fix: fix container list

* refactor: select bind

* refactor: simplify table selection state handling

---------

Co-authored-by: ssongliu <sloooop1x@gmail.com>

agent/app/api/v2/container.go

@@ -28,7 +28,7 @@ func (b *BaseApi) SearchContainer(c *gin.Context) {
 		return
 	}
 
-	total, list, err := containerService.Page(req)
+	total, list, err := containerService.Page(c.Request.Context(), req)
 	if err != nil {
 		helper.InternalServer(c, err)
 		return
@@ -249,9 +249,10 @@ func (b *BaseApi) ListContainerByImage(c *gin.Context) {
 // @Success 200 {object} dto.ContainerStatus
 // @Security ApiKeyAuth
 // @Security Timestamp
+// @Param containersOnly query boolean false "Only count containers"
 // @Router /containers/status [get]
 func (b *BaseApi) LoadContainerStatus(c *gin.Context) {
-	data, err := containerService.LoadStatus()
+	data, err := containerService.LoadStatus(c.Request.Context(), c.Query("containersOnly") == "true")
 	if err != nil {
 		helper.InternalServer(c, err)
 		return
@@ -415,9 +416,14 @@ func (b *BaseApi) LoadResourceLimit(c *gin.Context) {
 // @Success 200 {array} dto.ContainerListStats
 // @Security ApiKeyAuth
 // @Security Timestamp
+// @Param ids query string false "Comma-separated container IDs; omitted selects all containers"
 // @Router /containers/list/stats [get]
 func (b *BaseApi) ContainerListStats(c *gin.Context) {
-	data, err := containerService.ContainerListStats()
+	var ids []string
+	if _, supplied := c.Request.URL.Query()["ids"]; supplied {
+		ids = strings.FieldsFunc(c.Query("ids"), func(r rune) bool { return r == ',' })
+	}
+	data, err := containerService.ContainerListStats(c.Request.Context(), ids)
 	if err != nil {
 		helper.InternalServer(c, err)
 		return
@@ -964,3 +970,12 @@ func (b *BaseApi) ContainerStreamLogs(c *gin.Context) {
 
 	containerService.StreamLogs(c, streamLog)
 }
+
+func (b *BaseApi) CleanNetworks(c *gin.Context) {
+	result, err := containerService.CleanNetworks()
+	if err != nil {
+		helper.InternalServer(c, err)
+		return
+	}
+	helper.SuccessWithData(c, result)
+}

agent/app/api/v2/monitor.go

@@ -1,13 +1,9 @@
 package v2
 
 import (
-	"sort"
-
 	"github.com/1Panel-dev/1Panel/agent/app/api/v2/helper"
 	"github.com/1Panel-dev/1Panel/agent/app/dto"
 	"github.com/gin-gonic/gin"
-	"github.com/shirou/gopsutil/v4/disk"
-	"github.com/shirou/gopsutil/v4/net"
 )
 
 // @Tags Monitor
@@ -32,14 +28,19 @@ func (b *BaseApi) LoadMonitor(c *gin.Context) {
 }
 
 // @Tags Monitor
-// @Summary Clean monitor data
+// @Summary Clean host or GPU monitor data
+// @Param request body dto.MonitorClean true "request"
 // @Success 200
 // @Security ApiKeyAuth
 // @Security Timestamp
 // @Router /hosts/monitor/clean [post]
-// @x-panel-log {"bodyKeys":[],"paramKeys":[],"BeforeFunctions":[],"formatZH":"清空监控数据","formatEN":"clean monitor datas"}
+// @x-panel-log {"bodyKeys":["type"],"paramKeys":[],"BeforeFunctions":[],"formatZH":"清空 [type] 监控数据","formatEN":"clean [type] monitoring data"}
 func (b *BaseApi) CleanMonitor(c *gin.Context) {
-	if err := monitorService.CleanData(); err != nil {
+	var req dto.MonitorClean
+	if err := helper.CheckBindAndValidate(&req, c); err != nil {
+		return
+	}
+	if err := monitorService.CleanData(req.Type); err != nil {
 		helper.InternalServer(c, err)
 		return
 	}
@@ -91,14 +92,7 @@ func (b *BaseApi) UpdateMonitorSetting(c *gin.Context) {
 // @Security Timestamp
 // @Router /hosts/monitor/netoptions [get]
 func (b *BaseApi) GetNetworkOptions(c *gin.Context) {
-	netStat, _ := net.IOCounters(true)
-	var options []string
-	options = append(options, "all")
-	for _, net := range netStat {
-		options = append(options, net.Name)
-	}
-	sort.Strings(options)
-	helper.SuccessWithData(c, options)
+	helper.SuccessWithData(c, monitorService.LoadNetworkOptions())
 }
 
 // @Tags Monitor
@@ -108,12 +102,5 @@ func (b *BaseApi) GetNetworkOptions(c *gin.Context) {
 // @Security Timestamp
 // @Router /hosts/monitor/iooptions [get]
 func (b *BaseApi) GetIOOptions(c *gin.Context) {
-	diskStat, _ := disk.IOCounters()
-	var options []string
-	options = append(options, "all")
-	for _, net := range diskStat {
-		options = append(options, net.Name)
-	}
-	sort.Strings(options)
-	helper.SuccessWithData(c, options)
+	helper.SuccessWithData(c, monitorService.LoadIOOptions())
 }

agent/app/api/v2/monitor_vllm.go

@@ -0,0 +1,45 @@
+package v2
+
+import (
+	"github.com/1Panel-dev/1Panel/agent/app/api/v2/helper"
+	"github.com/1Panel-dev/1Panel/agent/app/dto"
+	"github.com/gin-gonic/gin"
+)
+
+func (b *BaseApi) LoadVLLMMonitor(c *gin.Context) {
+	var req dto.MonitorVLLMSearch
+	if err := helper.CheckBindAndValidate(&req, c); err != nil {
+		return
+	}
+	data, err := monitorService.LoadVLLMMonitorData(req)
+	if err != nil {
+		helper.InternalServer(c, err)
+		return
+	}
+	helper.SuccessWithData(c, data)
+}
+
+func (b *BaseApi) LoadVLLMCurrent(c *gin.Context) {
+	var req dto.MonitorVLLMCurrent
+	if err := helper.CheckBindAndValidate(&req, c); err != nil {
+		return
+	}
+	data, err := monitorService.LoadVLLMCurrent(c.Request.Context(), req)
+	if err != nil {
+		helper.InternalServer(c, err)
+		return
+	}
+	helper.SuccessWithData(c, data)
+}
+
+func (b *BaseApi) CleanVLLMMonitor(c *gin.Context) {
+	var req dto.MonitorVLLMClean
+	if err := helper.CheckBindAndValidate(&req, c); err != nil {
+		return
+	}
+	if err := monitorService.CleanVLLMMonitor(req); err != nil {
+		helper.InternalServer(c, err)
+		return
+	}
+	helper.Success(c)
+}

agent/app/api/v2/runtime.go

@@ -169,7 +169,7 @@ func (b *BaseApi) GetNodePackageRunScript(c *gin.Context) {
 // @Security ApiKeyAuth
 // @Security Timestamp
 // @Router /runtimes/operate [post]
-// @x-panel-log {"bodyKeys":["id"],"paramKeys":[],"BeforeFunctions":[],"formatZH":"操作运行环境 [id]","formatEN":"Operate runtime [id]"}
+// @x-panel-log {"bodyKeys":["ID"],"paramKeys":[],"BeforeFunctions":[{"input_column":"id","input_value":"ID","isList":false,"db":"runtimes","output_column":"name","output_value":"name"}],"formatZH":"操作运行环境 [name]","formatEN":"Operate runtime [name]"}
 func (b *BaseApi) OperateRuntime(c *gin.Context) {
 	var req request.RuntimeOperate
 	if err := helper.CheckBindAndValidate(&req, c); err != nil {

agent/app/api/v2/ssh.go

@@ -83,6 +83,7 @@ func (b *BaseApi) CreateRootCert(c *gin.Context) {
 	}
 	if err := loadCertAfterDecrypt(&req); err != nil {
 		helper.BadRequest(c, err)
+		return
 	}
 	if err := sshService.CreateRootCert(req); err != nil {
 		helper.InternalServer(c, err)
@@ -107,6 +108,7 @@ func (b *BaseApi) EditRootCert(c *gin.Context) {
 	}
 	if err := loadCertAfterDecrypt(&req); err != nil {
 		helper.BadRequest(c, err)
+		return
 	}
 	if err := sshService.EditRootCert(req); err != nil {
 		helper.InternalServer(c, err)

agent/app/dto/alert.go

@@ -21,6 +21,7 @@ type AlertBase struct {
 }
 
 type PushAlert struct {
+	Result    string `json:"result,omitempty"`
 	TaskName  string `json:"taskName"`
 	AlertType string `json:"alertType"`
 	EntryID   uint   `json:"entryID"`
@@ -53,6 +54,7 @@ type AlertDTO struct {
 	Method         string    `json:"method"`
 	Title          string    `json:"title"`
 	Project        string    `json:"project"`
+	TaskName       string    `json:"taskName,omitempty"`
 	Status         string    `json:"status"`
 	SendCount      uint      `json:"sendCount"`
 	AdvancedParams string    `json:"advancedParams"`

agent/app/dto/container_network_cleanup.go

@@ -0,0 +1,17 @@
+package dto
+
+type NetworkCleanupReport struct {
+	Deleted []NetworkCleanupItem `json:"deleted"`
+	Skipped []NetworkCleanupItem `json:"skipped"`
+	Failed  []NetworkCleanupItem `json:"failed"`
+}
+
+type NetworkCleanupItem struct {
+	ID     string `json:"id"`
+	Name   string `json:"name"`
+	Reason string `json:"reason,omitempty"`
+}
+
+type NetworkCleanupTask struct {
+	TaskID string `json:"taskID"`
+}

agent/app/dto/cronjob.go

@@ -51,9 +51,10 @@ type CronjobOperate struct {
 	Secret            string `json:"secret"`
 	Args              string `json:"args"`
 
-	AlertCount  uint   `json:"alertCount"`
-	AlertTitle  string `json:"alertTitle"`
-	AlertMethod string `json:"alertMethod"`
+	AlertCount       uint   `json:"alertCount"`
+	AlertTitle       string `json:"alertTitle"`
+	AlertMethod      string `json:"alertMethod"`
+	AlertTriggerMode string `json:"alertTriggerMode" validate:"omitempty,oneof=failed success both"`
 
 	CleanLogConfig
 }
@@ -126,7 +127,8 @@ type CronjobInfo struct {
 	Secret           string `json:"secret"`
 	Args             string `json:"args"`
 
-	AlertCount uint `json:"alertCount"`
+	AlertCount       uint   `json:"alertCount"`
+	AlertTriggerMode string `json:"alertTriggerMode"`
 }
 
 type CronjobImport struct {
@@ -169,9 +171,10 @@ type CronjobTrans struct {
 	SourceAccounts  []string `json:"sourceAccounts"`
 	DownloadAccount string   `json:"downloadAccount"`
 
-	AlertCount  uint   `json:"alertCount"`
-	AlertTitle  string `json:"alertTitle"`
-	AlertMethod string `json:"alertMethod"`
+	AlertCount       uint   `json:"alertCount"`
+	AlertTitle       string `json:"alertTitle"`
+	AlertMethod      string `json:"alertMethod"`
+	AlertTriggerMode string `json:"alertTriggerMode" validate:"omitempty,oneof=failed success both"`
 }
 type TransHelper struct {
 	Name       string `json:"name"`

agent/app/dto/monitor.go

@@ -1,6 +1,10 @@
 package dto
 
-import "time"
+import (
+	"time"
+
+	"github.com/1Panel-dev/1Panel/agent/app/model"
+)
 
 type MonitorSearch struct {
 	Param     string    `json:"param" validate:"required,oneof=all cpu memory load io network"`
@@ -26,24 +30,33 @@ type Process struct {
 }
 
 type MonitorSetting struct {
-	MonitorStatus    string `json:"monitorStatus"`
-	MonitorStoreDays string `json:"monitorStoreDays"`
-	MonitorInterval  string `json:"monitorInterval"`
-	DefaultNetwork   string `json:"defaultNetwork"`
-	DefaultIO        string `json:"defaultIO"`
+	GPUMonitorStatus     string `json:"gpuMonitorStatus"`
+	GPUMonitorStoreDays  string `json:"gpuMonitorStoreDays"`
+	GPUMonitorInterval   string `json:"gpuMonitorInterval"`
+	VLLMMonitorStatus    string `json:"vllmMonitorStatus"`
+	VLLMMonitorStoreDays string `json:"vllmMonitorStoreDays"`
+	VLLMMonitorInterval  string `json:"vllmMonitorInterval"`
+	MonitorStatus        string `json:"monitorStatus"`
+	MonitorStoreDays     string `json:"monitorStoreDays"`
+	MonitorInterval      string `json:"monitorInterval"`
+	DefaultNetwork       string `json:"defaultNetwork"`
+	DefaultIO            string `json:"defaultIO"`
 }
 
 type MonitorSettingUpdate struct {
-	Key   string `json:"key" validate:"required,oneof=MonitorStatus MonitorStoreDays MonitorInterval DefaultNetwork DefaultIO"`
+	Key   string `json:"key" validate:"required,oneof=MonitorStatus MonitorStoreDays MonitorInterval GPUMonitorStatus GPUMonitorStoreDays GPUMonitorInterval VLLMMonitorStatus VLLMMonitorStoreDays VLLMMonitorInterval DefaultNetwork DefaultIO"`
 	Value string `json:"value"`
 }
 
 type MonitorGPUOptions struct {
+	Supported bool           `json:"supported"`
 	GPUType   string         `json:"gpuType"`
 	ChartHide []GPUChartHide `json:"chartHide"`
 	Options   []string       `json:"options"`
 }
 type GPUChartHide struct {
+	DeviceID    string `json:"deviceID"`
+	Legacy      bool   `json:"legacy"`
 	ProductName string `json:"productName"`
 	Type        string `json:"type"`
 	Process     bool   `json:"process"`
@@ -55,23 +68,54 @@ type GPUChartHide struct {
 	Speed       bool   `json:"speed"`
 }
 type MonitorGPUSearch struct {
+	Aggregation string    `json:"aggregation" validate:"omitempty,oneof=avg max"`
+	DeviceID    string    `json:"deviceID"`
+	Legacy      bool      `json:"legacy"`
 	ProductName string    `json:"productName"`
 	StartTime   time.Time `json:"startTime"`
 	EndTime     time.Time `json:"endTime"`
 }
 type MonitorGPUData struct {
-	Date             []time.Time `json:"date"`
-	GPUValue         []float64   `json:"gpuValue"`
-	TemperatureValue []float64   `json:"temperatureValue"`
-	PowerTotal       []float64   `json:"powerTotal"`
-	PowerUsed        []float64   `json:"powerUsed"`
-	PowerPercent     []float64   `json:"powerPercent"`
-	MemoryTotal      []float64   `json:"memoryTotal"`
-	MemoryUsed       []float64   `json:"memoryUsed"`
-	MemoryPercent    []float64   `json:"memoryPercent"`
-	SpeedValue       []int       `json:"speedValue"`
+	MemoryActivity     []*float64 `json:"memoryActivity"`
+	EncoderUtil        []*float64 `json:"encoderUtil"`
+	DecoderUtil        []*float64 `json:"decoderUtil"`
+	JPEGUtil           []*float64 `json:"jpegUtil"`
+	OFAUtil            []*float64 `json:"ofaUtil"`
+	MediaUtil          []*float64 `json:"mediaUtil"`
+	ComputeUtil        []*float64 `json:"computeUtil"`
+	CopyUtil           []*float64 `json:"copyUtil"`
+	HotspotTemperature []*float64 `json:"hotspotTemperature"`
+	FanRPM             []*float64 `json:"fanRPM"`
+	AICPUUtil          []*float64 `json:"aiCPUUtil"`
+	CtrlCPUUtil        []*float64 `json:"ctrlCPUUtil"`
+	DDRUsed            []*float64 `json:"ddrUsed"`
+	DDRTotal           []*float64 `json:"ddrTotal"`
+	HBMUsed            []*float64 `json:"hbmUsed"`
+	HBMTotal           []*float64 `json:"hbmTotal"`
+	DDRBandwidth       []*float64 `json:"ddrBandwidth"`
+	HBMBandwidth       []*float64 `json:"hbmBandwidth"`
+	MemoryBandwidth    []*float64 `json:"memoryBandwidth"`
+	MediaFrequency     []*float64 `json:"mediaFrequency"`
+	HugepagesUsed      []*float64 `json:"hugepagesUsed"`
+	HugepagesTotal     []*float64 `json:"hugepagesTotal"`
 
-	ProcessCount []int          `json:"processCount"`
+	BucketSeconds          int64       `json:"bucketSeconds"`
+	SampleCount            int64       `json:"sampleCount"`
+	MemoryTemperatureValue []*float64  `json:"memoryTemperatureValue"`
+	FrequencyValue         []*float64  `json:"frequencyValue"`
+	MemoryFrequencyValue   []*float64  `json:"memoryFrequencyValue"`
+	Date                   []time.Time `json:"date"`
+	GPUValue               []*float64  `json:"gpuValue"`
+	TemperatureValue       []*float64  `json:"temperatureValue"`
+	PowerTotal             []*float64  `json:"powerTotal"`
+	PowerUsed              []*float64  `json:"powerUsed"`
+	PowerPercent           []*float64  `json:"powerPercent"`
+	MemoryTotal            []*float64  `json:"memoryTotal"`
+	MemoryUsed             []*float64  `json:"memoryUsed"`
+	MemoryPercent          []*float64  `json:"memoryPercent"`
+	SpeedValue             []*float64  `json:"speedValue"`
+
+	ProcessCount []*float64     `json:"processCount"`
 	GPUProcesses [][]GPUProcess `json:"gpuProcesses"`
 }
 
@@ -81,3 +125,28 @@ type GPUProcess struct {
 	ProcessName string `json:"processName"`
 	UsedMemory  string `json:"usedMemory"`
 }
+
+type MonitorVLLMSearch struct {
+	AppInstallID uint      `json:"appInstallID" validate:"required"`
+	StartTime    time.Time `json:"startTime" validate:"required"`
+	EndTime      time.Time `json:"endTime" validate:"required"`
+	Aggregation  string    `json:"aggregation" validate:"omitempty,oneof=avg max"`
+}
+
+type MonitorVLLMData struct {
+	SampleCount   int64               `json:"sampleCount"`
+	BucketSeconds int64               `json:"bucketSeconds"`
+	Points        []model.MonitorVLLM `json:"points"`
+}
+
+type MonitorVLLMCurrent struct {
+	AppInstallID uint `json:"appInstallID" validate:"required"`
+}
+
+type MonitorVLLMClean struct {
+	AppInstallID uint `json:"appInstallID" validate:"required"`
+}
+
+type MonitorClean struct {
+	Type string `json:"type" validate:"required,oneof=host gpu"`
+}

agent/app/dto/ssh.go

@@ -25,7 +25,7 @@ type RootCertOperate struct {
 	ID             uint   `json:"id"`
 	Name           string `json:"name"`
 	Mode           string `json:"mode"`
-	EncryptionMode string `json:"encryptionMode" validate:"required,oneof=rsa ed25519 ecdsa dsa"`
+	EncryptionMode string `json:"encryptionMode"`
 	PassPhrase     string `json:"passPhrase"`
 	PublicKey      string `json:"publicKey"`
 	PrivateKey     string `json:"privateKey"`

agent/app/model/monitor.go

@@ -1,5 +1,7 @@
 package model
 
+import "time"
+
 type MonitorBase struct {
 	BaseModel
 	Cpu         float64     `json:"cpu"`
@@ -33,14 +35,78 @@ type MonitorNetwork struct {
 }
 
 type MonitorGPU struct {
+	MemoryUtil         *float64 `json:"memoryUtil"`
+	MemoryActivity     *float64 `json:"memoryActivity"`
+	EncoderUtil        *float64 `json:"encoderUtil"`
+	DecoderUtil        *float64 `json:"decoderUtil"`
+	JPEGUtil           *float64 `json:"jpegUtil"`
+	OFAUtil            *float64 `json:"ofaUtil"`
+	MediaUtil          *float64 `json:"mediaUtil"`
+	ComputeUtil        *float64 `json:"computeUtil"`
+	CopyUtil           *float64 `json:"copyUtil"`
+	HotspotTemperature *float64 `json:"hotspotTemperature"`
+	FanRPM             *float64 `json:"fanRPM"`
+	AICPUUtil          *float64 `json:"aiCPUUtil"`
+	CtrlCPUUtil        *float64 `json:"ctrlCPUUtil"`
+	DDRUsed            *float64 `json:"ddrUsed"`
+	DDRTotal           *float64 `json:"ddrTotal"`
+	HBMUsed            *float64 `json:"hbmUsed"`
+	HBMTotal           *float64 `json:"hbmTotal"`
+	DDRBandwidth       *float64 `json:"ddrBandwidth"`
+	HBMBandwidth       *float64 `json:"hbmBandwidth"`
+	MemoryBandwidth    *float64 `json:"memoryBandwidth"`
+	MediaFrequency     *float64 `json:"mediaFrequency"`
+	HugepagesUsed      *float64 `json:"hugepagesUsed"`
+	HugepagesTotal     *float64 `json:"hugepagesTotal"`
+
+	MemoryTemperature *float64 `json:"memoryTemperature"`
+	DeviceID          string   `json:"deviceID"`
+	DeviceType        string   `json:"deviceType"`
+	ProcessStatus     string   `json:"processStatus"`
+	Frequency         *float64 `json:"frequency"`
+	MemoryFrequency   *float64 `json:"memoryFrequency"`
+	IntervalSeconds   int      `json:"intervalSeconds"`
 	BaseModel
-	ProductName   string  `json:"productName"`
-	GPUUtil       float64 `json:"gpuUtil"`
-	Temperature   float64 `json:"temperature"`
-	PowerDraw     float64 `json:"powerDraw"`
-	MaxPowerLimit float64 `json:"maxPowerLimit"`
-	MemUsed       float64 `json:"memUsed"`
-	MemTotal      float64 `json:"memTotal"`
-	FanSpeed      int     `json:"fanSpeed"`
-	Processes     string  `json:"processes"`
+	ProductName   string   `json:"productName"`
+	GPUUtil       *float64 `json:"gpuUtil"`
+	Temperature   *float64 `json:"temperature"`
+	PowerDraw     *float64 `json:"powerDraw"`
+	MaxPowerLimit *float64 `json:"maxPowerLimit"`
+	MemUsed       *float64 `json:"memUsed"`
+	MemTotal      *float64 `json:"memTotal"`
+	FanSpeed      *float64 `json:"fanSpeed"`
+	Processes     string   `json:"processes"`
+}
+
+type MonitorVLLM struct {
+	ID              uint      `json:"-" gorm:"primarykey;autoIncrement"`
+	CreatedAt       time.Time `json:"createdAt"`
+	AppInstallID    uint      `json:"appInstallID"`
+	Status          string    `json:"status"`
+	RawMetrics      string    `json:"-"`
+	HistogramDeltas string    `json:"-"`
+
+	Running               *float64 `json:"running"`
+	Waiting               *float64 `json:"waiting"`
+	CacheUsage            *float64 `json:"cacheUsage"`
+	PromptThroughput      *float64 `json:"promptThroughput"`
+	GenerationThroughput  *float64 `json:"generationThroughput"`
+	RequestThroughput     *float64 `json:"requestThroughput"`
+	TimeToFirstToken      *float64 `json:"timeToFirstToken"`
+	TimePerOutputToken    *float64 `json:"timePerOutputToken"`
+	RequestLatency        *float64 `json:"requestLatency"`
+	PrefillTime           *float64 `json:"prefillTime"`
+	DecodeTime            *float64 `json:"decodeTime"`
+	TimeToFirstTokenP50   *float64 `json:"timeToFirstTokenP50"`
+	TimeToFirstTokenP90   *float64 `json:"timeToFirstTokenP90"`
+	TimeToFirstTokenP95   *float64 `json:"timeToFirstTokenP95"`
+	TimeToFirstTokenP99   *float64 `json:"timeToFirstTokenP99"`
+	TimePerOutputTokenP50 *float64 `json:"timePerOutputTokenP50"`
+	TimePerOutputTokenP90 *float64 `json:"timePerOutputTokenP90"`
+	TimePerOutputTokenP95 *float64 `json:"timePerOutputTokenP95"`
+	TimePerOutputTokenP99 *float64 `json:"timePerOutputTokenP99"`
+	RequestLatencyP50     *float64 `json:"requestLatencyP50"`
+	RequestLatencyP90     *float64 `json:"requestLatencyP90"`
+	RequestLatencyP95     *float64 `json:"requestLatencyP95"`
+	RequestLatencyP99     *float64 `json:"requestLatencyP99"`
 }

agent/app/repo/monitor.go

@@ -1,6 +1,8 @@
 package repo
 
 import (
+	"fmt"
+	"strings"
 	"time"
 
