width=guests.length?(running.length/guests.length*100)+'%':'0%';el('guestCPU').textContent=g.available?fixed(guestCPU)+'%':'Не найдено';el('guestMemory').textContent=g.available?(size(guestMem)+' / '+size(guestMaxMem)):'—';
  el('guestList').innerHTML=guests.length?guests.map(v=>{const live=v.status==='running',type=v.type==='qemu'?'VM':'LXC',memPercent=v.maxmem?v.mem/v.maxmem*100:0,diskPercent=v.maxdisk?v.disk/v.maxdisk*100:0,ips=(v.ipAddresses||[]).join(', ')||'IP не найден';return '<article class="disk-item"><div class="disk-top"><div><div class="disk-model">'+safe(v.name||('Guest '+v.vmid))+'</div><div class="disk-meta">VMID '+v.vmid+' · '+type+' · '+safe(v.node)+' · '+safe(ips)+'</div></div><span class="badge '+(live?'':'warn')+'">'+(live?'Запущен':'Остановлен')+'</span></div><div class="disk-stats"><div class="disk-stat"><span>CPU</span><b>'+(live?fixed(v.cpu*100)+'%':'—')+' / '+v.maxcpu+'</b></div><div class="disk-stat"><span>Память</span><b>'+(live?size(v.mem)+' · '+fixed(memPercent)+'%':size(v.maxmem))+'</b></div><div class="disk-stat"><span>Диск</span><b>'+(v.maxdisk?size(v.disk)+' · '+fixed(diskPercent)+'%':'—')+'</b></div><div class="disk-stat"><span>Сеть</span><b>↓ '+size(v.netin)+' · ↑ '+size(v.netout)+'</b></div><div class="disk-stat"><span>Guest Agent</span><b>'+(v.agent?'Доступен':'Нет')+'</b></div><div class="disk-stat"><span>Автозапуск</span><b>'+(v.onboot?'Да':'Нет')+'</b></div><div class="disk-stat"><span>Storage</span><b>'+safe((v.storage||[]).join(', ')||'—')+'</b></div><div class="disk-stat"><span>Uptime</span><b>'+(live?uptime(v.uptime):'—')+'</b></div></div></article>'}).join(''):'<span class="empty">'+(g.available?'VM и LXC не найдены':'Команда pvesh недоступна')+'</span>';
  const pveShort=(mt.proxmoxVersion||'Proxmox VE').split('(')[0].trim();el('maintenanceVersion').textContent=pveShort;el('maintenanceUpdates').textContent=mt.availableUpdates??'—';el('maintenanceBar').style.width=Math.min((mt.availableUpdates||0)*5,100)+'%';el('maintenanceReboot').textContent=(mt.rebootRequired||mt.newKernelAvailable)?'Требуется':'Не требуется';el('maintenanceKernel').textContent=mt.kernelVersion||'—';el('dPveVersion').textContent=pveShort;el('dPveVersion').title=mt.proxmoxVersion||'';el('dKernel').textContent=mt.kernelVersion||'—';el('dLatestKernel').textContent=mt.latestInstalledKernel||'—';el('dUpdates').textContent=(mt.availableUpdates??'—')+' пакетов';el('dReboot').textContent=(mt.rebootRequired||mt.newKernelAvailable)?'Требуется':'Не требуется';el('dLastUpdate').textContent=mt.lastSystemUpdate?new Date(mt.lastSystemUpdate).toLocaleString('ru-RU'):'—';el('dNextBackup').textContent=mt.nextBackup||'Расписание не найдено';el('dNextScrub').textContent=mt.nextScrub||'Таймер не найден';el('dMaintenanceChecked').textContent=mt.checkedAt?new Date(mt.checkedAt).toLocaleString('ru-RU'):'—';
@@ -667,7 +718,7 @@ el('backupList').onclick=async event=>{const button=event.target.closest('[data-
 el('alertList').onclick=async event=>{const button=event.target.closest('[data-ack]');if(!button)return;button.disabled=true;const response=await fetch('/api/alerts/acknowledge',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({eventId:Number(button.dataset.ack)})});if(response.ok)button.closest('.alert-item').remove();else button.disabled=false};