 	"github.com/1Panel-dev/1Panel/agent/app/model"
@@ -10,11 +12,25 @@ import (
 
 type MonitorRepo struct{}
 
+type GPUHistoryPoint struct {
+	model.MonitorGPU
+	Bucket        int64
+	PowerPercent  *float64
+	MemoryPercent *float64
+	ProcessCount  *float64
+}
+
 type IMonitorRepo interface {
+	CleanHost() error
+	CleanGPU() error
 	GetBase(opts ...DBOption) ([]model.MonitorBase, error)
-	GetGPU(opts ...DBOption) ([]model.MonitorGPU, error)
+	CountGPU(opts ...DBOption) (int64, error)
+	GetGPUHistory(start time.Time, bucketSeconds int64, aggregation string, opts ...DBOption) ([]GPUHistoryPoint, error)
+	GetGPUDevices() ([]model.MonitorGPU, error)
 	GetIO(opts ...DBOption) ([]model.MonitorIO, error)
 	GetNetwork(opts ...DBOption) ([]model.MonitorNetwork, error)
+	GetIONames() ([]string, error)
+	GetNetworkNames() ([]string, error)
 
 	CreateMonitorBase(model model.MonitorBase) error
 	BatchCreateMonitorGPU(list []model.MonitorGPU) error
@@ -25,13 +41,26 @@ type IMonitorRepo interface {
 	DelMonitorIO(timeForDelete time.Time) error
 	DelMonitorNet(timeForDelete time.Time) error
 
-	WithByProductName(name string) DBOption
+	WithByGPUDevice(deviceID, name string, legacy bool) DBOption
 }
 
 func NewIMonitorRepo() IMonitorRepo {
 	return &MonitorRepo{}
 }
 
+func (s *MonitorRepo) CleanHost() error {
+	for _, item := range []interface{}{&model.MonitorBase{}, &model.MonitorIO{}, &model.MonitorNetwork{}} {
+		if err := global.MonitorDB.Where("1 = 1").Delete(item).Error; err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func (s *MonitorRepo) CleanGPU() error {
+	return global.GPUMonitorDB.Where("1 = 1").Delete(&model.MonitorGPU{}).Error
+}
+
 func (u *MonitorRepo) GetBase(opts ...DBOption) ([]model.MonitorBase, error) {
 	var data []model.MonitorBase
 	db := global.MonitorDB
@@ -59,14 +88,16 @@ func (u *MonitorRepo) GetNetwork(opts ...DBOption) ([]model.MonitorNetwork, erro
 	err := db.Find(&data).Error
 	return data, err
 }
-func (u *MonitorRepo) GetGPU(opts ...DBOption) ([]model.MonitorGPU, error) {
-	var data []model.MonitorGPU
-	db := global.GPUMonitorDB
-	for _, opt := range opts {
-		db = opt(db)
-	}
-	err := db.Find(&data).Error
-	return data, err
+func (u *MonitorRepo) GetIONames() ([]string, error) {
+	var names []string
+	err := global.MonitorDB.Model(&model.MonitorIO{}).Distinct().Pluck("name", &names).Error
+	return names, err
+}
+
+func (u *MonitorRepo) GetNetworkNames() ([]string, error) {
+	var names []string
+	err := global.MonitorDB.Model(&model.MonitorNetwork{}).Distinct().Pluck("name", &names).Error
+	return names, err
 }
 
 func (u *MonitorRepo) CreateMonitorBase(model model.MonitorBase) error {
@@ -97,8 +128,67 @@ func (s *MonitorRepo) DelMonitorGPU(timeForDelete time.Time) error {
 	return global.GPUMonitorDB.Where("created_at < ?", timeForDelete).Delete(&model.MonitorGPU{}).Error
 }
 
-func (s *MonitorRepo) WithByProductName(name string) DBOption {
-	return func(g *gorm.DB) *gorm.DB {
-		return g.Where("product_name = ?", name)
+func (u *MonitorRepo) GetGPUDevices() ([]model.MonitorGPU, error) {
+	var data []model.MonitorGPU
+	err := global.GPUMonitorDB.Model(&model.MonitorGPU{}).Select("device_id, product_name, device_type").Group("device_id, product_name, device_type").Order("product_name, device_id").Find(&data).Error
+	return data, err
+}
+
+func (u *MonitorRepo) WithByGPUDevice(deviceID, name string, legacy bool) DBOption {
+	return func(db *gorm.DB) *gorm.DB {
+		if deviceID != "" {
+			return db.Where("device_id = ?", deviceID)
+		}
+		db = db.Where("product_name = ?", name)
+		if legacy {
+			db = db.Where("device_id IS NULL OR device_id = ''")
+		}
+		return db
 	}
 }
+
+func (u *MonitorRepo) CountGPU(opts ...DBOption) (int64, error) {
+	db := global.GPUMonitorDB.Model(&model.MonitorGPU{})
+	for _, opt := range opts {
+		db = opt(db)
+	}
+	var count int64
+	err := db.Count(&count).Error
+	return count, err
+}
+
+func (u *MonitorRepo) GetGPUHistory(start time.Time, bucketSeconds int64, aggregation string, opts ...DBOption) ([]GPUHistoryPoint, error) {
+	db := global.GPUMonitorDB.Model(&model.MonitorGPU{})
+	for _, opt := range opts {
+		db = opt(db)
+	}
+	expressions := []string{
+		"CASE WHEN max_power_limit > 0 THEN 100.0 * power_draw / max_power_limit END",
+		"CASE WHEN mem_total > 0 AND mem_used IS NOT NULL THEN 100.0 * mem_used / mem_total ELSE memory_util END",
+		"CASE WHEN (process_status = 'ok' OR process_status IS NULL OR process_status = '') AND json_valid(processes) THEN CASE WHEN json_type(processes) = 'array' THEN json_array_length(processes) END END",
+	}
+	aliases := []string{"power_percent", "memory_percent", "process_count"}
+	columns := []string{"*"}
+	if bucketSeconds > 0 {
+		operation := "AVG"
+		if aggregation == "max" {
+			operation = "MAX"
+		}
+		columns = []string{fmt.Sprintf("(CAST(strftime('%%s', created_at) AS INTEGER) - %d) / %d AS bucket", start.Unix(), bucketSeconds)}
+		for _, column := range []string{"memory_activity", "encoder_util", "decoder_util", "jpeg_util", "ofa_util", "media_util", "compute_util", "copy_util", "hotspot_temperature", "fan_rpm", "ai_cpu_util", "ctrl_cpu_util", "ddr_used", "ddr_total", "hbm_used", "hbm_total", "ddr_bandwidth", "hbm_bandwidth", "memory_bandwidth", "media_frequency", "hugepages_used", "hugepages_total", "gpu_util", "temperature", "memory_temperature", "power_draw", "max_power_limit", "mem_used", "mem_total", "frequency", "memory_frequency", "fan_speed"} {
+			columns = append(columns, operation+"("+column+") AS "+column)
+		}
+		for i := range expressions {
+			expressions[i] = operation + "(" + expressions[i] + ")"
+		}
+		db = db.Group("bucket").Order("bucket ASC")
+	} else {
+		db = db.Order("created_at ASC, id ASC")
+	}
+	for i, expression := range expressions {
+		columns = append(columns, expression+" AS "+aliases[i])
+	}
+	var data []GPUHistoryPoint
+	err := db.Select(strings.Join(columns, ", ")).Scan(&data).Error
+	return data, err
+}

agent/app/repo/monitor_vllm.go

@@ -0,0 +1,78 @@
+package repo
+
+import (
+	"errors"
+	"fmt"
+	"strings"
+	"time"
+
+	"github.com/1Panel-dev/1Panel/agent/app/model"
+	"github.com/1Panel-dev/1Panel/agent/global"
+	"gorm.io/gorm"
+)
+
+type VLLMMonitorRepo struct{}
+
+type VLLMHistoryPoint struct {
+	model.MonitorVLLM
+	Bucket           int64
+	HistogramSamples string
+}
+
+func (r *VLLMMonitorRepo) Create(point *model.MonitorVLLM) error {
+	return global.VLLMMonitorDB.Create(point).Error
+}
+
+func (r *VLLMMonitorRepo) Latest(id uint) (model.MonitorVLLM, error) {
+	var point model.MonitorVLLM
+	db := global.VLLMMonitorDB.Where("app_install_id = ?", id)
+	err := db.Order("created_at DESC, id DESC").First(&point).Error
+	if errors.Is(err, gorm.ErrRecordNotFound) {
+		return point, nil
+	}
+	return point, err
+}
+
+func (r *VLLMMonitorRepo) CleanTarget(id uint) error {
+	return global.VLLMMonitorDB.Where("app_install_id = ?", id).Delete(&model.MonitorVLLM{}).Error
+}
+
+func (r *VLLMMonitorRepo) DeleteBefore(before time.Time) error {
+	return global.VLLMMonitorDB.Where("created_at < ?", before).Delete(&model.MonitorVLLM{}).Error
+}
+
+func (r *VLLMMonitorRepo) Count(id uint, start, end time.Time) (int64, error) {
+	var count int64
+	db := global.VLLMMonitorDB.Model(&model.MonitorVLLM{}).Where("app_install_id = ? AND created_at >= ? AND created_at <= ?", id, start, end)
+	err := db.Count(&count).Error
+	return count, err
+}
+
+func (r *VLLMMonitorRepo) History(id uint, start, end time.Time, seconds int64, aggregation string) ([]VLLMHistoryPoint, error) {
+	db := global.VLLMMonitorDB.Model(&model.MonitorVLLM{}).Where("app_install_id = ? AND created_at >= ? AND created_at <= ?", id, start, end)
+	metrics := []string{"running", "waiting", "cache_usage", "prompt_throughput", "generation_throughput", "request_throughput", "time_to_first_token", "time_per_output_token", "request_latency", "prefill_time", "decode_time", "time_to_first_token_p50", "time_to_first_token_p90", "time_to_first_token_p95", "time_to_first_token_p99", "time_per_output_token_p50", "time_per_output_token_p90", "time_per_output_token_p95", "time_per_output_token_p99", "request_latency_p50", "request_latency_p90", "request_latency_p95", "request_latency_p99"}
+	var columns []string
+	if seconds > 0 {
+		operation := "AVG"
+		if aggregation == "max" {
+			operation = "MAX"
+		}
+		columns = []string{fmt.Sprintf("(CAST(strftime('%%s', created_at) AS INTEGER) - %d) / %d AS bucket", start.Unix(), seconds)}
+		for _, column := range metrics {
+			if aggregation != "max" && (strings.HasPrefix(column, "time_to_first_token_p") || strings.HasPrefix(column, "time_per_output_token_p") || strings.HasPrefix(column, "request_latency_p")) {
+				continue
+			}
+			columns = append(columns, operation+"("+column+") AS "+column)
+		}
+		if aggregation != "max" {
+			columns = append(columns, "json_group_array(json(NULLIF(histogram_deltas, ''))) AS histogram_samples")
+		}
+		db = db.Group("bucket").Order("bucket ASC")
+	} else {
+		columns = append([]string{"id", "created_at", "app_install_id", "status"}, metrics...)
+		db = db.Order("created_at ASC, id ASC")
+	}
+	var points []VLLMHistoryPoint
+	err := db.Select(strings.Join(columns, ", ")).Scan(&points).Error
+	return points, err
+}

agent/app/service/alert.go

@@ -17,6 +17,7 @@ import (
 	"github.com/1Panel-dev/1Panel/agent/constant"
 	"github.com/1Panel-dev/1Panel/agent/global"
 	"github.com/1Panel-dev/1Panel/agent/i18n"
+	alertUtil "github.com/1Panel-dev/1Panel/agent/utils/alert"
 	alertconfig "github.com/1Panel-dev/1Panel/agent/utils/alert_config"
 	alertwebhook "github.com/1Panel-dev/1Panel/agent/utils/alert_webhook"
 	"github.com/1Panel-dev/1Panel/agent/utils/cmd"
@@ -109,7 +110,38 @@ func (a AlertService) PageAlert(search dto.AlertSearch) (int64, []dto.AlertDTO,
 		return 0, nil, err
 	}
 
+	cronjobProjects := make(map[string]uint)
+	var cronjobIDs []uint
 	for _, item := range alerts {
+		if alertUtil.GetCronJobType(item.Type) != "cronJob" {
+			continue
+		}
+		if _, exists := cronjobProjects[item.Project]; exists {
+			continue
+		}
+		id, parseErr := strconv.ParseUint(item.Project, 10, strconv.IntSize)
+		if parseErr != nil || id == 0 {
+			continue
+		}
+		cronjobProjects[item.Project] = uint(id)
+		cronjobIDs = append(cronjobIDs, uint(id))
+	}
+	cronjobsByID := make(map[uint]model.Cronjob)
+	if len(cronjobIDs) > 0 {
+		cronjobs, err := cronjobRepo.List(repo.WithByIDs(cronjobIDs))
+		if err != nil {
+			return 0, nil, err
+		}
+		for _, cronjob := range cronjobs {
+			cronjobsByID[cronjob.ID] = cronjob
+		}
+	}
+
+	for _, item := range alerts {
+		var taskName string
+		if cronjob, exists := cronjobsByID[cronjobProjects[item.Project]]; exists && cronjob.Type == item.Type {
+			taskName = cronjob.Name
+		}
 
 		result = append(result, dto.AlertDTO{
 			ID:             item.ID,
@@ -119,6 +151,7 @@ func (a AlertService) PageAlert(search dto.AlertSearch) (int64, []dto.AlertDTO,
 			Method:         item.Method,
 			Title:          item.Title,
 			Project:        item.Project,
+			TaskName:       taskName,
 			Status:         item.Status,
 			SendCount:      item.SendCount,
 			AdvancedParams: item.AdvancedParams,
@@ -190,6 +223,16 @@ func (a AlertService) CreateAlert(create dto.AlertCreate, operator string) error
 			return err
 		}
 	} else {
+		advanced, err := prepareCronJobAlertParams(create.Type, "", create.AdvancedParams)
+		if err != nil {
+			return err
+		}
+		create.AdvancedParams = advanced
+		if create.Status != constant.AlertDisable {
+			if err := a.validateCronJobAlertChannels(create.Type, advanced, create.Method); err != nil {
+				return err
+			}
+		}
 		alertInfo.Status = constant.AlertEnable
 		if err := copier.Copy(&alertInfo, &create); err != nil {
 			return buserr.WithErr("ErrStructTransform", err)
@@ -207,11 +250,24 @@ func (a AlertService) CreateAlert(create dto.AlertCreate, operator string) error
 }
 
 func (a AlertService) UpdateAlert(req dto.AlertUpdate, operator string) error {
+	if alertUtil.GetCronJobType(req.Type) == "cronJob" {
+		previous, err := alertRepo.Get(repo.WithByID(req.ID))
+		if err != nil {
+			return err
+		}
+		req.AdvancedParams, err = prepareCronJobAlertParams(req.Type, previous.AdvancedParams, req.AdvancedParams)
+		if err != nil {
+			return err
+		}
+	}
 	methodTypes, err := a.validateAlertMethodReferences(req.Method)
 	if err != nil {
 		return err
 	}
 	if req.Status != constant.AlertDisable {
+		if err := a.validateCronJobAlertChannels(req.Type, req.AdvancedParams, req.Method); err != nil {
+			return err
+		}
 		if err := a.validateAlertMethodEntitlement(methodTypes); err != nil {
 			return err
 		}
@@ -278,6 +334,9 @@ func (a AlertService) UpdateStatus(id uint, status string) error {
 		return err
 	}
 	if status == constant.AlertEnable {
+		if err := a.validateCronJobAlertChannels(alertInfo.Type, alertInfo.AdvancedParams, alertInfo.Method); err != nil {
+			return err
+		}
 		if err := a.validateAlertMethodEntitlement(methodTypes); err != nil {
 			return err
 		}
@@ -1021,6 +1080,23 @@ func (a AlertService) ExternalUpdateAlert(updateAlert dto.AlertCreate, operator
 		alertRepo.WithByType(updateAlert.Type),
 		alertRepo.WithByProject(updateAlert.Project),
 	)
+	advanced, err := prepareCronJobAlertParams(updateAlert.Type, alertInfo.AdvancedParams, updateAlert.AdvancedParams)
+	if err != nil {
+		return err
+	}
+	updateAlert.AdvancedParams = advanced
+	if alertUtil.GetCronJobType(updateAlert.Type) == "cronJob" {
+		upMap["advanced_params"] = advanced
+	}
+	if newStatus == constant.AlertEnable {
+		method := updateAlert.Method
+		if method == "" {
+			method = alertInfo.Method
+		}
+		if err := a.validateCronJobAlertChannels(updateAlert.Type, advanced, method); err != nil {
+			return err
+		}
+	}
 
 	if alertInfo.ID > 0 {
 		shouldUpdate := false
@@ -1034,6 +1110,9 @@ func (a AlertService) ExternalUpdateAlert(updateAlert dto.AlertCreate, operator
 		if val, ok := upMap["method"]; ok && val != "" && val != alertInfo.Method {
 			shouldUpdate = true
 		}
+		if val, ok := upMap["advanced_params"]; ok && val != alertInfo.AdvancedParams {
+			shouldUpdate = true
+		}
 
 		if shouldUpdate {
 			if err := alertRepo.Update(
@@ -1055,3 +1134,22 @@ func (a AlertService) ExternalUpdateAlert(updateAlert dto.AlertCreate, operator
 
 	return nil
 }
+
+func prepareCronJobAlertParams(alertType, previous, incoming string) (string, error) {
+	if alertUtil.GetCronJobType(alertType) != "cronJob" {
+		return incoming, nil
+	}
+	return alertUtil.MergeCronJobAlertParams(previous, incoming)
+}
+
+func (a AlertService) validateCronJobAlertChannels(alertType, advanced, method string) error {
+	if alertUtil.GetCronJobType(alertType) != "cronJob" {
+		return nil
+	}
+	mode, err := alertUtil.CronJobAlertTriggerMode(advanced)
+	if err != nil || mode != alertUtil.CronJobAlertSuccess {
+		return err
+	}
+	_, err = a.validateAlertMethodReferences(method)
+	return err
+}

agent/app/service/alert_helper.go

@@ -530,11 +530,15 @@ func loadSSHLogin(alert dto.AlertDTO) {
 	if err != nil {
 		global.LOG.Errorf("Failed to load ssh login logs: %v", err)
 	}
+	interfaceAddrs, err := net.InterfaceAddrs()
+	if err != nil {
+		global.LOG.Warnf("Failed to load local IP addresses for ssh login alert: %v", err)
+	}
 	count, records := summarizeSSHLoginHistories(
 		histories,
 		now,
 		failedWindow,
-		strings.Split(strings.TrimSpace(alert.AdvancedParams), "\n"),
+		sshSuccessLoginWhitelist(alert.AdvancedParams, interfaceAddrs),
 	)
 	isAlert := count >= int(alert.Count)
 	if isAlert {
@@ -570,6 +574,19 @@ func loadSSHLogin(alert dto.AlertDTO) {
 	}
 }
 
+func sshSuccessLoginWhitelist(configured string, interfaceAddrs []net.Addr) []string {
+	whitelist := strings.Split(strings.TrimSpace(configured), "\n")
+	whitelist = append(whitelist, "127.0.0.0/8", "::1")
+	for _, addr := range interfaceAddrs {
+		ipNet, ok := addr.(*net.IPNet)
+		if !ok || ipNet.IP == nil || ipNet.IP.IsUnspecified() {
+			continue
+		}
+		whitelist = append(whitelist, ipNet.IP.String())
+	}
+	return whitelist
+}
+
 func filterLoginLogsNotInWhitelist(records []model.LoginLog, whitelist []string) []model.LoginLog {
 	filtered := make([]model.LoginLog, 0, len(records))
 	for _, record := range records {

agent/app/service/container.go

@@ -58,10 +58,10 @@ type ContainerService struct{}
 var containerLogAnsiRegex = regexp.MustCompile("\x1b\\[[0-9;?]*[A-Za-z]|\x1b=|\x1b>")
 
 type IContainerService interface {
-	Page(req dto.PageContainer) (int64, interface{}, error)
+	Page(ctx context.Context, req dto.PageContainer) (int64, interface{}, error)
 	List() []dto.ContainerOptions
 	ListByImage(imageName string) []dto.ContainerOptions
-	LoadStatus() (dto.ContainerStatus, error)
+	LoadStatus(ctx context.Context, containersOnly bool) (dto.ContainerStatus, error)
 	PageNetwork(req dto.SearchWithPage) (int64, interface{}, error)
 	ListNetwork() ([]dto.Options, error)
 	PageVolume(req dto.SearchWithPage) (int64, interface{}, error)
@@ -80,7 +80,7 @@ type IContainerService interface {
 	ContainerUpdate(req dto.ContainerOperate) error
 	ContainerUpgrade(req dto.ContainerUpgrade) error
 	ContainerInfo(req dto.OperationWithName) (*dto.ContainerOperate, error)
-	ContainerListStats() ([]dto.ContainerListStats, error)
+	ContainerListStats(ctx context.Context, ids []string) ([]dto.ContainerListStats, error)
 	ContainerItemStats(ctx context.Context, req dto.OperationWithName) (dto.ContainerItemStats, error)
 	LoadResourceLimit() (*dto.ResourceLimit, error)
 	ContainerRename(req dto.ContainerRename) error
@@ -92,6 +92,7 @@ type IContainerService interface {
 
 	Inspect(req dto.InspectReq) (string, error)
 	DeleteNetwork(req dto.BatchDelete) error
+	CleanNetworks() (*dto.NetworkCleanupTask, error)
 	CreateNetwork(req dto.NetworkCreate) error
 	DeleteVolume(req dto.BatchDelete) error
 	CreateVolume(req dto.VolumeCreate) error
@@ -112,7 +113,7 @@ func NewIContainerService() IContainerService {
 	return &ContainerService{}
 }
 
-func (u *ContainerService) Page(req dto.PageContainer) (int64, interface{}, error) {
+func (u *ContainerService) Page(ctx context.Context, req dto.PageContainer) (int64, interface{}, error) {
 	client, err := docker.NewDockerClient()
 	if err != nil {
 		return 0, nil, err
@@ -123,7 +124,7 @@ func (u *ContainerService) Page(req dto.PageContainer) (int64, interface{}, erro
 		options.Filters = filters.NewArgs()
 		options.Filters.Add("label", req.Filters)
 	}
-	containers, err := client.ContainerList(context.Background(), options)
+	containers, err := client.ContainerList(ctx, options)
 	if err != nil {
 		return 0, nil, err
 	}
@@ -206,27 +207,30 @@ func (u *ContainerService) ListByImage(imageName string) []dto.ContainerOptions
 	return options
 }
 
-func (u *ContainerService) LoadStatus() (dto.ContainerStatus, error) {
+func (u *ContainerService) LoadStatus(ctx context.Context, containersOnly bool) (dto.ContainerStatus, error) {
 	var data dto.ContainerStatus
 	client, err := docker.NewDockerClient()
 	if err != nil {
 		return data, err
 	}
 	defer client.Close()
-	c := context.Background()
-
-	images, _ := client.ImageList(c, image.ListOptions{All: true})
-	data.ImageCount = len(images)
-	repo, _ := imageRepoRepo.List()
-	data.RepoCount = len(repo)
-	templates, _ := composeRepo.List()
-	data.ComposeTemplateCount = len(templates)
-	networks, _ := client.NetworkList(c, network.ListOptions{})
-	data.NetworkCount = len(networks)
-	volumes, _ := client.VolumeList(c, volume.ListOptions{})
-	data.VolumeCount = len(volumes.Volumes)
-	data.ComposeCount = loadComposeCount(client)
-	containers, _ := client.ContainerList(c, container.ListOptions{All: true})
+	if !containersOnly {
+		images, _ := client.ImageList(ctx, image.ListOptions{All: true})
+		data.ImageCount = len(images)
+		repo, _ := imageRepoRepo.List()
+		data.RepoCount = len(repo)
+		templates, _ := composeRepo.List()
+		data.ComposeTemplateCount = len(templates)
+		networks, _ := client.NetworkList(ctx, network.ListOptions{})
+		data.NetworkCount = len(networks)
+		volumes, _ := client.VolumeList(ctx, volume.ListOptions{})
+		data.VolumeCount = len(volumes.Volumes)
+		data.ComposeCount = loadComposeCount(ctx, client)
+	}
+	containers, err := client.ContainerList(ctx, container.ListOptions{All: true})
+	if err != nil {
+		return data, err
+	}
 	data.ContainerCount = len(containers)
 	for _, item := range containers {
 		switch item.State {
@@ -292,27 +296,65 @@ func (u *ContainerService) ContainerItemStats(ctx context.Context, req dto.Opera
 	}
 	return data, nil
 }
-func (u *ContainerService) ContainerListStats() ([]dto.ContainerListStats, error) {
+func (u *ContainerService) ContainerListStats(ctx context.Context, ids []string) ([]dto.ContainerListStats, error) {
 	client, err := docker.NewDockerClient()
 	if err != nil {
 		return nil, err
 	}
 	defer client.Close()
-	list, err := client.ContainerList(context.Background(), container.ListOptions{All: true})
+	options := container.ListOptions{All: true}
+	if ids != nil {
+		if len(ids) == 0 {
+			return []dto.ContainerListStats{}, nil
+		}
+		options.Filters = filters.NewArgs()
+		for _, id := range ids {
+			options.Filters.Add("id", id)
+		}
+	}
+	list, err := client.ContainerList(ctx, options)
 	if err != nil {
 		return nil, err
 	}
+	return collectContainerStats(ctx, list, func(ctx context.Context, id string) dto.ContainerListStats {
+		return loadCpuAndMem(ctx, client, id)
+	}), nil
+}
+
+// A fixed worker pool bounds Docker stats requests, including for legacy callers
+// that request all containers. Stopped containers do not need a stats sample.
+func collectContainerStats(ctx context.Context, list []container.Summary, load func(context.Context, string) dto.ContainerListStats) []dto.ContainerListStats {
 	datas := make([]dto.ContainerListStats, len(list))
-	var wg sync.WaitGroup
-	wg.Add(len(list))
-	for i := 0; i < len(list); i++ {
-		go func(index int, item container.Summary) {
-			datas[index] = loadCpuAndMem(client, item.ID)
-			wg.Done()
-		}(i, list[i])
+	for i, item := range list {
+		datas[i].ContainerID = item.ID
 	}
+	jobs := make(chan int)
+	var wg sync.WaitGroup
+	for worker := 0; worker < min(8, len(list)); worker++ {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			for index := range jobs {
+				if ctx.Err() != nil || list[index].State != "running" {
+					continue
+				}
+				sampleCtx, cancel := context.WithTimeout(ctx, 3*time.Second)
+				datas[index] = load(sampleCtx, list[index].ID)
+				cancel()
+			}
+		}()
+	}
+dispatch:
+	for index := range list {
+		select {
+		case <-ctx.Done():
+			break dispatch
+		case jobs <- index:
+		}
+	}
+	close(jobs)
 	wg.Wait()
-	return datas, nil
+	return datas
 }
 
 func (u *ContainerService) Inspect(req dto.InspectReq) (string, error) {
@@ -1715,11 +1757,11 @@ func selectImageRepo(imageName string, repos []model.ImageRepo) *model.ImageRepo
 	return selected
 }
 
-func loadCpuAndMem(client *client.Client, containerItem string) dto.ContainerListStats {
+func loadCpuAndMem(ctx context.Context, client *client.Client, containerItem string) dto.ContainerListStats {
 	data := dto.ContainerListStats{
 		ContainerID: containerItem,
 	}
-	res, err := client.ContainerStats(context.Background(), containerItem, false)
+	res, err := client.ContainerStats(ctx, containerItem, false)
 	if err != nil {
 		return data
 	}
@@ -1942,11 +1984,11 @@ func transPortToStr(ports []container.Port) []string {
 	return docker.SimplifyPorts(ports)
 }
 
-func loadComposeCount(client *client.Client) int {
+func loadComposeCount(ctx context.Context, client *client.Client) int {
 	options := container.ListOptions{All: true}
 	options.Filters = filters.NewArgs()
 	options.Filters.Add("label", composeProjectLabel)
-	list, err := client.ContainerList(context.Background(), options)
+	list, err := client.ContainerList(ctx, options)
 	if err != nil {
 		return 0
 	}

agent/app/service/container_network_cleanup.go

@@ -0,0 +1,71 @@
+package service
+
+import (
+	"fmt"
+	"sync"
+
+	"github.com/1Panel-dev/1Panel/agent/app/dto"
+	"github.com/1Panel-dev/1Panel/agent/app/task"
+	"github.com/1Panel-dev/1Panel/agent/global"
+	"github.com/1Panel-dev/1Panel/agent/i18n"
+	"github.com/1Panel-dev/1Panel/agent/utils/docker"
+)
+
+var networkCleanupMu sync.Mutex
+
+func (u *ContainerService) CleanNetworks() (*dto.NetworkCleanupTask, error) {
+	taskItem, err := task.NewTaskWithOps(i18n.GetMsgByKey("Network"), task.TaskClean, task.TaskScopeContainer, "", 0)
+	if err != nil {
+		return nil, err
+	}
+	taskItem.AddSubTask(i18n.GetMsgByKey("TaskClean"), func(t *task.Task) error {
+		networkCleanupMu.Lock()
+		defer networkCleanupMu.Unlock()
+		if err := t.TaskCtx.Err(); err != nil {
+			return err
+		}
+		cli, err := docker.NewDockerClient()
+		if err != nil {
+			return err
+		}
+		defer cli.Close()
+		return executeNetworkCleanup(t, cli)
+	}, nil)
+	go func() {
+		if err := taskItem.Execute(); err != nil {
+			global.LOG.Errorf("network cleanup task %s failed: %v", taskItem.TaskID, err)
+		}
+	}()
+	return &dto.NetworkCleanupTask{TaskID: taskItem.TaskID}, nil
+}
+
+func executeNetworkCleanup(t *task.Task, cli docker.NetworkCleanupClient) error {
+	t.Log(i18n.GetMsgByKey("PruneStart"))
+	report, err := docker.CleanUnusedNetworks(t.TaskCtx, cli, func(status string, item dto.NetworkCleanupItem) {
+		key := "NetworkCleanupDeleted"
+		if status == "skipped" || status == "failed" {
+			key = map[string]string{
+				"protected":           "NetworkCleanupProtected",
+				"container_connected": "NetworkCleanupConnected",
+				"network_in_use":      "NetworkCleanupConnected",
+				"unsupported_network": "NetworkCleanupUnsupported",
+				"already_removed":     "NetworkCleanupGone",
+				"inspect_failed":      "NetworkCleanupInspectFailed",
+				"remove_failed":       "NetworkCleanupRemoveFailed",
+			}[item.Reason]
+		}
+		t.Log(i18n.GetMsgWithMap(key, map[string]interface{}{"name": item.Name, "id": item.ID}))
+	})
+	if report != nil {
+		t.Log(i18n.GetMsgWithMap("NetworkCleanupSummary", map[string]interface{}{
+			"deleted": len(report.Deleted), "skipped": len(report.Skipped), "failed": len(report.Failed),
+		}))
+	}
+	if err != nil {
+		return err
+	}
+	if len(report.Failed) > 0 {
+		return fmt.Errorf("%s", i18n.GetMsgByKey("NetworkCleanupPartialFailure"))
+	}
+	return nil
+}

agent/app/service/container_network_cleanup_test.go

@@ -0,0 +1,113 @@
+package service
+
+import (
+	"context"
+	"errors"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"github.com/1Panel-dev/1Panel/agent/app/model"
+	"github.com/1Panel-dev/1Panel/agent/app/task"
+	"github.com/1Panel-dev/1Panel/agent/constant"
+	"github.com/1Panel-dev/1Panel/agent/global"
+	"github.com/1Panel-dev/1Panel/agent/i18n"
+	"github.com/docker/docker/api/types/container"
+	"github.com/docker/docker/api/types/network"
+	"github.com/docker/docker/errdefs"
+	"github.com/glebarez/sqlite"
+	"gorm.io/gorm"
+)
+
+type networkTaskClient struct{ fail bool }
+
+func (f networkTaskClient) NetworkList(context.Context, network.ListOptions) ([]network.Inspect, error) {
+	return []network.Inspect{{ID: "reserved", Name: "1panel-network", Scope: "local"}, {ID: "connected", Name: "busy", Scope: "local"}, {ID: "unused", Name: "free", Scope: "local"}, {ID: "race", Name: "race", Scope: "local"}}, nil
+}
+func (f networkTaskClient) ContainerList(context.Context, container.ListOptions) ([]container.Summary, error) {
+	return nil, nil
+}
+func (f networkTaskClient) NetworkInspect(_ context.Context, id string, _ network.InspectOptions) (network.Inspect, error) {
+	n := network.Inspect{}
+	if id == "connected" {
+		n.Containers = map[string]network.EndpointResource{"container-id": {}}
+	}
+	return n, nil
+}
+func (f networkTaskClient) NetworkRemove(_ context.Context, id string) error {
+	if id == "race" {
+		return errdefs.Conflict(errors.New("has active endpoints"))
+	}
+	if id != "unused" {
+		return errors.New("unexpected removal")
+	}
+	if f.fail {
+		return errors.New("remove failed")
+	}
+	return nil
+}
+
+func TestNetworkCleanupTaskPersistsLogsAndStatus(t *testing.T) {
+	oldDB, oldTaskDB, oldDir, oldI18n := global.DB, global.TaskDB, global.Dir, global.I18n
+	t.Cleanup(func() { global.DB = oldDB; global.TaskDB = oldTaskDB; global.Dir = oldDir; global.I18n = oldI18n })
+	dir := t.TempDir()
+	db, err := gorm.Open(sqlite.Open(filepath.Join(dir, "tasks.db")), &gorm.Config{})
+	if err != nil {
+		t.Fatal(err)
+	}
+	sqlDB, err := db.DB()
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer sqlDB.Close()
+	if err := db.AutoMigrate(&model.Task{}); err != nil {
+		t.Fatal(err)
+	}
+	global.DB = nil
+	global.TaskDB = db
+	global.Dir.TaskDir = dir
+	i18n.Init()
+	for _, fail := range []bool{false, true} {
+		name := "success"
+		wantStatus := constant.StatusSuccess
+		if fail {
+			name = "partial failure"
+			wantStatus = constant.StatusFailed
+		}
+		t.Run(name, func(t *testing.T) {
+			item, err := task.NewTaskWithOps("Network", task.TaskClean, task.TaskScopeContainer, "", 0)
+			if err != nil {
+				t.Fatal(err)
+			}
+			item.AddSubTask("Clean", func(t *task.Task) error { return executeNetworkCleanup(t, networkTaskClient{fail: fail}) }, nil)
+			err = item.Execute()
+			if (err != nil) != fail {
+				t.Fatalf("unexpected execution error: %v", err)
+			}
+			var saved model.Task
+			if err := db.First(&saved, "id = ?", item.TaskID).Error; err != nil {
+				t.Fatal(err)
+			}
+			if saved.Status != wantStatus {
+				t.Fatalf("status %s, want %s", saved.Status, wantStatus)
+			}
+			content, err := os.ReadFile(saved.LogFile)
+			if err != nil {
+				t.Fatal(err)
+			}
+			for _, want := range []string{"[busy] (connected): containers connected", "[race] (race): containers connected", "[1panel-network] (reserved): reserved network", "Network cleanup finished:", "[TASK-END]"} {
+				if !strings.Contains(string(content), want) {
+					t.Fatalf("missing %q in log: %s", want, content)
+				}
+			}
+			want := "Deleted network [free]"
+			if fail {
+				want = "Failed to remove network [free]"
+			}
+			if !strings.Contains(string(content), want) {
+				t.Fatalf("missing %q", want)
+			}
+		})
+	}
+}

agent/app/service/cronjob.go

@@ -16,6 +16,7 @@ import (
 	"github.com/1Panel-dev/1Panel/agent/buserr"
 	"github.com/1Panel-dev/1Panel/agent/constant"
 	"github.com/1Panel-dev/1Panel/agent/global"
+	alertUtil "github.com/1Panel-dev/1Panel/agent/utils/alert"
 	"github.com/1Panel-dev/1Panel/agent/utils/docker"
 	"github.com/jinzhu/copier"
 	"github.com/pkg/errors"
@@ -75,6 +76,7 @@ func (u *CronjobService) SearchWithPage(search dto.PageCronjob) (int64, interfac
 			EntryID:   cronjob.ID,
 		}
 		alertInfo, _ := alertRepo.Get(alertRepo.WithByType(alertBase.AlertType), alertRepo.WithByProject(strconv.Itoa(int(alertBase.EntryID))), repo.WithByStatus(constant.AlertEnable))
+		item.AlertTriggerMode, _ = alertUtil.CronJobAlertTriggerMode(alertInfo.AdvancedParams)
 		if alertInfo.SendCount != 0 {
 			item.AlertCount = alertInfo.SendCount
 		} else {
@@ -98,9 +100,11 @@ func (u *CronjobService) LoadInfo(req dto.OperateByID) (*dto.CronjobOperate, err
 		AlertType: cronjob.Type,
 		EntryID:   cronjob.ID,
 	}
-	alertInfo, _ := alertRepo.Get(alertRepo.WithByType(alertBase.AlertType), alertRepo.WithByProject(strconv.Itoa(int(alertBase.EntryID))), repo.WithByStatus(constant.AlertEnable))
+	alertInfo, _ := alertRepo.Get(alertRepo.WithByType(alertBase.AlertType), alertRepo.WithByProject(strconv.Itoa(int(alertBase.EntryID))))
 	item.AlertMethod = alertInfo.Method
-	if alertInfo.SendCount != 0 {
+	item.AlertTitle = alertInfo.Title
+	item.AlertTriggerMode, _ = alertUtil.CronJobAlertTriggerMode(alertInfo.AdvancedParams)
+	if alertInfo.Status == constant.AlertEnable {
 		item.AlertCount = alertInfo.SendCount
 	} else {
 		item.AlertCount = 0
@@ -195,11 +199,12 @@ func (u *CronjobService) Export(req dto.OperateByIDs) (string, error) {
 			}
 		}
 		item.SourceAccounts, item.DownloadAccount, _ = loadBackupNamesByID(cronjob.SourceAccountIDs, cronjob.DownloadAccountID)
-		alertInfo, _ := alertRepo.Get(alertRepo.WithByType(cronjob.Type), alertRepo.WithByProject(strconv.Itoa(int(cronjob.ID))), repo.WithByStatus(constant.AlertEnable))
-		if alertInfo.SendCount != 0 {
+		alertInfo, _ := alertRepo.Get(alertRepo.WithByType(cronjob.Type), alertRepo.WithByProject(strconv.Itoa(int(cronjob.ID))))
+		item.AlertTitle = alertInfo.Title
+		item.AlertMethod = alertInfo.Method
+		item.AlertTriggerMode, _ = alertUtil.CronJobAlertTriggerMode(alertInfo.AdvancedParams)
+		if alertInfo.Status == constant.AlertEnable {
 			item.AlertCount = alertInfo.SendCount
-			item.AlertTitle = alertInfo.Title
-			item.AlertMethod = alertInfo.Method
 		} else {
 			item.AlertCount = 0
 		}
@@ -213,6 +218,17 @@ func (u *CronjobService) Export(req dto.OperateByIDs) (string, error) {
 }
 
 func (u *CronjobService) Import(req []dto.CronjobTrans, operator string) error {
+	for _, item := range req {
+		advanced, err := cronJobAlertAdvancedParams(item.AlertTriggerMode)
+		if err != nil {
+			return err
+		}
+		if item.AlertCount != 0 {
+			if err := (AlertService{}).validateCronJobAlertChannels(item.Type, advanced, item.AlertMethod); err != nil {
+				return err
+			}
+		}
+	}
 	for _, item := range req {
 		cronjobItem, _ := cronjobRepo.Get(repo.WithByName(item.Name))
 		if cronjobItem.ID != 0 {
@@ -395,17 +411,27 @@ func (u *CronjobService) Import(req []dto.CronjobTrans, operator string) error {
 		} else {
 			cronjob.Status = constant.StatusDisable
 		}
-		_ = cronjobRepo.Create(&cronjob)
-		if item.AlertCount != 0 && item.AlertTitle != "" && item.AlertMethod != "" {
+		if err := cronjobRepo.Create(&cronjob); err != nil {
+			return err
+		}
+		if item.AlertTitle != "" && item.AlertMethod != "" {
+			advanced, _ := cronJobAlertAdvancedParams(item.AlertTriggerMode)
+			status := constant.AlertEnable
+			if item.AlertCount == 0 {
+				status = constant.AlertDisable
+			}
 			createAlert := dto.AlertCreate{
-				Title:     item.AlertTitle,
-				SendCount: item.AlertCount,
-				Method:    item.AlertMethod,
-				Type:      cronjob.Type,
-				Project:   strconv.Itoa(int(cronjob.ID)),
-				Status:    constant.AlertEnable,
+				Title:          item.AlertTitle,
+				SendCount:      item.AlertCount,
+				Method:         item.AlertMethod,
+				Type:           cronjob.Type,
+				Project:        strconv.Itoa(int(cronjob.ID)),
+				Status:         status,
+				AdvancedParams: advanced,
+			}
+			if err := NewIAlertService().CreateAlert(createAlert, operator); err != nil {
+				return err
 			}
-			_ = NewIAlertService().CreateAlert(createAlert, operator)
 		}
 	}
 	return nil
@@ -562,6 +588,15 @@ func (u *CronjobService) HandleOnce(id uint) error {
 }
 
 func (u *CronjobService) Create(req dto.CronjobOperate, operator string) error {
+	advanced, err := cronJobAlertAdvancedParams(req.AlertTriggerMode)
+	if err != nil {
+		return err
+	}
+	if req.AlertCount != 0 {
+		if err := (AlertService{}).validateCronJobAlertChannels(req.Type, advanced, req.AlertMethod); err != nil {
+			return err
+		}
+	}
 	cronjob, _ := cronjobRepo.Get(repo.WithByName(req.Name))
 	if cronjob.ID != 0 {
 		return buserr.New("ErrRecordExist")
@@ -603,12 +638,13 @@ func (u *CronjobService) Create(req dto.CronjobOperate, operator string) error {
 	}
 	if req.AlertCount != 0 && req.AlertTitle != "" && req.AlertMethod != "" {
 		createAlert := dto.AlertCreate{
-			Title:     req.AlertTitle,
-			SendCount: req.AlertCount,
-			Method:    req.AlertMethod,
-			Type:      cronjob.Type,
-			Project:   strconv.Itoa(int(cronjob.ID)),
-			Status:    constant.AlertEnable,
+			Title:          req.AlertTitle,
+			SendCount:      req.AlertCount,
+			Method:         req.AlertMethod,
+			Type:           cronjob.Type,
+			Project:        strconv.Itoa(int(cronjob.ID)),
+			Status:         constant.AlertEnable,
+			AdvancedParams: advanced,
 		}
 		err := NewIAlertService().CreateAlert(createAlert, operator)
 		if err != nil {
@@ -682,6 +718,10 @@ func (u *CronjobService) Delete(req dto.CronjobBatchDelete) error {
 }
 
 func (u *CronjobService) Update(id uint, req dto.CronjobOperate, operator string) error {
+	advanced, err := cronJobAlertAdvancedParams(req.AlertTriggerMode)
+	if err != nil {
+		return err
+	}
 	var cronjob model.Cronjob
 	if err := copier.Copy(&cronjob, &req); err != nil {
 		return buserr.WithDetail("ErrStructTransform", err.Error(), nil)
@@ -697,6 +737,20 @@ func (u *CronjobService) Update(id uint, req dto.CronjobOperate, operator string
 	if err != nil {
 		return buserr.New("ErrRecordNotFound")
 	}
+	if req.AlertCount != 0 {
+		previous, _ := alertRepo.Get(alertRepo.WithByType(cronModel.Type), alertRepo.WithByProject(strconv.Itoa(int(id))))
+		merged, err := prepareCronJobAlertParams(cronModel.Type, previous.AdvancedParams, advanced)
+		if err != nil {
+			return err
+		}
+		method := req.AlertMethod
+		if method == "" {
+			method = previous.Method
+		}
+		if err := (AlertService{}).validateCronJobAlertChannels(cronModel.Type, merged, method); err != nil {
+			return err
+		}
+	}
 	upMap := make(map[string]interface{})
 	cronjob.EntryIDs = cronModel.EntryIDs
 	cronjob.Type = cronModel.Type
@@ -753,11 +807,12 @@ func (u *CronjobService) Update(id uint, req dto.CronjobOperate, operator string
 		return err
 	}
 	updateAlert := dto.AlertCreate{
-		Title:     req.AlertTitle,
-		SendCount: req.AlertCount,
-		Method:    req.AlertMethod,
-		Type:      cronjob.Type,
-		Project:   strconv.Itoa(int(cronModel.ID)),
+		Title:          req.AlertTitle,
+		SendCount:      req.AlertCount,
+		Method:         req.AlertMethod,
+		Type:           cronjob.Type,
+		Project:        strconv.Itoa(int(cronModel.ID)),
+		AdvancedParams: advanced,
 	}
 	err = NewIAlertService().ExternalUpdateAlert(updateAlert, operator)
 	if err != nil {
@@ -766,6 +821,17 @@ func (u *CronjobService) Update(id uint, req dto.CronjobOperate, operator string
 	return nil
 }
 
+func cronJobAlertAdvancedParams(mode string) (string, error) {
+	if mode == "" {
+		return "", nil
+	}
+	data, err := json.Marshal(map[string]string{"alertTriggerMode": mode})
+	if err != nil {
+		return "", err
+	}
+	return alertUtil.MergeCronJobAlertParams("", string(data))
+}
+
 func (u *CronjobService) UpdateStatus(id uint, status string) error {
 	cronjob, _ := cronjobRepo.Get(repo.WithByID(id))
 	if cronjob.ID == 0 {

agent/app/service/cronjob_backup.go

@@ -412,6 +412,7 @@ func addSkipTask(source string, taskItem *task.Task) {
 		taskItem.Log(i18n.GetMsgByKey("NoSuchResource"))
 		return nil
 	}, nil)
+	taskItem.SubTasks[len(taskItem.SubTasks)-1].StepAlias = cronJobSkippedStep
 }
 
 func loadDbsForJob(cronjob model.Cronjob) []DatabaseHelper {

agent/app/service/cronjob_helper.go

@@ -4,6 +4,7 @@ import (
 	"bufio"
 	"context"
 	"encoding/json"
+	"errors"
 	"fmt"
 	"io"
 	"net/http"
@@ -23,6 +24,7 @@ import (
 	"github.com/1Panel-dev/1Panel/agent/constant"
 	"github.com/1Panel-dev/1Panel/agent/global"
 	"github.com/1Panel-dev/1Panel/agent/i18n"
+	alertUtil "github.com/1Panel-dev/1Panel/agent/utils/alert"
 	"github.com/1Panel-dev/1Panel/agent/utils/cmd"
 	"github.com/1Panel-dev/1Panel/agent/utils/files"
 	"github.com/1Panel-dev/1Panel/agent/utils/ntp"
@@ -56,10 +58,11 @@ func (u *CronjobService) HandleJob(cronjob *model.Cronjob) {
 					_ = taskRepo.Save(context.Background(), taskItem.Task)
 				}
 				cronjobRepo.EndRecords(record, constant.StatusFailed, err.Error(), record.Records)
-				handleCronJobAlert(cronjob)
+				handleCronJobAlert(cronjob, cronJobAlertResult(taskItem, err))
 				return
 			}
 			cronjobRepo.EndRecords(record, constant.StatusSuccess, "", record.Records)
+			handleCronJobAlert(cronjob, cronJobAlertResult(taskItem, nil))
 		}()
 		return
 	}
@@ -70,19 +73,20 @@ func (u *CronjobService) HandleJob(cronjob *model.Cronjob) {
 			record.TaskID = ""
 		}
 		cronjobRepo.EndRecords(record, constant.StatusFailed, err.Error(), record.Records)
-		handleCronJobAlert(cronjob)
+		handleCronJobAlert(cronjob, cronJobAlertResult(taskItem, err))
 		return
 	}
 	go func() {
 		if err := taskItem.Execute(); err != nil {
-			taskItem, _ := taskRepo.GetFirst(taskRepo.WithByID(record.TaskID))
-			if len(taskItem.ID) == 0 {
+			storedTask, _ := taskRepo.GetFirst(taskRepo.WithByID(record.TaskID))
+			if len(storedTask.ID) == 0 {
 				record.TaskID = ""
 			}
 			cronjobRepo.EndRecords(record, constant.StatusFailed, err.Error(), record.Records)
-			handleCronJobAlert(cronjob)
+			handleCronJobAlert(cronjob, cronJobAlertResult(taskItem, err))
 		} else {
 			cronjobRepo.EndRecords(record, constant.StatusSuccess, "", record.Records)
+			handleCronJobAlert(cronjob, cronJobAlertResult(taskItem, nil))
 		}
 	}()
 }
@@ -482,8 +486,33 @@ func hasBackup(cronjobType string) bool {
 	return cronjobType == "app" || cronjobType == "database" || cronjobType == "website" || cronjobType == "directory" || cronjobType == "snapshot" || cronjobType == "log" || cronjobType == "cutWebsiteLog"
 }
 