 async function loadAlertHistory(){const response=await fetch('/api/alerts/history');if(!response.ok)return;const events=await response.json()||[];el('alertHistory').innerHTML=events.length?events.map(a=>{const state=a.acknowledgedAt?'Принято · '+new Date(a.acknowledgedAt*1000).toLocaleString('ru-RU'):a.resolvedAt?'Завершено · '+new Date(a.resolvedAt*1000).toLocaleString('ru-RU'):'Активно';return '<article class="alert-item '+safe(a.severity)+'"><i class="alert-icon"></i><div><div class="alert-source">'+safe(a.source)+' · '+new Date(a.firstSeen*1000).toLocaleString('ru-RU')+'</div><div class="alert-title">'+safe(a.title)+'</div><div class="alert-message">'+safe(a.message)+'</div><span class="history-state">'+state+'</span></div></article>'}).join(''):'<div class="all-good"><strong>История пока пуста</strong><span>Новые события появятся здесь автоматически.</span></div>'}
 document.querySelectorAll('[data-alert-tab]').forEach(tab=>tab.onclick=()=>{document.querySelectorAll('[data-alert-tab]').forEach(t=>t.classList.toggle('active',t===tab));document.querySelectorAll('[data-alert-panel]').forEach(p=>p.hidden=p.dataset.alertPanel!==tab.dataset.alertTab);if(tab.dataset.alertTab==='history')loadAlertHistory()});
-const cardNames=['alerts','cpu','memory','storage','network','zfs','ups','guests','maintenance','backups','services','system','snapshots','tasks','map'],cardLabels={alerts:'Центр уведомлений',cpu:'CPU',memory:'Память',storage:'Диски',network:'Сеть',zfs:'ZFS',ups:'UPS',guests:'VM и LXC',maintenance:'Proxmox',backups:'Резервные копии',services:'Сервисы',system:'Система и ядро',snapshots:'Snapshots',tasks:'Задачи Proxmox',map:'Карта homelab'};let hiddenCards=[],cardOrder=[];try{hiddenCards=JSON.parse(localStorage.getItem('proxmox-hidden-cards')||'[]');cardOrder=JSON.parse(localStorage.getItem('proxmox-card-order')||'[]')}catch(e){}cardOrder=[...cardOrder.filter(x=>cardNames.includes(x)),...cardNames.filter(x=>!cardOrder.includes(x))];
+const cardNames=['alerts','cpu','memory','storage','network','zfs','ups','monitor','guests','maintenance','backups','services','system','snapshots','tasks','map'],cardLabels={alerts:'Центр уведомлений',cpu:'CPU',memory:'Память',storage:'Диски',network:'Сеть',zfs:'ZFS',ups:'UPS',monitor:'Самоконтроль',guests:'VM и LXC',maintenance:'Proxmox',backups:'Резервные копии',services:'Сервисы',system:'Система и ядро',snapshots:'Snapshots',tasks:'Задачи Proxmox',map:'Карта homelab'};let hiddenCards=[],cardOrder=[];try{hiddenCards=JSON.parse(localStorage.getItem('proxmox-hidden-cards')||'[]');cardOrder=JSON.parse(localStorage.getItem('proxmox-card-order')||'[]')}catch(e){}cardOrder=[...cardOrder.filter(x=>cardNames.includes(x)),...cardNames.filter(x=>!cardOrder.includes(x))];
 function renderCardOrder(){el('cardOrder').innerHTML=cardOrder.map(name=>'<div class="order-item" draggable="true" data-order-card="'+name+'"><span class="order-handle">⠿</span><span>'+cardLabels[name]+'</span><button class="order-button" data-move="up" aria-label="Выше">↑</button><button class="order-button" data-move="down" aria-label="Ниже">↓</button></div>').join('')}
 function saveCardOrder(){localStorage.setItem('proxmox-card-order',JSON.stringify(cardOrder));applyCardSettings();renderCardOrder()}
 function applyCardSettings(){const container=document.querySelector('.cards');cardOrder.forEach(name=>{const card=document.querySelector('[data-card="'+name+'"]');if(card)container.appendChild(card)});cardNames.forEach(name=>{const visible=!hiddenCards.includes(name),card=document.querySelector('[data-card="'+name+'"]'),toggle=document.querySelector('[data-card-toggle="'+name+'"]');if(card)card.hidden=!visible;if(toggle)toggle.checked=visible})}