-func handleCronJobAlert(cronjob *model.Cronjob) {
+const cronJobSkippedStep = "cronjob-skipped"
+
+func cronJobAlertResult(taskItem *task.Task, err error) string {
+	if errors.Is(err, context.Canceled) || taskItem.Task.Status == constant.StatusCanceled ||
+		(taskItem.TaskCtx != nil && taskItem.TaskCtx.Err() != nil) {
+		return ""
+	}
+	if err != nil {
+		return alertUtil.CronJobAlertFailed
+	}
+	if taskItem.Task.Status != constant.StatusSuccess {
+		return ""
+	}
+	for _, subTask := range taskItem.SubTasks {
+		if subTask.StepAlias != cronJobSkippedStep {
+			return alertUtil.CronJobAlertSuccess
+		}
+	}
+	return ""
+}
+
+func handleCronJobAlert(cronjob *model.Cronjob, result string) {
+	if result == "" {
+		return
+	}
 	pushAlert := dto.PushAlert{
+		Result:    result,
 		TaskName:  cronjob.Name,
 		AlertType: cronjob.Type,
 		EntryID:   cronjob.ID,

agent/app/service/dashboard.go

@@ -588,6 +588,9 @@ func loadAcceleratorInfo() ([]dto.GPUInfo, []dto.NPUInfo, []dto.XPUInfo) {
 		xpuData []dto.XPUInfo
 	)
 	for _, device := range snapshot.Devices {
+		if device.ParentID != "" {
+			continue
+		}
 		switch device.Kind {
 		case accelerator.KindGPU:
 			if device.GPU == nil {
@@ -597,7 +600,11 @@ func loadAcceleratorInfo() ([]dto.GPUInfo, []dto.NPUInfo, []dto.XPUInfo) {
 			if err := copier.Copy(&dataItem, device.GPU); err != nil {
 				continue
 			}
-			dataItem.PowerUsage = dataItem.PowerDraw + " / " + dataItem.MaxPowerLimit
+			dataItem.MaxPowerLimit = device.GPU.PowerLimit
+			dataItem.PowerUsage = dataItem.PowerDraw
+			if dataItem.MaxPowerLimit != "" {
+				dataItem.PowerUsage += " / " + dataItem.MaxPowerLimit
+			}
 			dataItem.MemoryUsage = dataItem.MemUsed + " / " + dataItem.MemTotal
 			gpuData = append(gpuData, dataItem)
 		case accelerator.KindNPU:

agent/app/service/entry.go

@@ -32,10 +32,11 @@ var (
 	scriptRepo  = repo.NewIScriptRepo()
 	cronjobRepo = repo.NewICronjobRepo()
 
-	hostRepo    = repo.NewIHostRepo()
-	ftpRepo     = repo.NewIFtpRepo()
-	clamRepo    = repo.NewIClamRepo()
-	monitorRepo = repo.NewIMonitorRepo()
+	hostRepo        = repo.NewIHostRepo()
+	ftpRepo         = repo.NewIFtpRepo()
+	clamRepo        = repo.NewIClamRepo()
+	monitorRepo     = repo.NewIMonitorRepo()
+	vllmMonitorRepo = &repo.VLLMMonitorRepo{}
 
 	settingRepo        = repo.NewISettingRepo()
 	forwardingRuleRepo = repo.NewIForwardingRuleRepo()

agent/app/service/image.go

@@ -28,7 +28,7 @@ import (
 	"github.com/docker/docker/api/types/container"
 	"github.com/docker/docker/api/types/image"
 	"github.com/docker/docker/api/types/registry"
-	"github.com/docker/docker/pkg/archive"
+	"github.com/moby/go-archive"
 )
 
 type ImageService struct{}

agent/app/service/monitor.go

@@ -33,11 +33,14 @@ import (
 type MonitorService struct {
 	DiskIO chan ([]disk.IOCountersStat)
 	NetIO  chan ([]net.IOCountersStat)
+	ctx    context.Context
 }
 
 var (
-	monitorCancel context.CancelFunc
-	hostSysPath   = loadHostSysPath()
+	monitorSettingMutex sync.Mutex
+	gpuMonitorMutex     sync.Mutex
+	monitorCancel       context.CancelFunc
+	hostSysPath         = loadHostSysPath()
 
 	blockDevicePartitionCache sync.Map
 )
@@ -47,8 +50,13 @@ type IMonitorService interface {
 	LoadMonitorData(req dto.MonitorSearch) ([]dto.MonitorData, error)
 	LoadSetting() (*dto.MonitorSetting, error)
 	UpdateSetting(key, value string) error
-	CleanData() error
+	CleanData(monitorType string) error
 
+	LoadIOOptions() []string
+	LoadNetworkOptions() []string
+	LoadVLLMMonitorData(req dto.MonitorVLLMSearch) (dto.MonitorVLLMData, error)
+	LoadVLLMCurrent(ctx context.Context, req dto.MonitorVLLMCurrent) (model.MonitorVLLM, error)
+	CleanVLLMMonitor(req dto.MonitorVLLMClean) error
 	LoadGPUOptions() dto.MonitorGPUOptions
 	LoadGPUMonitorData(req dto.MonitorGPUSearch) (dto.MonitorGPUData, error)
 
@@ -60,6 +68,7 @@ func NewIMonitorService() IMonitorService {
 	return &MonitorService{
 		DiskIO: make(chan []disk.IOCountersStat, 2),
 		NetIO:  make(chan []net.IOCountersStat, 2),
+		ctx:    context.Background(),
 	}
 }
 
@@ -85,7 +94,7 @@ func (m *MonitorService) LoadMonitorData(req dto.MonitorSearch) ([]dto.MonitorDa
 				base.TopCPUItems = processes
 				base.TopCPU = ""
 			}
-			if req.Param == "all" || req.Param == "mem" {
+			if req.Param == "all" || req.Param == "memory" {
 				var processes []dto.Process
 				_ = json.Unmarshal([]byte(base.TopMem), &processes)
 				base.TopMemItems = processes
@@ -96,7 +105,11 @@ func (m *MonitorService) LoadMonitorData(req dto.MonitorSearch) ([]dto.MonitorDa
 		data = append(data, itemData)
 	}
 	if req.Param == "all" || req.Param == "io" {
-		bases, err := monitorRepo.GetIO(repo.WithByName(req.IO), repo.WithByCreatedAt(req.StartTime, req.EndTime))
+		ioOpts := []repo.DBOption{repo.WithByCreatedAt(req.StartTime, req.EndTime)}
+		if len(req.IO) != 0 {
+			ioOpts = append(ioOpts, repo.WithByName(req.IO))
+		}
+		bases, err := monitorRepo.GetIO(ioOpts...)
 		if err != nil {
 			return nil, err
 		}
@@ -110,7 +123,11 @@ func (m *MonitorService) LoadMonitorData(req dto.MonitorSearch) ([]dto.MonitorDa
 		data = append(data, itemData)
 	}
 	if req.Param == "all" || req.Param == "network" {
-		bases, err := monitorRepo.GetNetwork(repo.WithByName(req.Network), repo.WithByCreatedAt(req.StartTime, req.EndTime))
+		netOpts := []repo.DBOption{repo.WithByCreatedAt(req.StartTime, req.EndTime)}
+		if len(req.Network) != 0 {
+			netOpts = append(netOpts, repo.WithByName(req.Network))
+		}
+		bases, err := monitorRepo.GetNetwork(netOpts...)
 		if err != nil {
 			return nil, err
 		}
@@ -126,117 +143,171 @@ func (m *MonitorService) LoadMonitorData(req dto.MonitorSearch) ([]dto.MonitorDa
 	return data, nil
 }
 
+func (m *MonitorService) LoadIOOptions() []string {
+	optionSet := make(map[string]struct{})
+	if diskStat, err := disk.IOCounters(); err == nil {
+		for _, item := range diskStat {
+			optionSet[item.Name] = struct{}{}
+		}
+	}
+	// union with names recorded in the monitor db so removed devices stay selectable
+	if names, err := monitorRepo.GetIONames(); err == nil {
+		for _, name := range names {
+			optionSet[name] = struct{}{}
+		}
+	}
+	return sortedMonitorOptions(optionSet)
+}
+
+func (m *MonitorService) LoadNetworkOptions() []string {
+	optionSet := make(map[string]struct{})
+	if netStat, err := net.IOCounters(true); err == nil {
+		for _, item := range netStat {
+			optionSet[item.Name] = struct{}{}
+		}
+	}
+	if names, err := monitorRepo.GetNetworkNames(); err == nil {
+		for _, name := range names {
+			optionSet[name] = struct{}{}
+		}
+	}
+	return sortedMonitorOptions(optionSet)
+}
+
 func (m *MonitorService) LoadGPUOptions() dto.MonitorGPUOptions {
 	var data dto.MonitorGPUOptions
 	exist, client := accelerator.New()
 	if !exist {
 		return data
 	}
+	seen := make(map[string]bool)
 	snapshot, err := client.Collect(context.Background())
 	if err != nil {
-		global.LOG.Errorf("Load accelerator info failed, err: %v", err)
+		global.LOG.Warnf("Load accelerator options failed: %v", err)
+	} else {
+		data = loadGPUOptions(snapshot)
+		for _, item := range data.ChartHide {
+			seen[item.DeviceID] = true
+		}
+	}
+	data.Supported = true
+	devices, err := monitorRepo.GetGPUDevices()
+	if err != nil {
+		global.LOG.Warnf("Load accelerator history options failed: %v", err)
 		return data
 	}
-	if warning := snapshot.Warning(); warning != nil {
-		global.LOG.Warnf("Load accelerator info partially failed, err: %v", warning)
-	}
-	return loadGPUOptions(snapshot)
-}
-
-func loadGPUOptions(snapshot *accelerator.Snapshot) dto.MonitorGPUOptions {
-	var data dto.MonitorGPUOptions
-	hasGPUOrNPU := false
-	hasXPU := false
-	for _, item := range snapshot.Devices {
-		if item.Kind == accelerator.KindXPU {
-			hasXPU = true
-		} else {
-			hasGPUOrNPU = true
+	for _, device := range devices {
+		key := device.DeviceID
+		if key == "" {
+			key = "legacy:" + device.ProductName
 		}
-	}
-	switch {
-	case hasGPUOrNPU && hasXPU:
-		data.GPUType = "mixed"
-	case hasXPU:
-		data.GPUType = "xpu"
-	case hasGPUOrNPU:
-		data.GPUType = "gpu"
-	}
-
-	sort.Slice(snapshot.Devices, func(i, j int) bool {
-		if snapshot.Devices[i].Kind != snapshot.Devices[j].Kind {
-			return snapshot.Devices[i].Kind < snapshot.Devices[j].Kind
+		if seen[key] {
+			continue
 		}
-		if snapshot.Devices[i].Vendor != snapshot.Devices[j].Vendor {
-			return snapshot.Devices[i].Vendor < snapshot.Devices[j].Vendor
-		}
-		if snapshot.Devices[i].NPUIndex != snapshot.Devices[j].NPUIndex {
-			return snapshot.Devices[i].NPUIndex < snapshot.Devices[j].NPUIndex
-		}
-		if snapshot.Devices[i].ChipIndex != snapshot.Devices[j].ChipIndex {
-			return snapshot.Devices[i].ChipIndex < snapshot.Devices[j].ChipIndex
-		}
-		return snapshot.Devices[i].Index < snapshot.Devices[j].Index
-	})
-	for _, item := range snapshot.Devices {
-		optionType := "gpu"
-		if item.Kind == accelerator.KindXPU {
-			optionType = "xpu"
-		}
-		chartHide := dto.GPUChartHide{
-			ProductName: item.Label,
-			Type:        optionType,
-			GPU:         !item.Capabilities.Utilization,
-			Memory:      !item.Capabilities.Memory,
-			Power:       !item.Capabilities.Power,
-			PowerLimit:  !item.Capabilities.PowerLimit,
-			Temperature: !item.Capabilities.Temperature,
-			Speed:       !item.Capabilities.FanSpeed,
-		}
-		data.ChartHide = append(data.ChartHide, chartHide)
-		data.Options = append(data.Options, chartHide.ProductName)
+		seen[key] = true
+		data.ChartHide = append(data.ChartHide, dto.GPUChartHide{DeviceID: device.DeviceID, ProductName: device.ProductName, Type: device.DeviceType, Legacy: device.DeviceID == ""})
+		data.Options = append(data.Options, device.ProductName)
 	}
 	return data
 }
 
 func (m *MonitorService) LoadGPUMonitorData(req dto.MonitorGPUSearch) (dto.MonitorGPUData, error) {
-	loc, _ := time.LoadLocation(common.LoadTimeZoneByCmd())
-	req.StartTime = req.StartTime.In(loc)
-	req.EndTime = req.EndTime.In(loc)
 	var data dto.MonitorGPUData
-	gpuList, err := monitorRepo.GetGPU(repo.WithByCreatedAt(req.StartTime, req.EndTime), monitorRepo.WithByProductName(req.ProductName))
+	if req.StartTime.IsZero() || req.EndTime.IsZero() || !req.EndTime.After(req.StartTime) {
+		return data, fmt.Errorf("invalid GPU history time range")
+	}
+	if req.DeviceID == "" && req.ProductName == "" {
+		return data, fmt.Errorf("GPU history requires a device")
+	}
+	if req.Aggregation != "" && req.Aggregation != "avg" && req.Aggregation != "max" {
+		return data, fmt.Errorf("invalid GPU history aggregation")
+	}
+	loc, err := time.LoadLocation(common.LoadTimeZoneByCmd())
 	if err != nil {
 		return data, err
 	}
-
-	for _, gpu := range gpuList {
-		data.Date = append(data.Date, gpu.CreatedAt)
-		data.GPUValue = append(data.GPUValue, gpu.GPUUtil)
-		data.TemperatureValue = append(data.TemperatureValue, gpu.Temperature)
-		data.PowerUsed = append(data.PowerUsed, gpu.PowerDraw)
-		data.PowerTotal = append(data.PowerTotal, gpu.MaxPowerLimit)
-		if gpu.MaxPowerLimit != 0 {
-			data.PowerPercent = append(data.PowerPercent, gpu.PowerDraw/gpu.MaxPowerLimit*100)
-		} else {
-			data.PowerPercent = append(data.PowerPercent, float64(0))
+	req.StartTime, req.EndTime = req.StartTime.In(loc), req.EndTime.In(loc)
+	opts := []repo.DBOption{repo.WithByCreatedAt(req.StartTime, req.EndTime), monitorRepo.WithByGPUDevice(req.DeviceID, req.ProductName, req.Legacy)}
+	data.SampleCount, err = monitorRepo.CountGPU(opts...)
+	if err != nil || data.SampleCount == 0 {
+		return data, err
+	}
+	if data.SampleCount > 1200 {
+		seconds := req.EndTime.Unix() - req.StartTime.Unix() + 1
+		data.BucketSeconds = (seconds + 599) / 600
+	}
+	points, err := monitorRepo.GetGPUHistory(req.StartTime, data.BucketSeconds, req.Aggregation, opts...)
+	if err != nil {
+		return data, err
+	}
+	samples := make([]repo.GPUHistoryPoint, 0, len(points))
+	if data.BucketSeconds > 0 {
+		next := 0
+		for bucket := int64(0); bucket <= (req.EndTime.Unix()-req.StartTime.Unix())/data.BucketSeconds; bucket++ {
+			point := repo.GPUHistoryPoint{}
+			if next < len(points) && points[next].Bucket == bucket {
+				point = points[next]
+				next++
+			}
+			point.CreatedAt = time.Unix(req.StartTime.Unix()+bucket*data.BucketSeconds, 0).In(loc)
+			if bucket == 0 {
+				point.CreatedAt = req.StartTime
+			}
+			samples = append(samples, point)
 		}
-
-		data.MemoryTotal = append(data.MemoryTotal, gpu.MemTotal)
-		data.MemoryUsed = append(data.MemoryUsed, gpu.MemUsed)
-		if gpu.MemTotal != 0 {
-			data.MemoryPercent = append(data.MemoryPercent, gpu.MemUsed/gpu.MemTotal*100)
-		} else {
-			data.MemoryPercent = append(data.MemoryPercent, float64(0))
+	} else {
+		for i, point := range points {
+			if i > 0 && points[i-1].IntervalSeconds > 0 {
+				interval := time.Duration(points[i-1].IntervalSeconds) * time.Second
+				if point.CreatedAt.Sub(points[i-1].CreatedAt) > 2*interval {
+					samples = append(samples, repo.GPUHistoryPoint{MonitorGPU: model.MonitorGPU{BaseModel: model.BaseModel{CreatedAt: points[i-1].CreatedAt.Add(interval)}}})
+				}
+			}
+			samples = append(samples, point)
 		}
-		var process []dto.GPUProcess
-		if err := json.Unmarshal([]byte(gpu.Processes), &process); err == nil {
-			data.ProcessCount = append(data.ProcessCount, len(process))
-			data.GPUProcesses = append(data.GPUProcesses, process)
-		} else {
-			data.ProcessCount = append(data.ProcessCount, 0)
-			data.GPUProcesses = append(data.GPUProcesses, []dto.GPUProcess{})
+	}
+	for _, point := range samples {
+		data.Date = append(data.Date, point.CreatedAt)
+		data.MemoryActivity = append(data.MemoryActivity, point.MemoryActivity)
+		data.EncoderUtil = append(data.EncoderUtil, point.EncoderUtil)
+		data.DecoderUtil = append(data.DecoderUtil, point.DecoderUtil)
+		data.JPEGUtil = append(data.JPEGUtil, point.JPEGUtil)
+		data.OFAUtil = append(data.OFAUtil, point.OFAUtil)
+		data.MediaUtil = append(data.MediaUtil, point.MediaUtil)
+		data.ComputeUtil = append(data.ComputeUtil, point.ComputeUtil)
+		data.CopyUtil = append(data.CopyUtil, point.CopyUtil)
+		data.HotspotTemperature = append(data.HotspotTemperature, point.HotspotTemperature)
+		data.FanRPM = append(data.FanRPM, point.FanRPM)
+		data.AICPUUtil = append(data.AICPUUtil, point.AICPUUtil)
+		data.CtrlCPUUtil = append(data.CtrlCPUUtil, point.CtrlCPUUtil)
+		data.DDRUsed = append(data.DDRUsed, point.DDRUsed)
+		data.DDRTotal = append(data.DDRTotal, point.DDRTotal)
+		data.HBMUsed = append(data.HBMUsed, point.HBMUsed)
+		data.HBMTotal = append(data.HBMTotal, point.HBMTotal)
+		data.DDRBandwidth = append(data.DDRBandwidth, point.DDRBandwidth)
+		data.HBMBandwidth = append(data.HBMBandwidth, point.HBMBandwidth)
+		data.MemoryBandwidth = append(data.MemoryBandwidth, point.MemoryBandwidth)
+		data.MediaFrequency = append(data.MediaFrequency, point.MediaFrequency)
+		data.HugepagesUsed = append(data.HugepagesUsed, point.HugepagesUsed)
+		data.HugepagesTotal = append(data.HugepagesTotal, point.HugepagesTotal)
+		data.GPUValue = append(data.GPUValue, point.GPUUtil)
+		data.TemperatureValue = append(data.TemperatureValue, point.Temperature)
+		data.MemoryTemperatureValue = append(data.MemoryTemperatureValue, point.MemoryTemperature)
+		data.PowerUsed = append(data.PowerUsed, point.PowerDraw)
+		data.PowerTotal = append(data.PowerTotal, point.MaxPowerLimit)
+		data.PowerPercent = append(data.PowerPercent, point.PowerPercent)
+		data.MemoryPercent = append(data.MemoryPercent, point.MemoryPercent)
+		data.MemoryTotal = append(data.MemoryTotal, point.MemTotal)
+		data.MemoryUsed = append(data.MemoryUsed, point.MemUsed)
+		data.SpeedValue = append(data.SpeedValue, point.FanSpeed)
+		data.FrequencyValue = append(data.FrequencyValue, point.Frequency)
+		data.MemoryFrequencyValue = append(data.MemoryFrequencyValue, point.MemoryFrequency)
+		data.ProcessCount = append(data.ProcessCount, point.ProcessCount)
+		var processes []dto.GPUProcess
+		if data.BucketSeconds == 0 && point.ProcessCount != nil {
+			_ = json.Unmarshal([]byte(point.Processes), &processes)
 		}
-		data.SpeedValue = append(data.SpeedValue, gpu.FanSpeed)
+		data.GPUProcesses = append(data.GPUProcesses, processes)
 	}
 	return data, nil
 }
@@ -262,9 +333,27 @@ func (m *MonitorService) LoadSetting() (*dto.MonitorSetting, error) {
 }
 
 func (m *MonitorService) UpdateSetting(key, value string) error {
+	monitorSettingMutex.Lock()
+	defer monitorSettingMutex.Unlock()
+	switch key {
+	case "MonitorStatus", "GPUMonitorStatus", "VLLMMonitorStatus":
+		if value != constant.StatusEnable && value != constant.StatusDisable {
+			return fmt.Errorf("invalid monitoring status")
+		}
+	case "MonitorInterval", "GPUMonitorInterval", "VLLMMonitorInterval":
+		interval, err := strconv.Atoi(value)
+		if err != nil || interval < 10 || interval > 43200 {
+			return fmt.Errorf("monitoring interval must be between 10 and 43200 seconds")
+		}
+	case "MonitorStoreDays", "GPUMonitorStoreDays", "VLLMMonitorStoreDays":
+		days, err := strconv.Atoi(value)
+		if err != nil || days < 1 {
+			return fmt.Errorf("monitoring retention must be a positive integer")
+		}
+	}
 	switch key {
 	case "MonitorStatus":
-		if value == constant.StatusEnable && global.MonitorCronID == 0 {
+		if value == constant.StatusEnable && monitorCancel == nil {
 			interval, err := settingRepo.Get(settingRepo.WithByKey("MonitorInterval"))
 			if err != nil {
 				return err
@@ -273,8 +362,9 @@ func (m *MonitorService) UpdateSetting(key, value string) error {
 				return err
 			}
 		}
-		if value == constant.StatusDisable && global.MonitorCronID != 0 {
+		if value == constant.StatusDisable && monitorCancel != nil {
 			monitorCancel()
+			monitorCancel = nil
 			global.Cron.Remove(cron.EntryID(global.MonitorCronID))
 			global.MonitorCronID = 0
 		}
@@ -283,31 +373,80 @@ func (m *MonitorService) UpdateSetting(key, value string) error {
 		if err != nil {
 			return err
 		}
-		if status.Value == constant.StatusEnable && global.MonitorCronID != 0 {
+		if status.Value == constant.StatusEnable && monitorCancel != nil {
 			if err := StartMonitor(true, value); err != nil {
 				return err
 			}
 		}
+	case "GPUMonitorStatus":
+		if value == constant.StatusEnable && global.GPUMonitorCronID == 0 {
+			interval, err := settingRepo.GetValueByKey("GPUMonitorInterval")
+			if err != nil {
+				return err
+			}
+			if err := StartGPUMonitor(interval); err != nil {
+				return err
+			}
+		}
+		if value == constant.StatusDisable && global.GPUMonitorCronID != 0 {
+			global.Cron.Remove(global.GPUMonitorCronID)
+			global.GPUMonitorCronID = 0
+		}
+	case "GPUMonitorInterval":
+		status, err := settingRepo.GetValueByKey("GPUMonitorStatus")
+		if err != nil {
+			return err
+		}
+		if status == constant.StatusEnable {
+			if err := StartGPUMonitor(value); err != nil {
+				return err
+			}
+		}
+	case "VLLMMonitorStatus":
+		if value == constant.StatusEnable && global.VLLMMonitorCronID == 0 {
+			interval, err := settingRepo.GetValueByKey("VLLMMonitorInterval")
+			if err != nil {
+				return err
+			}
+			if err := StartVLLMMonitor(interval); err != nil {
+				return err
+			}
+		}
+		if value == constant.StatusDisable && global.VLLMMonitorCronID != 0 {
+			global.Cron.Remove(global.VLLMMonitorCronID)
+			global.VLLMMonitorCronID = 0
+		}
+	case "VLLMMonitorInterval":
+		status, err := settingRepo.GetValueByKey("VLLMMonitorStatus")
+		if err != nil {
+			return err
+		}
+		if status == constant.StatusEnable {
+			if err := StartVLLMMonitor(value); err != nil {
+				return err
+			}
+		}
 	}
 	return settingRepo.Update(key, value)
 }
 