@@ -682,10 +733,15 @@ async function loadServices(){const response=await fetch('/api/services');if(!re
 function resetServiceForm(){el('serviceId').value='';el('serviceName').value='';el('serviceType').value='http';el('serviceURL').value='';el('serviceLocal').value='';el('serviceCode').value=200;el('serviceText').value='';el('serviceTimeout').value=5;el('serviceEnabled').checked=true}
 el('saveService').onclick=async()=>{const value={id:Number(el('serviceId').value)||0,name:el('serviceName').value.trim(),checkType:el('serviceType').value,url:el('serviceURL').value.trim(),localAddress:el('serviceLocal').value.trim(),expectedCode:Number(el('serviceCode').value)||200,expectedText:el('serviceText').value,timeoutSeconds:Number(el('serviceTimeout').value)||5,enabled:el('serviceEnabled').checked},status=el('serviceStatus');status.textContent='Сохранение…';const response=await fetch('/api/services',{method:value.id?'PUT':'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify(value)});status.textContent=response.ok?'Сервис сохранён. Первая проверка появится в течение 15 секунд.':await response.text();if(response.ok){resetServiceForm();loadServices()}};
 el('serviceSettingsList').onclick=async event=>{const edit=event.target.closest('[data-service-edit]'),del=event.target.closest('[data-service-delete]');if(edit){const v=configuredServices.find(x=>x.id===Number(edit.dataset.serviceEdit));if(!v)return;el('serviceId').value=v.id;el('serviceName').value=v.name;el('serviceType').value=v.checkType;el('serviceURL').value=v.url;el('serviceLocal').value=v.localAddress||'';el('serviceCode').value=v.expectedCode;el('serviceText').value=v.expectedText||'';el('serviceTimeout').value=v.timeoutSeconds;el('serviceEnabled').checked=v.enabled;return}if(del&&confirm('Удалить сервис и его историю?')){const response=await fetch('/api/services?id='+del.dataset.serviceDelete,{method:'DELETE'});if(response.ok)loadServices()}};loadServices().catch(()=>{});
+function upsShutdownForm(){return {enabled:el('upsShutdownEnabled').checked,chargePercent:Number(el('upsShutdownCharge').value),runtimeSeconds:Number(el('upsShutdownRuntime').value),graceSeconds:Number(el('upsShutdownGrace').value),guestOrder:el('upsShutdownOrder').value.split(',').map(v=>Number(v.trim())).filter(v=>v>0)}}
+async function loadUPSShutdown(){const response=await fetch('/api/settings/ups-shutdown');if(!response.ok)return;const s=await response.json();el('upsShutdownEnabled').checked=!!s.enabled;el('upsShutdownCharge').value=s.chargePercent||15;el('upsShutdownRuntime').value=s.runtimeSeconds||600;el('upsShutdownGrace').value=s.graceSeconds||120;el('upsShutdownOrder').value=(s.guestOrder||[]).join(', ')}
+el('saveUPSShutdown').onclick=async()=>{const status=el('upsShutdownStatus');status.textContent='Сохранение…';const response=await fetch('/api/settings/ups-shutdown',{method:'PUT',headers:{'Content-Type':'application/json'},body:JSON.stringify(upsShutdownForm())});status.textContent=response.ok?(upsShutdownForm().enabled?'Сценарий сохранён и разрешён.':'Сценарий сохранён, автоматическое выполнение отключено.'):await response.text()};