-func (m *MonitorService) CleanData() error {
-	if err := global.MonitorDB.Exec("DELETE FROM monitor_bases").Error; err != nil {
-		return err
+func (m *MonitorService) CleanData(monitorType string) error {
+	switch monitorType {
+	case "host":
+		return monitorRepo.CleanHost()
+	case "gpu":
+		gpuMonitorMutex.Lock()
+		defer gpuMonitorMutex.Unlock()
+		return monitorRepo.CleanGPU()
+	default:
+		return fmt.Errorf("unsupported monitoring cleanup type: %s", monitorType)
 	}
-	if err := global.MonitorDB.Exec("DELETE FROM monitor_ios").Error; err != nil {
-		return err
-	}
-	if err := global.MonitorDB.Exec("DELETE FROM monitor_networks").Error; err != nil {
-		return err
-	}
-	_ = global.GPUMonitorDB.Exec("DELETE FROM monitor_gpus").Error
-	return nil
 }
 
 func (m *MonitorService) Run() {
-	saveAcceleratorDataToDB()
+	if m.ctx.Err() != nil {
+		return
+	}
 	var itemModel model.MonitorBase
 	totalPercent, _ := cpu.Percent(3*time.Second, false)
 	if len(totalPercent) == 1 {
@@ -353,30 +492,34 @@ func (m *MonitorService) Run() {
 	_ = monitorRepo.DelMonitorBase(timeForDelete)
 	_ = monitorRepo.DelMonitorIO(timeForDelete)
 	_ = monitorRepo.DelMonitorNet(timeForDelete)
-	_ = monitorRepo.DelMonitorGPU(timeForDelete)
 }
 
 func (m *MonitorService) loadDiskIO() {
-	ioStat, _ := disk.IOCounters()
+	ioStat, _ := disk.IOCountersWithContext(m.ctx)
 	var diskIOList []disk.IOCountersStat
 	for _, io := range ioStat {
 		diskIOList = append(diskIOList, io)
 	}
 	diskIOList = append(diskIOList, sumDiskIOCounters(ioStat))
-	m.DiskIO <- diskIOList
+	select {
+	case <-m.ctx.Done():
+	case m.DiskIO <- diskIOList:
+	}
 }
 
 func (m *MonitorService) loadNetIO() {
-	netStat, _ := net.IOCounters(true)
-	netStatAll, _ := net.IOCounters(false)
+	netStat, _ := net.IOCountersWithContext(m.ctx, true)
+	netStatAll, _ := net.IOCountersWithContext(m.ctx, false)
 	var netList []net.IOCountersStat
 	netList = append(netList, netStat...)
 	netList = append(netList, netStatAll...)
-	m.NetIO <- netList
+	select {
+	case <-m.ctx.Done():
+	case m.NetIO <- netList:
+	}
 }
 
 func (m *MonitorService) saveIODataToDB(ctx context.Context, interval float64) {
-	defer close(m.DiskIO)
 	for {
 		select {
 		case <-ctx.Done():
@@ -426,14 +569,17 @@ func (m *MonitorService) saveIODataToDB(ctx context.Context, interval float64) {
 					}
 				}
 				_ = monitorRepo.BatchCreateMonitorIO(ioList)
-				m.DiskIO <- ioStat2
+				select {
+				case <-ctx.Done():
+					return
+				case m.DiskIO <- ioStat2:
+				}
 			}
 		}
 	}
 }
 
 func (m *MonitorService) saveNetDataToDB(ctx context.Context, interval float64) {
-	defer close(m.NetIO)
 	for {
 		select {
 		case <-ctx.Done():
@@ -463,7 +609,11 @@ func (m *MonitorService) saveNetDataToDB(ctx context.Context, interval float64)
 				}
 
 				_ = monitorRepo.BatchCreateMonitorNet(netList)
-				m.NetIO <- netStat2
+				select {
+				case <-ctx.Done():
+					return
+				case m.NetIO <- netStat2:
+				}
 			}
 		}
 	}
@@ -572,27 +722,38 @@ func loadTopMem() []dto.Process {
 }
 
 func StartMonitor(removeBefore bool, interval string) error {
+	intervalItem, err := strconv.Atoi(interval)
+	if err != nil || intervalItem < 10 || intervalItem > 43200 {
+		return fmt.Errorf("invalid host monitoring interval: %s", interval)
+	}
 	if removeBefore {
 		monitorCancel()
 		global.Cron.Remove(cron.EntryID(global.MonitorCronID))
 	}
-	intervalItem, err := strconv.Atoi(interval)
-	if err != nil {
-		return err
-	}
-
-	service := NewIMonitorService()
 	ctx, cancel := context.WithCancel(context.Background())
-	monitorCancel = cancel
+	service := &MonitorService{
+		DiskIO: make(chan []disk.IOCountersStat, 2),
+		NetIO:  make(chan []net.IOCountersStat, 2),
+		ctx:    ctx,
+	}
 	now := time.Now()
 	nextMinute := now.Truncate(time.Minute).Add(time.Minute)
-	time.AfterFunc(time.Until(nextMinute), func() {
+	timer := time.AfterFunc(time.Until(nextMinute), func() {
+		monitorSettingMutex.Lock()
+		defer monitorSettingMutex.Unlock()
+		if ctx.Err() != nil {
+			return
+		}
 		monitorID, err := global.Cron.AddJob(fmt.Sprintf("@every %ss", interval), service)
 		if err != nil {
 			return
 		}
 		global.MonitorCronID = monitorID
 	})
+	monitorCancel = func() {
+		cancel()
+		timer.Stop()
+	}
 
 	service.Run()
 
@@ -602,7 +763,97 @@ func StartMonitor(removeBefore bool, interval string) error {
 	return nil
 }
 
-func saveAcceleratorDataToDB() {
+func loadGPUOptions(snapshot *accelerator.Snapshot) dto.MonitorGPUOptions {
+	var data dto.MonitorGPUOptions
+	hasGPUOrNPU := false
+	hasXPU := false
+	for _, item := range snapshot.Devices {
+		if item.Kind == accelerator.KindXPU {
+			hasXPU = true
+		} else {
+			hasGPUOrNPU = true
+		}
+	}
+	switch {
+	case hasGPUOrNPU && hasXPU:
+		data.GPUType = "mixed"
+	case hasXPU:
+		data.GPUType = "xpu"
+	case hasGPUOrNPU:
+		data.GPUType = "gpu"
+	}
+
+	sort.Slice(snapshot.Devices, func(i, j int) bool {
+		if snapshot.Devices[i].Kind != snapshot.Devices[j].Kind {
+			return snapshot.Devices[i].Kind < snapshot.Devices[j].Kind
+		}
+		if snapshot.Devices[i].Vendor != snapshot.Devices[j].Vendor {
+			return snapshot.Devices[i].Vendor < snapshot.Devices[j].Vendor
+		}
+		if snapshot.Devices[i].NPUIndex != snapshot.Devices[j].NPUIndex {
+			return snapshot.Devices[i].NPUIndex < snapshot.Devices[j].NPUIndex
+		}
+		if snapshot.Devices[i].ChipIndex != snapshot.Devices[j].ChipIndex {
+			return snapshot.Devices[i].ChipIndex < snapshot.Devices[j].ChipIndex
+		}
+		return snapshot.Devices[i].Index < snapshot.Devices[j].Index
+	})
+	for _, item := range snapshot.Devices {
+		chartHide := dto.GPUChartHide{
+			DeviceID:    item.ID,
+			ProductName: item.Label,
+			Type:        string(item.Kind),
+		}
+		data.ChartHide = append(data.ChartHide, chartHide)
+		data.Options = append(data.Options, chartHide.ProductName)
+	}
+	return data
+}
+
+func StartGPUMonitor(interval string) error {
+	seconds, err := strconv.Atoi(interval)
+	if err != nil || seconds < 10 || seconds > 43200 {
+		return fmt.Errorf("invalid GPU monitoring interval: %s", interval)
+	}
+	service := &MonitorService{}
+	job := cron.NewChain(cron.Recover(cron.DefaultLogger)).Then(cron.FuncJob(service.saveGPUData))
+	id, err := global.Cron.AddFunc(fmt.Sprintf("@every %ds", seconds), func() { go job.Run() })
+	if err != nil {
+		return err
+	}
+	if global.GPUMonitorCronID != 0 {
+		global.Cron.Remove(global.GPUMonitorCronID)
+	}
+	global.GPUMonitorCronID = id
+	return nil
+}
+
+func (m *MonitorService) saveGPUData() {
+	if !gpuMonitorMutex.TryLock() {
+		return
+	}
+	defer gpuMonitorMutex.Unlock()
+	status, err := settingRepo.GetValueByKey("GPUMonitorStatus")
+	if err != nil {
+		global.LOG.Errorf("Load GPU monitoring status failed: %v", err)
+		return
+	}
+	if status != constant.StatusEnable {
+		return
+	}
+	retention, err := settingRepo.GetValueByKey("GPUMonitorStoreDays")
+	if err != nil {
+		global.LOG.Errorf("Load GPU monitoring retention failed: %v", err)
+		return
+	}
+	days, err := strconv.Atoi(retention)
+	if err != nil || days < 1 {
+		global.LOG.Errorf("Invalid GPU monitoring retention: %s", retention)
+		return
+	}
+	if err := monitorRepo.DelMonitorGPU(time.Now().AddDate(0, 0, -days)); err != nil {
+		global.LOG.Errorf("Clean GPU monitoring data failed: %v", err)
+	}
 	exist, client := accelerator.New()
 	if !exist {
 		return
@@ -615,27 +866,64 @@ func saveAcceleratorDataToDB() {
 	if warning := snapshot.Warning(); warning != nil {
 		global.LOG.Warnf("load accelerator monitor data partially failed, err: %v", warning)
 	}
+	intervalSeconds := 0
+	if setting, err := settingRepo.Get(settingRepo.WithByKey("GPUMonitorInterval")); err == nil {
+		intervalSeconds, _ = strconv.Atoi(setting.Value)
+	}
 	list := make([]model.MonitorGPU, 0, len(snapshot.Devices))
 	for _, device := range snapshot.Devices {
-		list = append(list, newMonitorGPU(device))
+		item := newMonitorGPU(device)
+		item.CreatedAt = snapshot.Info.CollectedAt
+		item.IntervalSeconds = intervalSeconds
+		list = append(list, item)
 	}
-	if err := repo.NewIMonitorRepo().BatchCreateMonitorGPU(list); err != nil {
+	if err := monitorRepo.BatchCreateMonitorGPU(list); err != nil {
 		global.LOG.Errorf("batch create accelerator monitor data failed, err: %v", err)
 	}
 }
 
 func newMonitorGPU(device accelerator.Device) model.MonitorGPU {
 	item := model.MonitorGPU{
-		ProductName:   device.Label,
-		GPUUtil:       device.Metrics.Utilization.ValueOrZero(),
-		Temperature:   device.Metrics.Temperature.ValueOrZero(),
-		PowerDraw:     device.Metrics.Power.ValueOrZero(),
-		MaxPowerLimit: device.Metrics.PowerLimit.ValueOrZero(),
-		MemUsed:       device.Metrics.MemoryUsed.ValueOrZero(),
-		MemTotal:      device.Metrics.MemoryTotal.ValueOrZero(),
-		FanSpeed:      int(device.Metrics.FanSpeed.ValueOrZero()),
+		MemoryUtil:         device.Metrics.MemoryUtil.Value,
+		MemoryActivity:     device.Metrics.MemoryActivity.Value,
+		EncoderUtil:        device.Metrics.EncoderUtil.Value,
+		DecoderUtil:        device.Metrics.DecoderUtil.Value,
+		JPEGUtil:           device.Metrics.JPEGUtil.Value,
+		OFAUtil:            device.Metrics.OFAUtil.Value,
+		MediaUtil:          device.Metrics.MediaUtil.Value,
+		ComputeUtil:        device.Metrics.ComputeUtil.Value,
+		CopyUtil:           device.Metrics.CopyUtil.Value,
+		HotspotTemperature: device.Metrics.HotspotTemperature.Value,
+		FanRPM:             device.Metrics.FanRPM.Value,
+		AICPUUtil:          device.Metrics.AICPUUtil.Value,
+		CtrlCPUUtil:        device.Metrics.CtrlCPUUtil.Value,
+		DDRUsed:            device.Metrics.DDRUsed.Value,
+		DDRTotal:           device.Metrics.DDRTotal.Value,
+		HBMUsed:            device.Metrics.HBMUsed.Value,
+		HBMTotal:           device.Metrics.HBMTotal.Value,
+		DDRBandwidth:       device.Metrics.DDRBandwidth.Value,
+		HBMBandwidth:       device.Metrics.HBMBandwidth.Value,
+		MemoryBandwidth:    device.Metrics.MemoryBandwidth.Value,
+		MediaFrequency:     device.Metrics.MediaFrequency.Value,
+		HugepagesUsed:      device.Metrics.HugepagesUsed.Value,
+		HugepagesTotal:     device.Metrics.HugepagesTotal.Value,
+
+		ProductName:       device.Label,
+		DeviceID:          device.ID,
+		DeviceType:        string(device.Kind),
+		ProcessStatus:     device.ProcessStatus,
+		Frequency:         device.Metrics.Frequency.Value,
+		MemoryFrequency:   device.Metrics.MemoryFrequency.Value,
+		MemoryTemperature: device.Metrics.MemoryTemperature.Value,
+		GPUUtil:           device.Metrics.Utilization.Value,
+		Temperature:       device.Metrics.Temperature.Value,
+		PowerDraw:         device.Metrics.Power.Value,
+		MaxPowerLimit:     device.Metrics.PowerLimit.Value,
+		MemUsed:           device.Metrics.MemoryUsed.Value,
+		MemTotal:          device.Metrics.MemoryTotal.Value,
+		FanSpeed:          device.Metrics.FanSpeed.Value,
 	}
-	if len(device.Processes) == 0 {
+	if device.ProcessStatus != "ok" {
 		return item
 	}
 	processes := make([]dto.GPUProcess, 0, len(device.Processes))
@@ -695,3 +983,14 @@ func loadHostSysPath() string {
 	}
 	return hostSys
 }
+
+func sortedMonitorOptions(optionSet map[string]struct{}) []string {
+	options := make([]string, 0, len(optionSet))
+	for name := range optionSet {
+		if len(name) != 0 && name != "all" {
+			options = append(options, name)
+		}
+	}
+	sort.Strings(options)
+	return append([]string{"all"}, options...)
+}

agent/app/service/monitor_vllm.go

@@ -0,0 +1,287 @@
+package service
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net"
+	"net/http"
+	"strconv"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/1Panel-dev/1Panel/agent/app/dto"
+	"github.com/1Panel-dev/1Panel/agent/app/model"
+	"github.com/1Panel-dev/1Panel/agent/app/repo"
+	"github.com/1Panel-dev/1Panel/agent/constant"
+	"github.com/1Panel-dev/1Panel/agent/global"
+	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/vllm"
+	"github.com/1Panel-dev/1Panel/agent/utils/common"
+	"github.com/robfig/cron/v3"
+	"golang.org/x/sync/errgroup"
+)
+
+var vllmMonitorMutex sync.Mutex
+var vllmMetricsClient = &http.Client{
+	Timeout:       5 * time.Second,
+	Transport:     &http.Transport{DialContext: (&net.Dialer{Timeout: 3 * time.Second}).DialContext, IdleConnTimeout: 30 * time.Second},
+	CheckRedirect: func(req *http.Request, via []*http.Request) error { return http.ErrUseLastResponse },
+}
+
+func (m *MonitorService) LoadVLLMMonitorData(req dto.MonitorVLLMSearch) (dto.MonitorVLLMData, error) {
+	data := dto.MonitorVLLMData{Points: []model.MonitorVLLM{}}
+	if !req.EndTime.After(req.StartTime) {
+		return data, fmt.Errorf("invalid vLLM history request")
+	}
+	install, err := appInstallRepo.GetFirst(repo.WithByID(req.AppInstallID))
+	if err != nil {
+		return data, err
+	}
+	if install.App.Key != "vllm" {
+		return data, fmt.Errorf("not a vLLM installation")
+	}
+
+	loc, err := time.LoadLocation(common.LoadTimeZoneByCmd())
+	if err != nil {
+		return data, err
+	}
+	req.StartTime, req.EndTime = req.StartTime.In(loc), req.EndTime.In(loc)
+	data.SampleCount, err = vllmMonitorRepo.Count(req.AppInstallID, req.StartTime, req.EndTime)
+	if err != nil || data.SampleCount == 0 {
+		return data, err
+	}
+	if data.SampleCount > 1200 {
+		data.BucketSeconds = (req.EndTime.Unix() - req.StartTime.Unix() + 600) / 600
+	}
+	points, err := vllmMonitorRepo.History(req.AppInstallID, req.StartTime, req.EndTime, data.BucketSeconds, req.Aggregation)
+	if err != nil {
+		return data, err
+	}
+	if data.BucketSeconds > 0 {
+		if req.Aggregation != "max" {
+			for i := range points {
+				if err := vllm.AggregateHistograms(points[i].HistogramSamples, &points[i].MonitorVLLM); err != nil {
+					return data, err
+				}
+			}
+		}
+		next := 0
+		for bucket := int64(0); bucket <= (req.EndTime.Unix()-req.StartTime.Unix())/data.BucketSeconds; bucket++ {
+			var point model.MonitorVLLM
+			if next < len(points) && points[next].Bucket == bucket {
+				point = points[next].MonitorVLLM
+				next++
+			}
+			point.AppInstallID = req.AppInstallID
+			point.CreatedAt = time.Unix(req.StartTime.Unix()+bucket*data.BucketSeconds, 0).In(loc)
+			if bucket == 0 {
+				point.CreatedAt = req.StartTime
+			}
+			data.Points = append(data.Points, point)
+		}
+	} else {
+		setting, err := settingRepo.GetValueByKey("VLLMMonitorInterval")
+		if err != nil {
+			return data, err
+		}
+		intervalSeconds, err := strconv.Atoi(setting)
+		if err != nil || intervalSeconds <= 0 {
+			return data, fmt.Errorf("invalid vLLM monitoring interval: %s", setting)
+		}
+		interval := time.Duration(intervalSeconds) * time.Second
+		for i, point := range points {
+			if i > 0 && point.CreatedAt.Sub(points[i-1].CreatedAt) > 2*interval {
+				data.Points = append(data.Points, model.MonitorVLLM{CreatedAt: points[i-1].CreatedAt.Add(interval), AppInstallID: req.AppInstallID})
+			}
+			data.Points = append(data.Points, point.MonitorVLLM)
+		}
+	}
+	return data, nil
+}
+
+func (m *MonitorService) LoadVLLMCurrent(ctx context.Context, req dto.MonitorVLLMCurrent) (model.MonitorVLLM, error) {
+	install, err := appInstallRepo.GetFirst(repo.WithByID(req.AppInstallID))
+	if err != nil {
+		return model.MonitorVLLM{}, err
+	}
+	if install.App.Key != "vllm" {
+		return model.MonitorVLLM{}, fmt.Errorf("not a vLLM installation")
+	}
+	previous, err := collectVLLMMetrics(ctx, install, model.MonitorVLLM{})
+	if err != nil {
+		global.LOG.Debugf("Collect vLLM metrics on port %d failed: %v", install.HttpPort, err)
+		return previous, nil
+	}
+	timer := time.NewTimer(time.Second)
+	defer timer.Stop()
+	select {
+	case <-ctx.Done():
+		return model.MonitorVLLM{}, ctx.Err()
+	case <-timer.C:
+	}
+	point, err := collectVLLMMetrics(ctx, install, previous)
+	if err != nil {
+		global.LOG.Debugf("Collect vLLM metrics on port %d failed: %v", install.HttpPort, err)
+	}
+	return point, nil
+}
+
+func (m *MonitorService) CleanVLLMMonitor(req dto.MonitorVLLMClean) error {
+	install, err := appInstallRepo.GetFirst(repo.WithByID(req.AppInstallID))
+	if err != nil {
+		return err
+	}
+	if install.App.Key != "vllm" {
+		return fmt.Errorf("not a vLLM installation")
+	}
+	vllmMonitorMutex.Lock()
+	defer vllmMonitorMutex.Unlock()
+	return vllmMonitorRepo.CleanTarget(req.AppInstallID)
+}
+
+func StartVLLMMonitor(interval string) error {
+	seconds, err := strconv.Atoi(interval)
+	if err != nil || seconds < 10 || seconds > 43200 {
+		return fmt.Errorf("invalid vLLM monitoring interval: %s", interval)
+	}
+	service := &MonitorService{}
+	job := cron.NewChain(cron.Recover(cron.DefaultLogger)).Then(cron.FuncJob(service.saveVLLMData))
+	id, err := global.Cron.AddFunc(fmt.Sprintf("@every %ds", seconds), func() { go job.Run() })
+	if err != nil {
+		return err
+	}
+	if global.VLLMMonitorCronID != 0 {
+		global.Cron.Remove(global.VLLMMonitorCronID)
+	}
+	global.VLLMMonitorCronID = id
+	return nil
+}
+
+func (m *MonitorService) saveVLLMData() {
+	if !vllmMonitorMutex.TryLock() {
+		return
+	}
+	defer vllmMonitorMutex.Unlock()
+	status, err := settingRepo.GetValueByKey("VLLMMonitorStatus")
+	if err != nil {
+		global.LOG.Errorf("Load vLLM monitoring status failed: %v", err)
+		return
+	}
+	if status != constant.StatusEnable {
+		return
+	}
+	retention, err := settingRepo.GetValueByKey("VLLMMonitorStoreDays")
+	if err != nil {
+		global.LOG.Errorf("Load vLLM monitoring retention failed: %v", err)
+		return
+	}
+	days, err := strconv.Atoi(retention)
+	if err != nil || days < 1 {
+		global.LOG.Errorf("Invalid vLLM monitoring retention: %s", retention)
+		return
+	}
+	if err := vllmMonitorRepo.DeleteBefore(time.Now().AddDate(0, 0, -days)); err != nil {
+		global.LOG.Errorf("Clean vLLM monitoring data failed: %v", err)
+	}
+	apps, err := appRepo.GetBy(appRepo.WithKey("vllm"))
+	if err != nil {
+		global.LOG.Errorf("Load vLLM apps for monitoring failed: %v", err)
+		return
+	}
+	if len(apps) == 0 {
+		return
+	}
+	ids := make([]uint, 0, len(apps))
+	for _, app := range apps {
+		ids = append(ids, app.ID)
+	}
+	installs, err := appInstallRepo.ListBy(context.Background(), appInstallRepo.WithAppIdsIn(ids))
+	if err != nil {
+		global.LOG.Errorf("Load vLLM instances for monitoring failed: %v", err)
+		return
+	}
+	var group errgroup.Group
+	group.SetLimit(4)
+	for _, install := range installs {
+		group.Go(func() error {
+			previous, err := vllmMonitorRepo.Latest(install.ID)
+			if err != nil {
+				return err
+			}
+			point, err := collectVLLMMetrics(context.Background(), install, previous)
+			if err != nil {
+				global.LOG.Debugf("Collect vLLM metrics for %s failed: %v", install.Name, err)
+			}
+			return vllmMonitorRepo.Create(&point)
+		})
+	}
+	if err := group.Wait(); err != nil {
+		global.LOG.Errorf("Save vLLM monitoring data failed: %v", err)
+	}
+}
+
+func collectVLLMMetrics(ctx context.Context, install model.AppInstall, previous model.MonitorVLLM) (model.MonitorVLLM, error) {
+	point := model.MonitorVLLM{AppInstallID: install.ID, CreatedAt: time.Now(), Status: "unavailable"}
+	if install.HttpPort <= 0 || install.HttpPort > 65535 {
+		return point, fmt.Errorf("invalid vLLM port")
+	}
+	var env map[string]interface{}
+	if err := json.Unmarshal([]byte(install.Env), &env); err != nil {
+		return point, err
+	}
+	host, _ := env[constant.HostIP].(string)
+	host = strings.Trim(host, "[]")
+	switch host {
+	case "", "0.0.0.0":
+		host = "127.0.0.1"
+	case "::":
+		host = "::1"
+	}
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "http://"+net.JoinHostPort(host, strconv.Itoa(install.HttpPort))+"/metrics", nil)
+	if err != nil {
+		return point, err
+	}
+	response, err := vllmMetricsClient.Do(req)
+	if err != nil {
+		return point, err
+	}
+	defer response.Body.Close()
+	if response.StatusCode != http.StatusOK {
+		return point, fmt.Errorf("vLLM metrics returned HTTP %d", response.StatusCode)
+	}
+	body, err := io.ReadAll(io.LimitReader(response.Body, 8*1024*1024+1))
+	if err != nil {
+		return point, err
+	}
+	if len(body) > 8*1024*1024 {
+		return point, fmt.Errorf("vLLM metrics response too large")
+	}
+	metrics, err := vllm.Parse(bytes.NewReader(body))
+	if err != nil {
+		return point, err
+	}
+	point.CreatedAt = time.Now()
+	raw, err := json.Marshal(metrics)
+	if err != nil {
+		return point, err
+	}
+	var before vllm.Metrics
+	elapsed := point.CreatedAt.Sub(previous.CreatedAt).Seconds()
+	if previous.Status == "ok" && elapsed > 0 {
+		if err := json.Unmarshal([]byte(previous.RawMetrics), &before); err != nil {
+			return point, err
+		}
+	}
+	calculated, err := vllm.Calculate(metrics, before, elapsed)
+	if err != nil {
+		return point, err
+	}
+	calculated.AppInstallID = install.ID
+	calculated.CreatedAt = point.CreatedAt
+	calculated.RawMetrics = string(raw)
+	calculated.Status = "ok"
+	return calculated, nil
+}