+el('dryRunUPSShutdown').onclick=async()=>{const saved=await fetch('/api/settings/ups-shutdown',{method:'PUT',headers:{'Content-Type':'application/json'},body:JSON.stringify(upsShutdownForm())});if(!saved.ok){el('upsShutdownStatus').textContent=await saved.text();return}const response=await fetch('/api/ups/shutdown-dry-run',{method:'POST'}),steps=await response.json();el('upsShutdownPlan').innerHTML=(steps||[]).map(v=>'<div class="shutdown-step"><b>'+v.order+'. '+safe(v.target)+'</b><div class="disk-meta">'+safe(v.action)+' · '+safe(v.detail)+'</div></div>').join('');el('upsShutdownStatus').textContent='Dry-run завершён: ни одна VM и хост не были остановлены.'};loadUPSShutdown().catch(()=>{});
 function emailForm(){return {enabled:el('emailEnabled').checked,skipTlsVerify:el('emailSkipTLS').checked,host:el('emailHost').value.trim(),port:Number(el('emailPort').value),username:el('emailUsername').value.trim(),password:el('emailPassword').value,from:el('emailFrom').value.trim(),to:el('emailTo').value.trim(),intervalMinutes:Number(el('emailInterval').value),severities:[...document.querySelectorAll('[data-email-severity]:checked')].map(x=>x.dataset.emailSeverity),sources:[...document.querySelectorAll('[data-email-source]:checked')].map(x=>x.dataset.emailSource)}}
 async function loadEmail(){const response=await fetch('/api/settings/email');if(!response.ok)return;const s=await response.json();el('emailEnabled').checked=s.enabled;el('emailSkipTLS').checked=!!s.skipTlsVerify;el('emailHost').value=s.host||'';el('emailPort').value=s.port||587;el('emailUsername').value=s.username||'';el('emailPassword').value='';el('emailPassword').placeholder=s.hasPassword?'Сохранён · не изменять':'Пароль';el('emailFrom').value=s.from||'';el('emailTo').value=s.to||'';el('emailInterval').value=s.intervalMinutes||60;document.querySelectorAll('[data-email-severity]').forEach(x=>x.checked=(s.severities||[]).includes(x.dataset.emailSeverity));document.querySelectorAll('[data-email-source]').forEach(x=>x.checked=(s.sources||[]).includes(x.dataset.emailSource))}
 el('saveEmail').onclick=async()=>{const status=el('emailStatus');status.textContent='Сохранение…';const response=await fetch('/api/settings/email',{method:'PUT',headers:{'Content-Type':'application/json'},body:JSON.stringify(emailForm())});status.textContent=response.ok?'Настройки сохранены.':await response.text();if(response.ok)loadEmail()};
 el('testEmail').onclick=async()=>{const status=el('emailStatus');status.textContent='Отправляем тестовое письмо…';const saved=await fetch('/api/settings/email',{method:'PUT',headers:{'Content-Type':'application/json'},body:JSON.stringify(emailForm())});if(!saved.ok){status.textContent=await saved.text();return}const response=await fetch('/api/settings/email/test',{method:'POST'});status.textContent=response.ok?'Тестовое письмо отправлено.':await response.text()};loadEmail().catch(()=>{});
 el('installUpdate').onclick=async()=>{const button=el('installUpdate');button.disabled=true;button.textContent='Ставим в очередь…';el('updateMessage').textContent='Создаём запрос на обновление…';try{const response=await fetch('/api/update/request',{method:'POST'});if(!response.ok)throw new Error(await response.text());const status=await response.json();el('updateMessage').textContent=status.message;button.textContent='Обновление выполняется…'}catch(e){button.disabled=false;button.textContent='Попробовать ещё раз';el('updateMessage').textContent='Не удалось запустить обновление.'}};