agent/app/service/ssh.go

@@ -7,6 +7,7 @@ import (
 	"context"
 	"encoding/base64"
 	"encoding/json"
+	"encoding/pem"
 	"fmt"
 	"io"
 	"net"
@@ -37,6 +38,7 @@ import (
 	"github.com/1Panel-dev/1Panel/agent/utils/common"
 	"github.com/1Panel-dev/1Panel/agent/utils/re"
 	"github.com/pkg/errors"
+	"golang.org/x/crypto/ssh"
 )
 
 const sshPath = "/etc/ssh/sshd_config"
@@ -438,15 +440,33 @@ func (u *SSHService) SyncRootCert() error {
 			global.LOG.Errorf("read pubic key of %s for sync failed, err: %v", item, err)
 			continue
 		}
-		cert.EncryptionMode = loadEncryptioMode(string(pubItem))
+		publicKey, _, _, rest, err := ssh.ParseAuthorizedKey(pubItem)
+		if err != nil || len(bytes.TrimSpace(rest)) != 0 {
+			continue
+		}
+		cert.EncryptionMode = loadSSHKeyType(publicKey)
+		if cert.EncryptionMode == "" {
+			continue
+		}
 		rootCerts = append(rootCerts, cert)
 	}
 	return hostRepo.SyncCert(rootCerts)
 }
 
 func (u *SSHService) CreateRootCert(req dto.RootCertOperate) error {
-	if cmd.CheckIllegal(req.EncryptionMode, req.PassPhrase) {
-		return buserr.New("ErrCmdIllegal")
+	switch req.Mode {
+	case "generate":
+		switch req.EncryptionMode {
+		case "rsa", "ed25519", "ecdsa", "dsa":
+		default:
+			return buserr.WithName("ErrNotSupportType", req.EncryptionMode)
+		}
+	case "input", "import":
+		if err := validateSSHKeyPair(&req); err != nil {
+			return err
+		}
+	default:
+		return buserr.WithName("ErrNotSupportType", req.Mode)
 	}
 	certItem, _ := hostRepo.GetCert(repo.WithByName(req.Name))
 	if certItem.ID != 0 {
@@ -543,6 +563,9 @@ func (u *SSHService) CreateRootCert(req dto.RootCertOperate) error {
 }
 
 func (u *SSHService) EditRootCert(req dto.RootCertOperate) error {
+	if err := validateSSHKeyPair(&req); err != nil {
+		return err
+	}
 	currentUser, err := user.Current()
 	if err != nil {
 		return fmt.Errorf("load current user failed, err: %v", err)
@@ -1681,22 +1704,60 @@ func loadDate(currentYear int, DateStr string, nyc *time.Location) time.Time {
 	return itemDate
 }
 
-func loadEncryptioMode(content string) string {
-	if strings.HasPrefix(content, "ssh-rsa") {
+func loadSSHKeyType(publicKey ssh.PublicKey) string {
+	switch publicKey.Type() {
+	case ssh.KeyAlgoRSA:
 		return "rsa"
-	}
-	if strings.HasPrefix(content, "ssh-ed25519") {
+	case ssh.KeyAlgoED25519:
 		return "ed25519"
-	}
-	if strings.HasPrefix(content, "ssh-ecdsa") {
+	case ssh.KeyAlgoECDSA256, ssh.KeyAlgoECDSA384, ssh.KeyAlgoECDSA521:
 		return "ecdsa"
-	}
-	if strings.HasPrefix(content, "ssh-dsa") {
+	case ssh.KeyAlgoDSA:
 		return "dsa"
 	}
 	return ""
 }
 
+func validateSSHKeyPair(req *dto.RootCertOperate) error {
+	req.PublicKey = strings.TrimSpace(req.PublicKey)
+	req.PrivateKey = strings.TrimSpace(req.PrivateKey)
+	if req.PublicKey == "" || strings.ContainsAny(req.PublicKey, "\r\n") {
+		return buserr.New("ErrSSHPublicKey")
+	}
+	publicKey, comment, options, rest, err := ssh.ParseAuthorizedKey([]byte(req.PublicKey))
+	if err != nil || len(options) != 0 || len(bytes.TrimSpace(rest)) != 0 {
+		return buserr.New("ErrSSHPublicKey")
+	}
+	req.EncryptionMode = loadSSHKeyType(publicKey)
+	if req.EncryptionMode == "" {
+		return buserr.WithName("ErrNotSupportType", publicKey.Type())
+	}
+	block, rest := pem.Decode([]byte(req.PrivateKey))
+	if !strings.HasPrefix(req.PrivateKey, "-----BEGIN ") || block == nil || len(bytes.TrimSpace(rest)) != 0 {
+		return buserr.New("ErrSSHPrivateKey")
+	}
+	signer, err := ssh.ParsePrivateKey([]byte(req.PrivateKey))
+	var missingPassphrase *ssh.PassphraseMissingError
+	if errors.As(err, &missingPassphrase) {
+		signer, err = ssh.ParsePrivateKeyWithPassphrase([]byte(req.PrivateKey), []byte(req.PassPhrase))
+	} else if err == nil {
+		req.PassPhrase = ""
+	}
+	if err != nil {
+		return buserr.New("ErrSSHPrivateKey")
+	}
+	if !bytes.Equal(signer.PublicKey().Marshal(), publicKey.Marshal()) {
+		return buserr.New("ErrSSHKeyMismatch")
+	}
+	req.PublicKey = strings.TrimSpace(string(ssh.MarshalAuthorizedKey(publicKey)))
+	if comment != "" {
+		req.PublicKey += " " + comment
+	}
+	req.PublicKey += "\n"
+	req.PrivateKey += "\n"
+	return nil
+}
+
 func updateLocalConn(newPort uint) error {
 	conn, _ := settingRepo.GetValueByKey("LocalSSHConn")
 	if len(conn) == 0 {

agent/app/task/task.go

@@ -154,7 +154,7 @@ func NewTask(name, operate, taskScope, taskID string, resourceID uint) (*Task, e
 	logPath := path.Join(global.Dir.TaskDir, taskScope, taskID+".log")
 	logger := logrus.New()
 	logger.SetFormatter(&SimpleFormatter{})
-	logFile, err := os.OpenFile(logPath, os.O_TRUNC|os.O_CREATE|os.O_WRONLY, constant.FilePerm)
+	logFile, err := os.OpenFile(logPath, os.O_TRUNC|os.O_CREATE|os.O_WRONLY|os.O_APPEND, constant.FilePerm)
 	if err != nil {
 		return nil, fmt.Errorf("failed to open log file: %w", err)
 	}

agent/cmd/server/docs/x-log.json

@@ -644,6 +644,13 @@
 		"formatZH": "创建容器网络 name",
 		"formatEN": "create container network [name]"
 	},
+	"/containers/network/clean": {
+		"bodyKeys": [],
+		"paramKeys": [],
+		"beforeFunctions": [],
+		"formatZH": "清理未使用的容器网络",
+		"formatEN": "Clean unused container networks"
+	},
 	"/containers/network/del": {
 		"bodyKeys": [
 			"names"
@@ -2486,6 +2493,17 @@
 		"formatZH": "删除虚拟机 [name]",
 		"formatEN": "delete VM [name]"
 	},
+	"/core/xpack/vms/disks/resize": {
+		"bodyKeys": [
+			"name",
+			"target",
+			"sizeGiB"
+		],
+		"paramKeys": [],
+		"beforeFunctions": [],
+		"formatZH": "扩容虚拟机 [name] 磁盘 [target] 至 [sizeGiB] GiB",
+		"formatEN": "expand VM [name] disk [target] to [sizeGiB] GiB"
+	},
 	"/core/xpack/vms/environment/enable": {
 		"bodyKeys": [],
 		"paramKeys": [],
@@ -4028,12 +4046,21 @@
 	},
 	"/runtimes/operate": {
 		"bodyKeys": [
-			"id"
+			"ID"
 		],
 		"paramKeys": [],
-		"beforeFunctions": [],
-		"formatZH": "操作运行环境 [id]",
-		"formatEN": "Operate runtime [id]"
+		"beforeFunctions": [
+			{
+				"input_column": "id",
+				"input_value": "ID",
+				"isList": false,
+				"db": "runtimes",
+				"output_column": "name",
+				"output_value": "name"
+			}
+		],
+		"formatZH": "操作运行环境 [name]",
+		"formatEN": "Operate runtime [name]"
 	},
 	"/runtimes/php/config": {
 		"bodyKeys": [

agent/cron/cron.go

@@ -38,6 +38,28 @@ func Run() {
 		}
 	}
 
+	var gpuStatus, gpuInterval model.Setting
+	if err := global.DB.Where("key = ?", "GPUMonitorStatus").First(&gpuStatus).Error; err != nil {
+		global.LOG.Errorf("Load GPU monitoring status failed: %v", err)
+	} else if gpuStatus.Value == constant.StatusEnable {
+		if err := global.DB.Where("key = ?", "GPUMonitorInterval").First(&gpuInterval).Error; err != nil {
+			global.LOG.Errorf("Load GPU monitoring interval failed: %v", err)
+		} else if err := service.StartGPUMonitor(gpuInterval.Value); err != nil {
+			global.LOG.Errorf("Start GPU monitoring failed: %v", err)
+		}
+	}
+
+	var vllmStatus, vllmInterval model.Setting
+	if err := global.DB.Where("key = ?", "VLLMMonitorStatus").First(&vllmStatus).Error; err != nil {
+		global.LOG.Errorf("Load vLLM monitoring status failed: %v", err)
+	} else if vllmStatus.Value == constant.StatusEnable {
+		if err := global.DB.Where("key = ?", "VLLMMonitorInterval").First(&vllmInterval).Error; err != nil {
+			global.LOG.Errorf("Load vLLM monitoring interval failed: %v", err)
+		} else if err := service.StartVLLMMonitor(vllmInterval.Value); err != nil {
+			global.LOG.Errorf("Start vLLM monitoring failed: %v", err)
+		}
+	}
+
 	if _, err := global.Cron.AddJob("@daily", job.NewWebsiteJob()); err != nil {
 		global.LOG.Errorf("can not add  website corn job: %s", err.Error())
 	}

agent/global/global.go

@@ -14,12 +14,13 @@ import (
 )
 
 var (
-	DB           *gorm.DB
-	MonitorDB    *gorm.DB
-	GPUMonitorDB *gorm.DB
-	TaskDB       *gorm.DB
-	CoreDB       *gorm.DB
-	AlertDB      *gorm.DB
+	DB            *gorm.DB
+	MonitorDB     *gorm.DB
+	GPUMonitorDB  *gorm.DB
+	VLLMMonitorDB *gorm.DB
+	TaskDB        *gorm.DB
+	CoreDB        *gorm.DB
+	AlertDB       *gorm.DB
 
 	LOG   *logrus.Logger
 	CONF  ServerConfig
@@ -29,8 +30,10 @@ var (
 
 	Dir SystemDir
 
-	Cron          *cron.Cron
-	MonitorCronID cron.EntryID
+	Cron              *cron.Cron
+	MonitorCronID     cron.EntryID
+	VLLMMonitorCronID cron.EntryID
+	GPUMonitorCronID  cron.EntryID
 
 	IsMaster bool
 

agent/go.mod

@@ -1,14 +1,12 @@
 module github.com/1Panel-dev/1Panel/agent
 
-go 1.26.1
-
-replace github.com/moby/go-archive => github.com/moby/go-archive v0.1.0
+go 1.26.6
 
 replace github.com/go-acme/lego/v5 => github.com/1Panel-dev/lego/v5 v5.3.1
 
 require (
 	github.com/aliyun/aliyun-oss-go-sdk v3.0.2+incompatible
-	github.com/compose-spec/compose-go/v2 v2.14.0
+	github.com/compose-spec/compose-go/v2 v2.15.0
 	github.com/creack/pty v1.1.24
 	github.com/docker/cli v29.7.2+incompatible
 	github.com/docker/docker v28.5.2+incompatible
@@ -33,6 +31,7 @@ require (
 	github.com/mholt/archiver/v4 v4.0.0-alpha.8
 	github.com/miekg/dns v1.1.73
 	github.com/minio/minio-go/v7 v7.3.0
+	github.com/moby/go-archive v0.3.0
 	github.com/nicksnyder/go-i18n/v2 v2.6.1
 	github.com/opencontainers/image-spec v1.1.1
 	github.com/oschwald/maxminddb-golang v1.13.1
@@ -52,13 +51,13 @@ require (
 	github.com/tklauser/go-sysconf v0.4.0
 	github.com/tomasen/fcgi_client v0.0.0-20180423082037-2bb3d819fd19
 	github.com/upyun/go-sdk v2.1.0+incompatible
-	go.mongodb.org/mongo-driver/v2 v2.8.2
-	golang.org/x/crypto v0.55.0
+	go.mongodb.org/mongo-driver/v2 v2.9.0
+	golang.org/x/crypto v0.56.0
 	golang.org/x/net v0.58.0
-	golang.org/x/sync v0.22.0
+	golang.org/x/sync v0.23.0
 	golang.org/x/sys v0.47.0
 	golang.org/x/text v0.41.0
-	golang.org/x/time v0.15.0
+	golang.org/x/time v0.16.0
 	google.golang.org/genproto v0.0.0-20260414002931-afd174a4e478
 	gopkg.in/ini.v1 v1.67.3
 	gopkg.in/yaml.v3 v3.0.1
@@ -121,7 +120,7 @@ require (
 	github.com/go-acme/alidns-20150109/v5 v5.5.0 // indirect
 	github.com/go-acme/esa-20240910/v3 v3.4.0 // indirect
 	github.com/go-jose/go-jose/v4 v4.1.4 // indirect
-	github.com/go-logr/logr v1.4.3 // indirect
+	github.com/go-logr/logr v1.4.4 // indirect
 	github.com/go-logr/stdr v1.2.2 // indirect
 	github.com/go-ole/go-ole v1.3.0 // indirect
 	github.com/go-playground/locales v0.14.1 // indirect
@@ -132,7 +131,6 @@ require (
 	github.com/gofrs/flock v0.13.0 // indirect
 	github.com/golang/snappy v1.0.0 // indirect
 	github.com/google/go-querystring v1.2.0 // indirect
-	github.com/grpc-ecosystem/grpc-gateway/v2 v2.29.0 // indirect
 	github.com/hashicorp/golang-lru/v2 v2.0.7 // indirect
 	github.com/huaweicloud/huaweicloud-sdk-go-v3 v0.1.205 // indirect
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
@@ -155,11 +153,10 @@ require (
 	github.com/minio/md5-simd v1.1.2 // indirect
 	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/moby/docker-image-spec v1.3.1 // indirect
-	github.com/moby/go-archive v0.2.0 // indirect
 	github.com/moby/patternmatcher v0.6.1 // indirect
 	github.com/moby/sys/atomicwriter v0.1.0 // indirect
-	github.com/moby/sys/sequential v0.6.0 // indirect
-	github.com/moby/sys/user v0.4.0 // indirect
+	github.com/moby/sys/sequential v0.7.0 // indirect
+	github.com/moby/sys/user v0.4.1 // indirect
 	github.com/moby/sys/userns v0.1.0 // indirect
 	github.com/moby/term v0.5.2 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
@@ -209,18 +206,18 @@ require (
 	go.mongodb.org/mongo-driver v1.17.9 // indirect
 	go.opentelemetry.io/auto/sdk v1.2.1 // indirect
 	go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.68.0 // indirect
-	go.opentelemetry.io/otel v1.44.0 // indirect
-	go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.43.0 // indirect
-	go.opentelemetry.io/otel/metric v1.44.0 // indirect
-	go.opentelemetry.io/otel/sdk v1.44.0 // indirect
-	go.opentelemetry.io/otel/sdk/metric v1.44.0 // indirect
-	go.opentelemetry.io/otel/trace v1.44.0 // indirect
+	go.opentelemetry.io/otel v1.45.0 // indirect
+	go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.45.0 // indirect
+	go.opentelemetry.io/otel/metric v1.45.0 // indirect
+	go.opentelemetry.io/otel/sdk v1.45.0 // indirect
+	go.opentelemetry.io/otel/sdk/metric v1.45.0 // indirect
+	go.opentelemetry.io/otel/trace v1.45.0 // indirect
 	go.yaml.in/yaml/v3 v3.0.5 // indirect
 	go.yaml.in/yaml/v4 v4.0.0-rc.4 // indirect
 	go4.org v0.0.0-20260112195520-a5071408f32f // indirect
 	golang.org/x/arch v0.26.0 // indirect
 	golang.org/x/oauth2 v0.36.0 // indirect
-	google.golang.org/grpc v1.83.1 // indirect
+	google.golang.org/grpc v1.83.2 // indirect
 	google.golang.org/protobuf v1.36.11 // indirect
 	modernc.org/fileutil v1.4.0 // indirect
 	modernc.org/libc v1.72.0 // indirect

agent/go.sum

@@ -205,8 +205,8 @@ github.com/cloudwego/base64x v0.1.6/go.mod h1:OFcloc187FXDaYHvrNIjxSe8ncn0OOM8gE
 github.com/cncf/udpa/go v0.0.0-20191209042840-269d4d468f6f/go.mod h1:M8M6+tZqaGXZJjfX53e64911xZQV5JYwmTeXPW+k8Sc=
 github.com/cncf/udpa/go v0.0.0-20201120205902-5459f2c99403/go.mod h1:WmhPx2Nbnhtbo57+VJT5O0JRkEi1Wbu0z5j0R8u5Hbk=
 github.com/cncf/xds/go v0.0.0-20210312221358-fbca930ec8ed/go.mod h1:eXthEFrGJvWHgFFCl3hGmgk+/aYT6PnTQLykKQRLhEs=
-github.com/compose-spec/compose-go/v2 v2.14.0 h1:uaJeo5B3+OVlu+Rx2qLBcAdXPEUUzm5nQrRiGJafRAQ=
-github.com/compose-spec/compose-go/v2 v2.14.0/go.mod h1:ZU6zlcweCZKyiB7BVfCizQT9XmkEIMFE+PRZydVcsZg=
+github.com/compose-spec/compose-go/v2 v2.15.0 h1:tdQw+eMyT+P6ZIb09JfcIVvbMmIa+PjST7cWezVLf00=
+github.com/compose-spec/compose-go/v2 v2.15.0/go.mod h1:Q1+qtN4vhzEjGrnqRtzx1xa8raDZQlMUe3WJxndYNiQ=
 github.com/containerd/errdefs v1.0.0 h1:tg5yIfIlQIrxYtu9ajqY42W3lpS19XqdxRQeEwYG8PI=
 github.com/containerd/errdefs v1.0.0/go.mod h1:+YBYIdtsnF4Iw6nWZhJcqGSg/dwvV7tyJ/kCkyJ2k+M=
 github.com/containerd/errdefs/pkg v0.3.0 h1:9IKJ06FvyNlexW690DXuQNx2KA2cUJXx151Xdx3ZPPE=
@@ -307,8 +307,8 @@ github.com/go-logfmt/logfmt v0.4.0/go.mod h1:3RMwSq7FuexP4Kalkev3ejPJsZTpXXBr9+V
 github.com/go-logfmt/logfmt v0.5.0/go.mod h1:wCYkCAKZfumFQihp8CzCvQ3paCTfi41vtzG1KdI/P7A=
 github.com/go-logfmt/logfmt v0.5.1/go.mod h1:WYhtIu8zTZfxdn5+rREduYbwxfcBr/Vr6KEVveWlfTs=
 github.com/go-logr/logr v1.2.2/go.mod h1:jdQByPbusPIv2/zmleS9BjJVeZ6kBagPoEUsqbVz/1A=
-github.com/go-logr/logr v1.4.3 h1:CjnDlHq8ikf6E492q6eKboGOC0T8CDaOvkHCIg8idEI=
-github.com/go-logr/logr v1.4.3/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
+github.com/go-logr/logr v1.4.4 h1:tG4xh9yMsRCAiodLVTxyrkzSZ9+o0L1Kg/+cPVcbP/8=
+github.com/go-logr/logr v1.4.4/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
 github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
 github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
 github.com/go-ole/go-ole v1.2.6/go.mod h1:pprOEPIfldk/42T2oK7lQ4v4JSDwmV0As9GaiUsvbm0=
@@ -438,6 +438,7 @@ github.com/gorilla/websocket v1.5.3/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/ad
 github.com/gorilla/websocket v1.5.4-0.20250319132907-e064f32e3674 h1:JeSE6pjso5THxAzdVpqr6/geYxZytqFMBCOtn/ujyeo=
 github.com/gorilla/websocket v1.5.4-0.20250319132907-e064f32e3674/go.mod h1:r4w70xmWCQKmi1ONH4KIaBptdivuRPyosB9RmPlGEwA=
 github.com/grpc-ecosystem/go-grpc-prometheus v1.2.0/go.mod h1:8NvIoxWQoOIhqOTXgfV/d3M/q6VIi02HzZEHgUlZvzk=
+github.com/grpc-ecosystem/grpc-gateway v1.16.0 h1:gmcG1KaJ57LophUzW0Hy8NmPhnMZb4M0+kPpLofRdBo=
 github.com/grpc-ecosystem/grpc-gateway v1.16.0/go.mod h1:BDjrQk3hbvj6Nolgz8mAMFbcEtjT1g+wF4CSlocrBnw=
 github.com/grpc-ecosystem/grpc-gateway/v2 v2.29.0 h1:5VipnvEpbqr2gA2VbM+nYVbkIF28c5ZQfqCBQ5g2xfk=
 github.com/grpc-ecosystem/grpc-gateway/v2 v2.29.0/go.mod h1:Hyl3n6Twe1hvtd9XUXDec4pTvgMSEixRuQKPTMH2bNs=
@@ -597,16 +598,16 @@ github.com/mitchellh/mapstructure v1.5.0 h1:jeMsZIYE/09sWLaz43PL7Gy6RuMjD2eJVyua
 github.com/mitchellh/mapstructure v1.5.0/go.mod h1:bFUtVrKA4DC2yAKiSyO/QUcy7e+RRV2QTWOzhPopBRo=
 github.com/moby/docker-image-spec v1.3.1 h1:jMKff3w6PgbfSa69GfNg+zN/XLhfXJGnEx3Nl2EsFP0=
 github.com/moby/docker-image-spec v1.3.1/go.mod h1:eKmb5VW8vQEh/BAr2yvVNvuiJuY6UIocYsFu/DxxRpo=
-github.com/moby/go-archive v0.1.0 h1:Kk/5rdW/g+H8NHdJW2gsXyZ7UnzvJNOy6VKJqueWdcQ=
-github.com/moby/go-archive v0.1.0/go.mod h1:G9B+YoujNohJmrIYFBpSd54GTUB4lt9S+xVQvsJyFuo=
+github.com/moby/go-archive v0.3.0 h1:nos4BtzzUIqB406BgQnWGMI4qib9BZ8XUHU+ucv/n1c=
+github.com/moby/go-archive v0.3.0/go.mod h1:Npdv43fFqlhZW7Xo8fbm3ZMYFvAGNviUPqX21VERbcE=
 github.com/moby/patternmatcher v0.6.1 h1:qlhtafmr6kgMIJjKJMDmMWq7WLkKIo23hsrpR3x084U=
 github.com/moby/patternmatcher v0.6.1/go.mod h1:hDPoyOpDY7OrrMDLaYoY3hf52gNCR/YOUYxkhApJIxc=
 github.com/moby/sys/atomicwriter v0.1.0 h1:kw5D/EqkBwsBFi0ss9v1VG3wIkVhzGvLklJ+w3A14Sw=
 github.com/moby/sys/atomicwriter v0.1.0/go.mod h1:Ul8oqv2ZMNHOceF643P6FKPXeCmYtlQMvpizfsSoaWs=
-github.com/moby/sys/sequential v0.6.0 h1:qrx7XFUd/5DxtqcoH1h438hF5TmOvzC/lspjy7zgvCU=
-github.com/moby/sys/sequential v0.6.0/go.mod h1:uyv8EUTrca5PnDsdMGXhZe6CCe8U/UiTWd+lL+7b/Ko=
-github.com/moby/sys/user v0.4.0 h1:jhcMKit7SA80hivmFJcbB1vqmw//wU61Zdui2eQXuMs=
-github.com/moby/sys/user v0.4.0/go.mod h1:bG+tYYYJgaMtRKgEmuueC0hJEAZWwtIbZTB+85uoHjs=
+github.com/moby/sys/sequential v0.7.0 h1:ASQNGNROJSuOO6LL6bPHbKvuZu6NU8P4ldPWk31zj/8=
+github.com/moby/sys/sequential v0.7.0/go.mod h1:NfSTAp6V3fw4tmkD62PEcOKeZKquXT8VKCkf7aVR79o=
+github.com/moby/sys/user v0.4.1 h1:RgjRlaDKi/Xmyrz4t8lyzXT6v2ooFeO/7xtchmhVWE0=
+github.com/moby/sys/user v0.4.1/go.mod h1:E9QsW5WRe1kUAf7kW8hXKwu1uhsZEAdPLYHYSDudF4Y=
 github.com/moby/sys/userns v0.1.0 h1:tVLXkFOxVu9A64/yh59slHVv9ahO9UIev4JZusOLG/g=
 github.com/moby/sys/userns v0.1.0/go.mod h1:IHUYgu/kao6N8YZlp9Cf444ySSvCmDlmzUcYfDHOl28=
 github.com/moby/term v0.5.2 h1:6qk3FJAFDs6i/q3W/pQ97SX192qKfZgGjCQqfCJkgzQ=
@@ -877,8 +878,8 @@ go.etcd.io/etcd/client/v3 v3.5.0/go.mod h1:AIKXXVX/DQXtfTEqBryiLTUXwON+GuvO6Z7lL
 go.mongodb.org/mongo-driver v1.13.1/go.mod h1:wcDf1JBCXy2mOW0bWHwO/IOYqdca1MPCwDtFu/Z9+eo=
 go.mongodb.org/mongo-driver v1.17.9 h1:IexDdCuuNJ3BHrELgBlyaH9p60JXAvdzWR128q+U5tU=
 go.mongodb.org/mongo-driver v1.17.9/go.mod h1:LlOhpH5NUEfhxcAwG0UEkMqwYcc4JU18gtCdGudk/tQ=
-go.mongodb.org/mongo-driver/v2 v2.8.2 h1:b6o2m7zL8g2URuO8urBedAylxojybKXNZTxgkOcl+2w=
-go.mongodb.org/mongo-driver/v2 v2.8.2/go.mod h1:yOI9kBsufol30iFsl1slpdq1I0eHPzybRWdyYUs8K/0=
+go.mongodb.org/mongo-driver/v2 v2.9.0 h1:e2mQdOmbkiYz+dj3faM7lVDwl7WdnRD+g5VicafMhL0=
+go.mongodb.org/mongo-driver/v2 v2.9.0/go.mod h1:SHKN0IWkKmEVGHLjXnni6s4wPKX4v86FTgOeJJFuXcA=
 go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
 go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
 go.opencensus.io v0.22.2/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=
@@ -889,23 +890,23 @@ go.opentelemetry.io/auto/sdk v1.2.1 h1:jXsnJ4Lmnqd11kwkBV2LgLoFMZKizbCi5fNZ/ipaZ
 go.opentelemetry.io/auto/sdk v1.2.1/go.mod h1:KRTj+aOaElaLi+wW1kO/DZRXwkF4C5xPbEe3ZiIhN7Y=
 go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.68.0 h1:CqXxU8VOmDefoh0+ztfGaymYbhdB/tT3zs79QaZTNGY=
 go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.68.0/go.mod h1:BuhAPThV8PBHBvg8ZzZ/Ok3idOdhWIodywz2xEcRbJo=
-go.opentelemetry.io/otel v1.44.0 h1:JjwHmHpA4iZ3wBxluu2fbbE7j4kqlE8jXyAyPXH7HqU=
-go.opentelemetry.io/otel v1.44.0/go.mod h1:BMgjTHL9WPRlRjL2oZCBTL4whCGtXch2H4BhOPIAyYc=
-go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.43.0 h1:88Y4s2C8oTui1LGM6bTWkw0ICGcOLCAI5l6zsD1j20k=
-go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.43.0/go.mod h1:Vl1/iaggsuRlrHf/hfPJPvVag77kKyvrLeD10kpMl+A=
-go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.43.0 h1:3iZJKlCZufyRzPzlQhUIWVmfltrXuGyfjREgGP3UUjc=
-go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.43.0/go.mod h1:/G+nUPfhq2e+qiXMGxMwumDrP5jtzU+mWN7/sjT2rak=
-go.opentelemetry.io/otel/metric v1.44.0 h1:1w0gILTcHdr3YI+ixLyjemwrVnsMURbTZFrSYCdDdmc=
-go.opentelemetry.io/otel/metric v1.44.0/go.mod h1:8O7hanEPBNgEMmybD3s2VBKcgWOCsA6tzHBPODAiquo=
-go.opentelemetry.io/otel/sdk v1.44.0 h1:nHYwb9lK+fJPU/dnT6s7W7Z8itMWyqrnVfbheVYrZ58=
-go.opentelemetry.io/otel/sdk v1.44.0/go.mod h1:Osuydd3Se74nqjAKxid74N5eC+jfEqfTegHRnq58oK0=
-go.opentelemetry.io/otel/sdk/metric v1.44.0 h1:3LlKgI+VjbVsjNRFZJZAJ30WjXC5VkNRks6si09iEfI=
-go.opentelemetry.io/otel/sdk/metric v1.44.0/go.mod h1:5B5pMARnXxKhltooO4xUuCBorl65a4EpnTalObqOigA=
-go.opentelemetry.io/otel/trace v1.44.0 h1:jxF5CsGYCe74MCRx2X4g7WsY/VBKRqqpNvXlX/6gtIk=
-go.opentelemetry.io/otel/trace v1.44.0/go.mod h1:oLl1jrMQAVo6v3GAggN+1VH9VIz9iUSvW53sW1Q8PIE=
+go.opentelemetry.io/otel v1.45.0 h1:pdrWmLHofpubmArBv1LgFSv1Z0Ie/ppdZzu+kUN5EeU=
+go.opentelemetry.io/otel v1.45.0/go.mod h1:XZxIqPapzEYnhNSScF5DIqXhm/rYi0FzCe2XddAwZfQ=
+go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.45.0 h1:QRefszxJmfPdjXUUm3j6iDzY03mTPXMjqErFqQ67vUg=
+go.opentelemetry.io/otel/exporters/otlp/otlptrace v1.45.0/go.mod h1:Tiz03lTBVBrm7eWZBOidzEaYaJa8tjwGUGv6d8mlTyk=
+go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.45.0 h1:QBajQ2SrwQijzHyZbQlPsuIzpl/ll8DY6wPWsajeGcI=
+go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp v1.45.0/go.mod h1:08ZQLjrPLQ6R4kAXvuOvODEer5Yh4CoFvll5qB2BCI8=
+go.opentelemetry.io/otel/metric v1.45.0 h1:7Eg1uH7CJ5cXv9is6tnBe1FI6rj1nwUdbFypRm3br/M=
+go.opentelemetry.io/otel/metric v1.45.0/go.mod h1:HAPbm1nd3p1PmFH7v2dR+6BjXxw+Lq4a2+pndMAm08s=
+go.opentelemetry.io/otel/sdk v1.45.0 h1:4VVSMgQ83dUgW2aoX5f6JgLvHwIvzcuLnF9lUdCSpCw=
+go.opentelemetry.io/otel/sdk v1.45.0/go.mod h1:Sr40LgXV7DsKMMJMKOhUWOgMWTfAaqvm2kF0g7ilwuA=
+go.opentelemetry.io/otel/sdk/metric v1.45.0 h1:oVFszMfyj1Am6s24Vtc7wBb8BKLcwepJjNEYILuiE3o=
+go.opentelemetry.io/otel/sdk/metric v1.45.0/go.mod h1:vUWUxDZvu1WVRj8JA8S0AdhsPrZoDpA2DdZauIh4mDA=
+go.opentelemetry.io/otel/trace v1.45.0 h1:l/mP6Uv7oNO7/TblbhpbgMidxhq1uO/rPsikOyVhxag=
+go.opentelemetry.io/otel/trace v1.45.0/go.mod h1:qoJJA2xNMnxRrdISU/kLtfUH2wNeQbiv+jhs/CxI8bc=
 go.opentelemetry.io/proto/otlp v0.7.0/go.mod h1:PqfVotwruBrMGOCsRd/89rSnXhoiJIqeYNgFYFoEGnI=
-go.opentelemetry.io/proto/otlp v1.10.0 h1:IQRWgT5srOCYfiWnpqUYz9CVmbO8bFmKcwYxpuCSL2g=
-go.opentelemetry.io/proto/otlp v1.10.0/go.mod h1:/CV4QoCR/S9yaPj8utp3lvQPoqMtxXdzn7ozvvozVqk=
+go.opentelemetry.io/proto/otlp v1.11.0 h1:5rrYs0Ykyj50sdU/JU0x8etU+LubXWb+gED6TbEdMIk=
+go.opentelemetry.io/proto/otlp v1.11.0/go.mod h1:SmVizdCOAm3XBtG1g1NnOdhW6jtddT72hLMhv8VwA8E=
 go.uber.org/atomic v1.7.0/go.mod h1:fEN4uk6kAWBTFdckzkM89CLk9XfWZrxpCo0nPH17wJc=
 go.uber.org/atomic v1.9.0/go.mod h1:fEN4uk6kAWBTFdckzkM89CLk9XfWZrxpCo0nPH17wJc=
 go.uber.org/goleak v1.1.11-0.20210813005559-691160354723/go.mod h1:cwTWslyiVhfpKIDGSZEM2HlOvcqm+tG4zioyIeLoqMQ=
@@ -951,8 +952,8 @@ golang.org/x/crypto v0.19.0/go.mod h1:Iy9bg/ha4yyC70EfRS8jz+B6ybOBKMaSxLj6P6oBDf
 golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
 golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8=
 golang.org/x/crypto v0.24.0/go.mod h1:Z1PMYSOR5nyMcyAVAIQSKCDwalqy85Aqn1x3Ws4L5DM=
-golang.org/x/crypto v0.55.0 h1:+KWHjbgOaAQ66dh/YlkZKHlz9ZUlq61AFirAR9ntP8M=
-golang.org/x/crypto v0.55.0/go.mod h1:uq0V9dE/fzQuJtbnL+2EhWOE63vo164FY8xqEnV9xis=
+golang.org/x/crypto v0.56.0 h1:GUh5Ii4J5jtcseSMiRqr1jXCNHoxjeV9Fmekc2oLy6Y=
+golang.org/x/crypto v0.56.0/go.mod h1:OMW5y6CY9l38uPLmxU6l6pwcXp1obtLo3e6gT7gQR2I=
 golang.org/x/exp v0.0.0-20180321215751-8460e604b9de/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20180807140117-3d87b88a115f/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
 golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
@@ -1076,8 +1077,8 @@ golang.org/x/sync v0.1.0/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sync v0.3.0/go.mod h1:FU7BRWz2tNW+3quACPkgCx/L+uEAv1htQ0V83Z9Rj+Y=
 golang.org/x/sync v0.6.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
 golang.org/x/sync v0.7.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
-golang.org/x/sync v0.22.0 h1:SZjpbeLmrCk4xhRSZFNZW5gFUeCeFgjekvI/+gfScek=
-golang.org/x/sync v0.22.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
+golang.org/x/sync v0.23.0 h1:KameEIfc1IkluZyXWLn39Wd4tURc6GbCiISGiZm2bQk=
+golang.org/x/sync v0.23.0/go.mod h1:sUUOizhqBxiL6pEWpqNLUiaJn1ShEbZ6BBqskPbjZm0=
 golang.org/x/sys v0.0.0-20180823144017-11551d06cbcc/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20180830151530-49385e6e1522/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20180905080454-ebe1bf3edb33/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
@@ -1193,8 +1194,8 @@ golang.org/x/time v0.0.0-20190308202827-9d24e82272b4/go.mod h1:tRJNPiyCQ0inRvYxb
 golang.org/x/time v0.0.0-20191024005414-555d28b269f0/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
 golang.org/x/time v0.0.0-20200416051211-89c76fbcd5d1/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
 golang.org/x/time v0.0.0-20210723032227-1f47c861a9ac/go.mod h1:tRJNPiyCQ0inRvYxbN9jk5I+vvW/OXSQhTDSoE431IQ=
-golang.org/x/time v0.15.0 h1:bbrp8t3bGUeFOx08pvsMYRTCVSMk89u4tKbNOZbp88U=
-golang.org/x/time v0.15.0/go.mod h1:Y4YMaQmXwGQZoFaVFk4YpCt4FLQMYKZe9oeV/f4MSno=
+golang.org/x/time v0.16.0 h1:vMb6ptszcQMkcwiRTAuNNU50gom6++Q/6gY2hDM6VDE=
+golang.org/x/time v0.16.0/go.mod h1:rVKOqvZeKvrDKTQiAHJ7wmwP0RzleSphoEA9RcdLA0s=
 golang.org/x/tools v0.0.0-20180525024113-a5b4c53f6e8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20190114222345-bf090417da8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
@@ -1315,10 +1316,10 @@ google.golang.org/genproto v0.0.0-20210602131652-f16073e35f0c/go.mod h1:UODoCrxH
 google.golang.org/genproto v0.0.0-20210917145530-b395a37504d4/go.mod h1:eFjDcFEctNawg4eG61bRv87N7iHBWyVhJu7u1kqDUXY=
 google.golang.org/genproto v0.0.0-20260414002931-afd174a4e478 h1:aLsVTW0lZ8+IY5u/ERjZSCvAmhuR7slKzyha3YikDNA=
 google.golang.org/genproto v0.0.0-20260414002931-afd174a4e478/go.mod h1:YJAzKjfHIUHb9T+bfu8L7mthAp7VVXQBUs1PLdBWS7M=
-google.golang.org/genproto/googleapis/api v0.0.0-20260526163538-3dc84a4a5aaa h1:Kjn0N0tCrDgiAFW+lGO4JZ3ck44CehvJQMAwj9QF0G8=
-google.golang.org/genproto/googleapis/api v0.0.0-20260526163538-3dc84a4a5aaa/go.mod h1:q4lMZS6kskjT5HvCPrnnypcDPVJqT/f4nfxmkE7gryY=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20260630182238-925bb5da69e7 h1:eM/YSd5bBFagF51o1E745Ta7RwzpW0h+z+QDNZOgmQ8=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20260630182238-925bb5da69e7/go.mod h1:4Hqkh8ycfw05ld/3BWL7rJOSfebL2Q+DVDeRgYgxUU8=
+google.golang.org/genproto/googleapis/api v0.0.0-20260803160001-6ac0973c030d h1:FarXi840EJWSHYTN3ERkADbPWjl307+FGrA22KAVjjc=
+google.golang.org/genproto/googleapis/api v0.0.0-20260803160001-6ac0973c030d/go.mod h1:K/+WGbmBY7aNW1HDw1fJnKYo10i0DkAX6pows00dLig=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20260803160001-6ac0973c030d h1:IL4hdHzcUv2l/gcg98/Rj3FbtE6axwqslOW8SW0C+S0=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20260803160001-6ac0973c030d/go.mod h1:4Hqkh8ycfw05ld/3BWL7rJOSfebL2Q+DVDeRgYgxUU8=
 google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
 google.golang.org/grpc v1.20.1/go.mod h1:10oTOabMzJvdu6/UiuZezV6QK5dSlG84ov/aaiqXj38=
 google.golang.org/grpc v1.21.1/go.mod h1:oYelfM1adQP15Ek0mdvEgi9Df8B9CZIaU1084ijfRaM=
@@ -1336,8 +1337,8 @@ google.golang.org/grpc v1.33.2/go.mod h1:JMHMWHQWaTccqQQlmk3MJZS+GWXOdAesneDmEnv
 google.golang.org/grpc v1.36.0/go.mod h1:qjiiYl8FncCW8feJPdyg3v6XW24KsRHe+dy9BAGRRjU=
 google.golang.org/grpc v1.38.0/go.mod h1:NREThFqKR1f3iQ6oBuvc5LadQuXVGo9rkm5ZGrQdJfM=
 google.golang.org/grpc v1.40.0/go.mod h1:ogyxbiOoUXAkP+4+xa6PZSE9DZgIHtSpzjDTB9KAK34=
-google.golang.org/grpc v1.83.1 h1:HIO0+BEtBP6soyqvqC8sNUjZ7bTs+0hFQuFF+RAy++Y=
-google.golang.org/grpc v1.83.1/go.mod h1:kDyl6SKsiHKt0uylY5gtn5cEjkrIOhQOGDgIc4JGwzQ=
+google.golang.org/grpc v1.83.2 h1:EManeRomTObA0BU7I8vXgg/78uE5MJ9M8B39EX2WscU=
+google.golang.org/grpc v1.83.2/go.mod h1:YPI1hK3kDked6iHvgX3tR0y+nX/qpMFKhPgFsokw1S8=
 google.golang.org/protobuf v0.0.0-20200109180630-ec00e32a8dfd/go.mod h1:DFci5gLYBciE7Vtevhsrf46CRTquxDuWsQurQQe4oz8=
 google.golang.org/protobuf v0.0.0-20200221191635-4d8936d0db64/go.mod h1:kwYJMbMJ01Woi6D6+Kah6886xMZcty6N08ah7+eCXa0=
 google.golang.org/protobuf v0.0.0-20200228230310-ab0ca4ff8a60/go.mod h1:cfTl7dwQJ+fmap5saPgwCLgHXTUD7jkjRqWcaiX5VyM=

agent/i18n/lang/en.yaml

@@ -649,6 +649,7 @@ CommonAlert: "Panel {{ .node }}{{ .ip }}: {{ .msg }}. Log in to view details."
 NodeExceptionAlert: "Panel {{ .node }}{{ .ip }}: {{ .num }} nodes are abnormal. Log in to view details."
 LicenseExceptionAlert: "Panel {{ .node }}{{ .ip }}: {{ .num }} licenses are abnormal. Log in to view details."
 SSHAndPanelLoginAlert: "Panel {{ .node }}{{ .ip }}: abnormal {{ .name }} login from {{ .loginIp }}. Log in to view details."
+CronJobSuccessAlert: "Panel {{ .node }}{{ .ip }}: scheduled task {{ .name }} completed successfully. Log in to view details."
 
 # disk
 DeviceNotFound: "Device {{ .name }} not found"
@@ -730,3 +731,17 @@ FileTaskRename: 'Target filename: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "The current backend {{ .current }} still contains 1Panel rules. Clean it up before switching to {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "IPv4 forwarding is disabled. Set net.ipv4.ip_forward=1 before using Docker's firewall backend."
 ErrFirewallRuleSavedApplyFailed: "The new rule configuration was saved, but applying it to the firewall failed. Retry by synchronizing: {{ .detail }}"
+
+ErrSSHPublicKey: "Enter a valid SSH public key"
+ErrSSHPrivateKey: "Invalid SSH private key or incorrect passphrase; encrypted keys require the original passphrase"
+ErrSSHKeyMismatch: "The SSH public and private keys do not match"
+
+NetworkCleanupDeleted: "Deleted network [{{ .name }}] ({{ .id }})"
+NetworkCleanupProtected: "Skipped network [{{ .name }}] ({{ .id }}): reserved network"
+NetworkCleanupConnected: "Skipped network [{{ .name }}] ({{ .id }}): containers connected or network in use; not deleted"
+NetworkCleanupUnsupported: "Skipped network [{{ .name }}] ({{ .id }}): special network"
+NetworkCleanupGone: "Skipped network [{{ .name }}] ({{ .id }}): already removed"
+NetworkCleanupInspectFailed: "Failed to inspect network [{{ .name }}] ({{ .id }}); not deleted"
+NetworkCleanupRemoveFailed: "Failed to remove network [{{ .name }}] ({{ .id }})"
+NetworkCleanupSummary: "Network cleanup finished: deleted {{ .deleted }}, skipped {{ .skipped }}, failed {{ .failed }}"
+NetworkCleanupPartialFailure: "Some networks could not be cleaned; see the task log"

agent/i18n/lang/es-ES.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'Su Panel {{ .node }}{{ .ip }}, {{ .msg }}. Inicie sesión en el pa
 NodeExceptionAlert: 'Su Panel {{ .node }}{{ .ip }}, {{ .num }} nodos son anómalos. Inicie sesión en el panel para ver los detalles.'
 LicenseExceptionAlert: 'Su Panel {{ .node }}{{ .ip }}, {{ .num }} licencias son anómalas. Inicie sesión en el panel para ver los detalles.'
 SSHAndPanelLoginAlert: 'Su Panel {{ .node }}{{ .ip }}, el inicio de sesión {{ .name }} desde {{ .loginIp }} es anómalo. Inicie sesión en el panel para ver los detalles.'
+CronJobSuccessAlert: 'Panel {{ .node }}{{ .ip }}: la tarea programada {{ .name }} se completó correctamente. Inicie sesión para ver los detalles.'
 