 const events=new EventSource('/api/events');events.onmessage=e=>render(JSON.parse(e.data));events.onerror=()=>{el('dot').classList.remove('live');el('status').textContent='Переподключение…'};
+setInterval(()=>{if(lastDashboardEvent&&Date.now()-lastDashboardEvent>15000){el('dot').classList.remove('live');el('status').textContent='Данные устарели · переподключение…';el('monitorState').textContent='STALE';el('monitorBar').style.width='100%'}},3000);
 </script></body></html>`

zfs.go

@@ -2,7 +2,9 @@ package main
 
 import (
 	"context"
+	"os"
 	"os/exec"
+	"regexp"
 	"strconv"
 	"strings"
 	"sync"
@@ -10,28 +12,66 @@ import (
 )
 
 type ZFSPoolMetrics struct {
-	Name             string   `json:"name"`
-	Health           string   `json:"health"`
-	SizeBytes        uint64   `json:"sizeBytes"`
-	AllocatedBytes   uint64   `json:"allocatedBytes"`
-	FreeBytes        uint64   `json:"freeBytes"`
-	CapacityPercent  float64  `json:"capacityPercent"`
-	Fragmentation    float64  `json:"fragmentationPercent"`
-	DedupRatio       string   `json:"dedupRatio"`
-	Scan             string   `json:"scan"`
-	Errors           string   `json:"errors"`
-	AttentionReasons []string `json:"attentionReasons"`
+	Name             string    `json:"name"`
+	Health           string    `json:"health"`
+	SizeBytes        uint64    `json:"sizeBytes"`
+	AllocatedBytes   uint64    `json:"allocatedBytes"`
+	FreeBytes        uint64    `json:"freeBytes"`
+	CapacityPercent  float64   `json:"capacityPercent"`
+	Fragmentation    float64   `json:"fragmentationPercent"`
+	DedupRatio       string    `json:"dedupRatio"`
+	Scan             string    `json:"scan"`
+	Errors           string    `json:"errors"`
+	AttentionReasons []string  `json:"attentionReasons"`
+	ReadErrors       uint64    `json:"readErrors"`
+	WriteErrors      uint64    `json:"writeErrors"`
+	ChecksumErrors   uint64    `json:"checksumErrors"`
+	VDevs            []ZFSVDev `json:"vdevs"`
+	ScrubStatus      string    `json:"scrubStatus"`
+	LastScrubAt      string    `json:"lastScrubAt"`
+	ScrubDuration    string    `json:"scrubDuration"`
+	ScrubTooOld      bool      `json:"scrubTooOld"`
+	ErrorChanges     []string  `json:"errorChanges"`
+}
+
+type ZFSVDev struct {
+	Name     string `json:"name"`
+	Class    string `json:"class"`
+	State    string `json:"state"`
+	Read     uint64 `json:"read"`
+	Write    uint64 `json:"write"`
+	Checksum uint64 `json:"checksum"`
+}
+type ZFSDataset struct {
+	Name            string  `json:"name"`
+	Pool            string  `json:"pool"`
+	UsedBytes       uint64  `json:"usedBytes"`
+	AvailableBytes  uint64  `json:"availableBytes"`
+	ReferencedBytes uint64  `json:"referencedBytes"`
+	Mountpoint      string  `json:"mountpoint"`
+	CapacityPercent float64 `json:"capacityPercent"`
+}
+type ZFSARC struct {
+	Available   bool    `json:"available"`
+	SizeBytes   uint64  `json:"sizeBytes"`
+	TargetBytes uint64  `json:"targetBytes"`
+	HitRatio    float64 `json:"hitRatio"`
+	Hits        uint64  `json:"hits"`
+	Misses      uint64  `json:"misses"`
 }
 
 type ZFSMetrics struct {
 	Available bool             `json:"available"`
 	Pools     []ZFSPoolMetrics `json:"pools"`
+	Datasets  []ZFSDataset     `json:"datasets"`
+	ARC       ZFSARC           `json:"arc"`
 }
 
 type zfsCollector struct {
 	mu        sync.Mutex
 	cached    ZFSMetrics
 	updatedAt time.Time
+	previous  map[string]ZFSPoolMetrics
 }
 
 func (c *zfsCollector) collect() ZFSMetrics {
@@ -40,7 +80,26 @@ func (c *zfsCollector) collect() ZFSMetrics {
 	if time.Since(c.updatedAt) < 30*time.Second {
 		return c.cached
 	}
-	c.cached = readZFSPools()
+	current := readZFSPools()
+	if c.previous == nil {
+		c.previous = map[string]ZFSPoolMetrics{}
+	}
+	for i := range current.Pools {
+		pool := &current.Pools[i]
+		if old, ok := c.previous[pool.Name]; ok {
+			checks := []struct {
+				label       string
+				before, now uint64
+			}{{"read", old.ReadErrors, pool.ReadErrors}, {"write", old.WriteErrors, pool.WriteErrors}, {"checksum", old.ChecksumErrors, pool.ChecksumErrors}}
+			for _, check := range checks {
+				if check.now > check.before {
+					pool.ErrorChanges = append(pool.ErrorChanges, check.label+": "+strconv.FormatUint(check.before, 10)+" → "+strconv.FormatUint(check.now, 10))
+				}
+			}
+		}
+		c.previous[pool.Name] = *pool
+	}
+	c.cached = current
 	c.updatedAt = time.Now()
 	return c.cached
 }
@@ -67,15 +126,35 @@ func readZFSPools() ZFSMetrics {
 		pool.FreeBytes, _ = strconv.ParseUint(fields[3], 10, 64)
 		pool.Fragmentation = parsePercent(fields[4])
 		pool.CapacityPercent = parsePercent(fields[5])
-		pool.Scan, pool.Errors = readZFSStatus(pool.Name)
+		pool.Scan, pool.Errors, pool.VDevs = readZFSStatus(pool.Name)
+		for _, vdev := range pool.VDevs {
+			pool.ReadErrors += vdev.Read
+			pool.WriteErrors += vdev.Write
+			pool.ChecksumErrors += vdev.Checksum
+			if vdev.State != "ONLINE" {
+				pool.AttentionReasons = append(pool.AttentionReasons, vdev.Class+" "+vdev.Name+": "+vdev.State)
+			}
+		}
+		pool.ScrubStatus, pool.LastScrubAt, pool.ScrubDuration, pool.ScrubTooOld = parseScrub(pool.Scan)
 		if pool.Health != "ONLINE" {
 			pool.AttentionReasons = append(pool.AttentionReasons, "Состояние пула: "+pool.Health)
 		}
 		if pool.Errors != "" && !strings.Contains(pool.Errors, "No known data errors") {
 			pool.AttentionReasons = append(pool.AttentionReasons, pool.Errors)
 		}
+		if pool.ReadErrors+pool.WriteErrors+pool.ChecksumErrors > 0 {
+			pool.AttentionReasons = append(pool.AttentionReasons, "Ошибки vdev: read="+strconv.FormatUint(pool.ReadErrors, 10)+", write="+strconv.FormatUint(pool.WriteErrors, 10)+", checksum="+strconv.FormatUint(pool.ChecksumErrors, 10))
+		}
+		if pool.ScrubTooOld {
+			pool.AttentionReasons = append(pool.AttentionReasons, "Scrub не выполнялся более 35 дней")
+		}
+		if pool.CapacityPercent >= 80 {
+			pool.AttentionReasons = append(pool.AttentionReasons, "Мало свободного места для безопасной CoW-работы ZFS")
+		}
 		result.Pools = append(result.Pools, pool)
 	}
+	result.Datasets = readZFSDatasets()
+	result.ARC = readZFSARC()
 	return result
 }
 
@@ -84,11 +163,13 @@ func parsePercent(value string) float64 {
 	return parsed
 }
 
-func readZFSStatus(pool string) (string, string) {
+func readZFSStatus(pool string) (string, string, []ZFSVDev) {
 	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
 	defer cancel()
 	output, _ := exec.CommandContext(ctx, "zpool", "status", pool).Output()
-	var scan, errors string
+	var scan, errors, currentClass string
+	var vdevs []ZFSVDev
+	inConfig := false
 	for _, line := range strings.Split(string(output), "\n") {
 		trimmed := strings.TrimSpace(line)
 		if strings.HasPrefix(trimmed, "scan:") {
@@ -97,6 +178,98 @@ func readZFSStatus(pool string) (string, string) {
 		if strings.HasPrefix(trimmed, "errors:") {
 			errors = strings.TrimSpace(strings.TrimPrefix(trimmed, "errors:"))
 		}
+		if strings.HasPrefix(trimmed, "config:") {
+			inConfig = true
+			continue
+		}
+		if !inConfig || trimmed == "" || strings.HasPrefix(trimmed, "NAME") {
+			continue
+		}
+		fields := strings.Fields(trimmed)
+		if len(fields) == 1 && (fields[0] == "logs" || fields[0] == "cache" || fields[0] == "spares" || fields[0] == "special") {
+			currentClass = fields[0]
+			continue
+		}
+		if len(fields) >= 5 && (fields[1] == "ONLINE" || fields[1] == "DEGRADED" || fields[1] == "OFFLINE" || fields[1] == "UNAVAIL" || fields[1] == "FAULTED" || fields[1] == "REMOVED") {
+			read, _ := strconv.ParseUint(fields[2], 10, 64)
+			write, _ := strconv.ParseUint(fields[3], 10, 64)
+			checksum, _ := strconv.ParseUint(fields[4], 10, 64)
+			class := currentClass
+			if class == "" {
+				class = "data"
+			}
+			vdevs = append(vdevs, ZFSVDev{Name: fields[0], Class: class, State: fields[1], Read: read, Write: write, Checksum: checksum})
+		}
 	}
-	return scan, errors
+	return scan, errors, vdevs
+}
+
+var scrubDatePattern = regexp.MustCompile(`(?:on|since) (\w{3} \w{3}\s+\d+ \d\d:\d\d:\d\d \d{4})`)
+
+func parseScrub(scan string) (status, lastAt, duration string, tooOld bool) {
+	lower := strings.ToLower(scan)
+	status = scan
+	if strings.Contains(lower, "in progress") {
+		status = "Выполняется"
+	} else if strings.Contains(lower, "repaired") {
+		status = "Завершён"
+	} else if strings.Contains(lower, "none requested") || scan == "" {
+		status = "Никогда не запускался"
+		return status, "", "", true
+	}
+	if match := scrubDatePattern.FindStringSubmatch(scan); len(match) > 1 {
+		if t, err := time.Parse("Mon Jan 2 15:04:05 2006", match[1]); err == nil {
+			lastAt = t.Format(time.RFC3339)
+			tooOld = time.Since(t) > 35*24*time.Hour
+		}
+	}
+	if parts := strings.Split(scan, " in "); len(parts) > 1 {
+		duration = strings.Fields(parts[1])[0]
+	}
+	return
+}
+
+func readZFSDatasets() []ZFSDataset {
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+	out, err := exec.CommandContext(ctx, "zfs", "list", "-H", "-p", "-o", "name,used,avail,refer,mountpoint").Output()
+	if err != nil {
+		return nil
+	}
+	var result []ZFSDataset
+	for _, line := range strings.Split(strings.TrimSpace(string(out)), "\n") {
+		f := strings.Fields(line)
+		if len(f) < 5 {
+			continue
+		}
+		used, _ := strconv.ParseUint(f[1], 10, 64)
+		avail, _ := strconv.ParseUint(f[2], 10, 64)
+		refer, _ := strconv.ParseUint(f[3], 10, 64)
+		pct := 0.0
+		if used+avail > 0 {
+			pct = float64(used) / float64(used+avail) * 100
+		}
+		result = append(result, ZFSDataset{Name: f[0], Pool: strings.Split(f[0], "/")[0], UsedBytes: used, AvailableBytes: avail, ReferencedBytes: refer, Mountpoint: strings.Join(f[4:], " "), CapacityPercent: pct})
+	}
+	return result
+}
+
+func readZFSARC() ZFSARC {
+	data, err := os.ReadFile("/proc/spl/kstat/zfs/arcstats")
+	if err != nil {
+		return ZFSARC{}
+	}
+	values := map[string]uint64{}
+	for _, line := range strings.Split(string(data), "\n") {
+		f := strings.Fields(line)
+		if len(f) >= 3 {
+			values[f[0]], _ = strconv.ParseUint(f[2], 10, 64)
+		}
+	}
+	hits, misses := values["hits"], values["misses"]
+	ratio := 0.0
+	if hits+misses > 0 {
+		ratio = float64(hits) / float64(hits+misses) * 100
+	}
+	return ZFSARC{Available: true, SizeBytes: values["size"], TargetBytes: values["c"], HitRatio: ratio, Hits: hits, Misses: misses}
 }