 # disco
 DeviceNotFound: 'Dispositivo {{ .name }} no encontrado'
@@ -730,3 +731,7 @@ FileTaskRename: 'Nombre del archivo de destino: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "El backend actual {{ .current }} aún contiene reglas de 1Panel. Elimínelas antes de cambiar a {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "El reenvío IPv4 está desactivado. Configure net.ipv4.ip_forward=1 antes de usar el backend de cortafuegos de Docker."
 ErrFirewallRuleSavedApplyFailed: "Se guardó la configuración de la nueva regla, pero no se pudo aplicar al cortafuegos. Vuelva a intentarlo mediante la sincronización: {{ .detail }}"
+
+ErrSSHPublicKey: "Introduce una clave pública SSH válida"
+ErrSSHPrivateKey: "Clave privada SSH no válida o contraseña incorrecta; las claves cifradas requieren la contraseña original"
+ErrSSHKeyMismatch: "Las claves pública y privada SSH no coinciden"

agent/i18n/lang/fa.yaml

@@ -649,6 +649,8 @@ CommonAlert: "پنل {{ .node }}{{ .ip }}: {{ .msg }}. برای مشاهده ج
 NodeExceptionAlert: "پنل {{ .node }}{{ .ip }}: {{ .num }} گره غیرعادی هستند. برای مشاهده جزئیات وارد شوید."
 LicenseExceptionAlert: "پنل {{ .node }}{{ .ip }}: {{ .num }} مجوز غیرعادی است. برای مشاهده جزئیات وارد شوید."
 SSHAndPanelLoginAlert: "پنل {{ .node }}{{ .ip }}: ورود غیرعادی {{ .name }} از {{ .loginIp }}. برای مشاهده جزئیات وارد شوید."
+CronJobSuccessAlert: "پنل {{ .node }}{{ .ip }}: وظیفه زمان‌بندی‌شده {{ .name }} با موفقیت تکمیل شد. برای مشاهده جزئیات وارد شوید."
+
 
 # دیسک
 DeviceNotFound: "دستگاه {{ .name }} یافت نشد"
@@ -730,3 +732,7 @@ FileTaskRename: 'نام فایل مقصد: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "بک‌اند فعلی {{ .current }} هنوز شامل قوانین 1Panel است. پیش از تغییر به {{ .target }} آن‌ها را پاک کنید."
 ErrDockerIPv4ForwardingDisabled: "ارسال IPv4 غیرفعال است. پیش از استفاده از بک‌اند فایروال Docker، مقدار net.ipv4.ip_forward=1 را تنظیم کنید."
 ErrFirewallRuleSavedApplyFailed: "پیکربندی قانون جدید ذخیره شد، اما اعمال آن در دیوار آتش ناموفق بود. با همگام‌سازی دوباره تلاش کنید: {{ .detail }}"
+
+ErrSSHPublicKey: "یک کلید عمومی SSH معتبر وارد کنید"
+ErrSSHPrivateKey: "کلید خصوصی SSH نامعتبر یا عبارت عبور نادرست است؛ کلیدهای رمزگذاری‌شده به عبارت عبور اصلی نیاز دارند"
+ErrSSHKeyMismatch: "کلید عمومی و خصوصی SSH مطابقت ندارند"

agent/i18n/lang/ja.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'あなたの {{ .node }}{{ .ip }} パネル、{{ .msg }}。詳細
 NodeExceptionAlert: 'あなたの {{ .node }}{{ .ip }} パネル、{{ .num }} 個のノードに異常が発生しています。詳細はパネルにログインして'
 LicenseExceptionAlert: 'あなたの {{ .node }}{{ .ip }} パネル、{{ .num }} 個のライセンスに異常が発生しています。詳細はパネルにログインして'
 SSHAndPanelLoginAlert: 'あなたの {{ .node }}{{ .ip }} パネル、{{ .loginIp }} からの {{ .name }} ログインに異常があります。詳細はパネルにログインして'
+CronJobSuccessAlert: 'パネル {{ .node }}{{ .ip }}: スケジュールタスク {{ .name }} が正常に完了しました。詳細はパネルにログインして確認してください。'
 
 # ディスク
 DeviceNotFound: 'デバイス {{ .name }} が見つかりません'
@@ -730,3 +731,7 @@ FileTaskRename: '保存先ファイル名：{{ .name }}'
 ErrFirewallBackendCleanupRequired: "現在のバックエンド {{ .current }} に 1Panel ルールが残っています。{{ .target }} に切り替える前に削除してください。"
 ErrDockerIPv4ForwardingDisabled: "IPv4 転送が無効です。Docker のファイアウォールバックエンドを使用する前に net.ipv4.ip_forward=1 を設定してください。"
 ErrFirewallRuleSavedApplyFailed: "新しいルール設定は保存されましたが、ファイアウォールへの適用に失敗しました。同期で再試行してください：{{ .detail }}"
+
+ErrSSHPublicKey: "有効な SSH 公開鍵を入力してください"
+ErrSSHPrivateKey: "SSH 秘密鍵の形式が無効か、パスフレーズが間違っています。暗号化された鍵には元のパスフレーズが必要です"
+ErrSSHKeyMismatch: "SSH 公開鍵と秘密鍵が一致しません"

agent/i18n/lang/ko.yaml

@@ -649,6 +649,7 @@ CommonAlert: '귀하의 {{ .node }}{{ .ip }} 패널, {{ .msg }}。자세한 내
 NodeExceptionAlert: '귀하의 {{ .node }}{{ .ip }} 패널, {{ .num }}개의 노드에 이상이 있습니다. 자세한 내용은 패널에 로그인하십시오.'
 LicenseExceptionAlert: '귀하의 {{ .node }}{{ .ip }} 패널, {{ .num }}개의 라이센스에 이상이 있습니다. 자세한 내용은 패널에 로그인하십시오.'
 SSHAndPanelLoginAlert: '귀하의 {{ .node }}{{ .ip }} 패널, {{ .loginIp }}에서의 {{ .name }} 로그인에 이상이 있습니다. 자세한 내용은 패널에 로그인하십시오.'
+CronJobSuccessAlert: '패널 {{ .node }}{{ .ip }}: 예약 작업 {{ .name }}이(가) 성공적으로 완료되었습니다. 로그인하여 자세한 내용을 확인하십시오.'
 
 # 디스크
 DeviceNotFound: '장치 {{ .name }} 을(를) 찾을 수 없습니다'
@@ -730,3 +731,7 @@ FileTaskRename: '대상 파일 이름: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "현재 백엔드 {{ .current }}에 1Panel 규칙이 남아 있습니다. {{ .target }}로 전환하기 전에 정리하세요."
 ErrDockerIPv4ForwardingDisabled: "IPv4 전달이 비활성화되어 있습니다. Docker 방화벽 백엔드를 사용하기 전에 net.ipv4.ip_forward=1을 설정하세요."
 ErrFirewallRuleSavedApplyFailed: "새 규칙 설정이 저장되었지만 방화벽에 적용하지 못했습니다. 동기화하여 다시 시도하세요: {{ .detail }}"
+
+ErrSSHPublicKey: "유효한 SSH 공개 키를 입력하세요"
+ErrSSHPrivateKey: "SSH 개인 키 형식이 잘못되었거나 암호가 틀립니다. 암호화된 키에는 원래 암호가 필요합니다"
+ErrSSHKeyMismatch: "SSH 공개 키와 개인 키가 일치하지 않습니다"

agent/i18n/lang/lo.yaml

@@ -639,6 +639,7 @@ CommonAlert: "ແຜງຄວບຄຸມ {{ .node }}{{ .ip }}: {{ .msg }}. ເ
 NodeExceptionAlert: "ແຜງຄວບຄຸມ {{ .node }}{{ .ip }}: ໂນດ {{ .num }} ແຫ່ງຜິດປົກກະຕິ. ເຂົ້າສູ່ລະບົບເພື່ອເບິ່ງລາຍລະອຽດ."
 LicenseExceptionAlert: "ແຜງຄວບຄຸມ {{ .node }}{{ .ip }}: ໃບອະນຸຍາດ {{ .num }} ສະບັບຜິດປົກກະຕິ. ເຂົ້າສູ່ລະບົບເພື່ອເບິ່ງລາຍລະອຽດ."
 SSHAndPanelLoginAlert: "ແຜງຄວບຄຸມ {{ .node }}{{ .ip }}: ມີການເຂົ້າສູ່ລະບົບ {{ .name }} ທີ່ຜິດປົກກະຕິຈາກ {{ .loginIp }}. ເຂົ້າສູ່ລະບົບເພື່ອເບິ່ງລາຍລະອຽດ."
+CronJobSuccessAlert: "ແຜງຄວບຄຸມ {{ .node }}{{ .ip }}: ວຽກທີ່ຕັ້ງເວລາ {{ .name }} ສຳເລັດແລ້ວ. ເຂົ້າສູ່ລະບົບເພື່ອເບິ່ງລາຍລະອຽດ."
 
 #disk
 DeviceNotFound: "ບໍ່ພົບອຸປະກອນ {{ .name }}"
@@ -721,3 +722,7 @@ FileTaskRename: 'ຊື່ໄຟລ໌ປາຍທາງ: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "ແບັກເອນປັດຈຸບັນ {{ .current }} ຍັງມີກົດຂອງ 1Panel. ກະລຸນາລຶບອອກກ່ອນປ່ຽນໄປ {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "ການສົ່ງຕໍ່ IPv4 ຖືກປິດ. ກະລຸນາຕັ້ງ net.ipv4.ip_forward=1 ກ່ອນໃຊ້ແບັກເອນໄຟວໍຂອງ Docker."
 ErrFirewallRuleSavedApplyFailed: "ບັນທຶກການຕັ້ງຄ່າກົດໃໝ່ແລ້ວ ແຕ່ນຳໃຊ້ກັບໄຟວໍບໍ່ສຳເລັດ. ລອງອີກຄັ້ງດ້ວຍການຊິງຂໍ້ມູນ: {{ .detail }}"
+
+ErrSSHPublicKey: "ກະລຸນາປ້ອນຄີສາທາລະນະ SSH ທີ່ຖືກຕ້ອງ"
+ErrSSHPrivateKey: "ຄີສ່ວນຕົວ SSH ບໍ່ຖືກຕ້ອງ ຫຼື ລະຫັດຜ່ານຜິດ; ຄີທີ່ເຂົ້າລະຫັດຕ້ອງໃຊ້ລະຫັດຜ່ານເດີມ"
+ErrSSHKeyMismatch: "ຄີສາທາລະນະ ແລະ ຄີສ່ວນຕົວ SSH ບໍ່ກົງກັນ"

agent/i18n/lang/ms.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'Panel {{ .node }}{{ .ip }} Anda, {{ .msg }}. Sila log masuk ke pan
 NodeExceptionAlert: 'Panel {{ .node }}{{ .ip }} Anda, {{ .num }} nod tidak normal. Sila log masuk ke panel untuk butiran lanjut.'
 LicenseExceptionAlert: 'Panel {{ .node }}{{ .ip }} Anda, {{ .num }} lesen tidak normal. Sila log masuk ke panel untuk butiran lanjut.'
 SSHAndPanelLoginAlert: 'Panel {{ .node }}{{ .ip }} Anda, log masuk {{ .name }} dari {{ .loginIp }} tidak normal. Sila log masuk ke panel untuk butiran lanjut.'
+CronJobSuccessAlert: 'Panel {{ .node }}{{ .ip }}: tugas berjadual {{ .name }} berjaya diselesaikan. Log masuk untuk melihat butiran.'
 
 # cakera
 DeviceNotFound: 'Peranti {{ .name }} tidak ditemui'
@@ -730,3 +731,7 @@ FileTaskRename: 'Nama fail sasaran: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "Bahagian belakang semasa {{ .current }} masih mengandungi peraturan 1Panel. Buangkannya sebelum beralih kepada {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "Pemajuan IPv4 dilumpuhkan. Tetapkan net.ipv4.ip_forward=1 sebelum menggunakan bahagian belakang tembok api Docker."
 ErrFirewallRuleSavedApplyFailed: "Konfigurasi peraturan baharu telah disimpan, tetapi gagal digunakan pada tembok api. Cuba lagi melalui penyegerakan: {{ .detail }}"
+
+ErrSSHPublicKey: "Masukkan kunci awam SSH yang sah"
+ErrSSHPrivateKey: "Kunci peribadi SSH tidak sah atau frasa laluan salah; kunci yang disulitkan memerlukan frasa laluan asal"
+ErrSSHKeyMismatch: "Kunci awam dan peribadi SSH tidak sepadan"

agent/i18n/lang/pt-BR.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'Seu Painel {{ .node }}{{ .ip }}, {{ .msg }}. Faça login no painel
 NodeExceptionAlert: 'Seu Painel {{ .node }}{{ .ip }}, {{ .num }} nós estão anormais. Faça login no painel para obter detalhes.'
 LicenseExceptionAlert: 'Seu Painel {{ .node }}{{ .ip }}, {{ .num }} licenças estão anormais. Faça login no painel para obter detalhes.'
 SSHAndPanelLoginAlert: 'Seu Painel {{ .node }}{{ .ip }}, o login {{ .name }} a partir de {{ .loginIp }} é anormal. Faça login no painel para obter detalhes.'
+CronJobSuccessAlert: 'Painel {{ .node }}{{ .ip }}: a tarefa agendada {{ .name }} foi concluída com sucesso. Faça login para ver os detalhes.'
 
 # disco
 DeviceNotFound: 'Dispositivo {{ .name }} não encontrado'
@@ -730,3 +731,7 @@ FileTaskRename: 'Nome do arquivo de destino: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "O backend atual {{ .current }} ainda contém regras do 1Panel. Remova-as antes de mudar para {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "O encaminhamento IPv4 está desativado. Defina net.ipv4.ip_forward=1 antes de usar o backend de firewall do Docker."
 ErrFirewallRuleSavedApplyFailed: "A configuração da nova regra foi salva, mas não pôde ser aplicada ao firewall. Tente novamente por meio da sincronização: {{ .detail }}"
+
+ErrSSHPublicKey: "Informe uma chave pública SSH válida"
+ErrSSHPrivateKey: "Chave privada SSH inválida ou senha incorreta; chaves criptografadas exigem a senha original"
+ErrSSHKeyMismatch: "As chaves pública e privada SSH não correspondem"

agent/i18n/lang/ru.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'Ваш панель {{ .node }}{{ .ip }}, {{ .msg }}. Войдит
 NodeExceptionAlert: 'Ваш панель {{ .node }}{{ .ip }}, {{ .num }} узлов работают с ошибками. Войдите в панель для получения деталей.'
 LicenseExceptionAlert: 'Ваш панель {{ .node }}{{ .ip }}, {{ .num }} лицензий имеют ошибки. Войдите в панель для получения деталей.'
 SSHAndPanelLoginAlert: 'Ваш панель {{ .node }}{{ .ip }}, вход {{ .name }} с адреса {{ .loginIp }} является аномальным. Войдите в панель для получения деталей.'
+CronJobSuccessAlert: 'Панель {{ .node }}{{ .ip }}: запланированная задача {{ .name }} успешно завершена. Войдите для просмотра подробностей.'
 
 # диск
 DeviceNotFound: 'Устройство {{ .name }} не найдено'
@@ -730,3 +731,7 @@ FileTaskRename: 'Имя целевого файла: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "В текущем бэкенде {{ .current }} остались правила 1Panel. Удалите их перед переключением на {{ .target }}."
 ErrDockerIPv4ForwardingDisabled: "Пересылка IPv4 отключена. Перед использованием бэкенда межсетевого экрана Docker установите net.ipv4.ip_forward=1."
 ErrFirewallRuleSavedApplyFailed: "Настройки нового правила сохранены, но применить их к межсетевому экрану не удалось. Повторите попытку с помощью синхронизации: {{ .detail }}"
+
+ErrSSHPublicKey: "Введите действительный открытый ключ SSH"
+ErrSSHPrivateKey: "Недопустимый закрытый ключ SSH или неверная парольная фраза; для зашифрованных ключей требуется исходная парольная фраза"
+ErrSSHKeyMismatch: "Открытый и закрытый ключи SSH не соответствуют друг другу"

agent/i18n/lang/tr.yaml

@@ -649,6 +649,7 @@ CommonAlert: 'Paneliniz {{ .node }}{{ .ip }}, {{ .msg }}. Detaylar için panelin
 NodeExceptionAlert: 'Paneliniz {{ .node }}{{ .ip }}, {{ .num }} düğüm anormal durumda. Detaylar için paneline giriş yapın.'
 LicenseExceptionAlert: 'Paneliniz {{ .node }}{{ .ip }}, {{ .num }} lisans anormal durumda. Detaylar için paneline giriş yapın.'
 SSHAndPanelLoginAlert: 'Paneliniz {{ .node }}{{ .ip }}, {{ .loginIp }} adresinden {{ .name }} girişi anormal. Detaylar için paneline giriş yapın.'
+CronJobSuccessAlert: 'Panel {{ .node }}{{ .ip }}: zamanlanmış görev {{ .name }} başarıyla tamamlandı. Ayrıntıları görmek için giriş yapın.'
 
 # disk
 DeviceNotFound: 'Cihaz {{ .name }} bulunamadı'
@@ -730,3 +731,7 @@ FileTaskRename: 'Hedef dosya adı: {{ .name }}'
 ErrFirewallBackendCleanupRequired: "Mevcut {{ .current }} arka ucunda hâlâ 1Panel kuralları var. {{ .target }} arka ucuna geçmeden önce bunları temizleyin."
 ErrDockerIPv4ForwardingDisabled: "IPv4 yönlendirmesi devre dışı. Docker güvenlik duvarı arka ucunu kullanmadan önce net.ipv4.ip_forward=1 ayarını yapın."
 ErrFirewallRuleSavedApplyFailed: "Yeni kural yapılandırması kaydedildi, ancak güvenlik duvarına uygulanamadı. Eşitleme yaparak yeniden deneyin: {{ .detail }}"
+
+ErrSSHPublicKey: "Geçerli bir SSH açık anahtarı girin"
+ErrSSHPrivateKey: "SSH özel anahtarı geçersiz veya parola yanlış; şifreli anahtarlar mevcut parolayı gerektirir"
+ErrSSHKeyMismatch: "SSH açık ve özel anahtarları eşleşmiyor"

agent/i18n/lang/zh-Hant.yaml

@@ -649,6 +649,7 @@ CommonAlert: '您的 {{ .node }}{{ .ip }} 面板，{{ .msg }}，詳情請登入
 NodeExceptionAlert: '您的 {{ .node }}{{ .ip }} 面板，{{ .num }} 個節點出現異常，詳情請登入面板檢視。'
 LicenseExceptionAlert: '您的 {{ .node }}{{ .ip }} 面板，{{ .num }} 個授權出現異常，詳情請登入面板檢視。'
 SSHAndPanelLoginAlert: '您的 {{ .node }}{{ .ip }} 面板，來自 {{ .loginIp }} 的 {{ .name }} 登入出現異常，詳情請登入面板檢視。'
+CronJobSuccessAlert: '您的 {{ .node }}{{ .ip }} 面板，排程任務 {{ .name }} 執行成功，詳情請登入面板檢視。'
 
 # 磁碟
 DeviceNotFound: '裝置 {{ .name }} 未找到'
@@ -730,3 +731,17 @@ FileTaskRename: '目標檔名：{{ .name }}'
 ErrFirewallBackendCleanupRequired: "目前後端 {{ .current }} 中仍有 1Panel 規則，請先清理後再切換至 {{ .target }}。"
 ErrDockerIPv4ForwardingDisabled: "IPv4 轉送尚未啟用，請先設定 net.ipv4.ip_forward=1，再使用 Docker 防火牆後端。"
 ErrFirewallRuleSavedApplyFailed: "新規則設定已儲存，但套用至防火牆失敗。可透過同步重試：{{ .detail }}"
+
+ErrSSHPublicKey: "請輸入有效的 SSH 公鑰"
+ErrSSHPrivateKey: "SSH 私鑰格式無效或密碼錯誤；加密私鑰需要填寫原密碼"
+ErrSSHKeyMismatch: "SSH 公鑰與私鑰不相符"
+
+NetworkCleanupDeleted: "網路 [{{ .name }}] ({{ .id }}) 已刪除"
+NetworkCleanupProtected: "跳過網路 [{{ .name }}] ({{ .id }})：預設保留網路，未刪除"
+NetworkCleanupConnected: "跳過網路 [{{ .name }}] ({{ .id }})：仍有容器連接或正在使用，未刪除"
+NetworkCleanupUnsupported: "跳過網路 [{{ .name }}] ({{ .id }})：特殊網路，未刪除"
+NetworkCleanupGone: "跳過網路 [{{ .name }}] ({{ .id }})：網路已不存在"
+NetworkCleanupInspectFailed: "網路 [{{ .name }}] ({{ .id }}) 檢查失敗，未刪除"
+NetworkCleanupRemoveFailed: "網路 [{{ .name }}] ({{ .id }}) 刪除失敗"
+NetworkCleanupSummary: "網路清理完成：已刪除 {{ .deleted }}，已跳過 {{ .skipped }}，失敗 {{ .failed }}"
+NetworkCleanupPartialFailure: "部分網路清理失敗，請查看任務日誌"

agent/i18n/lang/zh.yaml

@@ -649,6 +649,7 @@ CommonAlert: "您的 {{ .node }}{{ .ip }} 面板，{{ .msg }}，请登录面板
 NodeExceptionAlert: "您的 {{ .node }}{{ .ip }} 面板，{{ .num }} 个节点存在异常，请登录面板查看详情。"
 LicenseExceptionAlert: "您的 {{ .node }}{{ .ip }} 面板，{{ .num }} 个许可证存在异常，请登录面板查看详情。"
 SSHAndPanelLoginAlert: "您的 {{ .node }}{{ .ip }} 面板，面板 {{ .name }} 登录 {{ .loginIp }} 异常，请登录面板查看详情。"
+CronJobSuccessAlert: "您的 {{ .node }}{{ .ip }} 面板，计划任务-{{ .name }}执行成功，请登录面板查看详情。"
 
 # 磁盘
 DeviceNotFound: "设备 {{ .name }} 未找到"
@@ -730,3 +731,17 @@ FileTaskRename: '目标文件名：{{ .name }}'
 ErrFirewallBackendCleanupRequired: "当前后端 {{ .current }} 中仍有 1Panel 规则，请先清理后再切换到 {{ .target }}。"
 ErrDockerIPv4ForwardingDisabled: "IPv4 转发未开启，请先设置 net.ipv4.ip_forward=1，再使用 Docker 防火墙后端。"
 ErrFirewallRuleSavedApplyFailed: "新规则配置已保存，但应用到防火墙失败。可通过同步重试：{{ .detail }}"
+
+ErrSSHPublicKey: "请输入有效的 SSH 公钥"
+ErrSSHPrivateKey: "SSH 私钥格式无效或密码错误；加密私钥需要填写原密码"
+ErrSSHKeyMismatch: "SSH 公钥与私钥不匹配"
+
+NetworkCleanupDeleted: "网络 [{{ .name }}] ({{ .id }}) 已删除"
+NetworkCleanupProtected: "跳过网络 [{{ .name }}] ({{ .id }})：默认保留网络，未删除"
+NetworkCleanupConnected: "跳过网络 [{{ .name }}] ({{ .id }})：仍有容器连接或正在使用，未删除"
+NetworkCleanupUnsupported: "跳过网络 [{{ .name }}] ({{ .id }})：特殊网络，未删除"
+NetworkCleanupGone: "跳过网络 [{{ .name }}] ({{ .id }})：网络已不存在"
+NetworkCleanupInspectFailed: "网络 [{{ .name }}] ({{ .id }}) 检查失败，未删除"
+NetworkCleanupRemoveFailed: "网络 [{{ .name }}] ({{ .id }}) 删除失败"
+NetworkCleanupSummary: "网络清理完成：已删除 {{ .deleted }}，已跳过 {{ .skipped }}，失败 {{ .failed }}"
+NetworkCleanupPartialFailure: "部分网络清理失败，请查看任务日志"

agent/init/db/db.go

@@ -13,6 +13,7 @@ func Init() {
 	global.TaskDB = common.LoadDBConnByPath(path.Join(global.Dir.DbDir, "task.db"), "task")
 	global.MonitorDB = common.LoadDBConnByPath(path.Join(global.Dir.DbDir, "monitor.db"), "monitor")
 	global.GPUMonitorDB = common.LoadDBConnByPath(path.Join(global.Dir.DbDir, "gpu_monitor.db"), "gpu_monitor")
+	global.VLLMMonitorDB = common.LoadDBConnByPath(path.Join(global.Dir.DbDir, "vllm_monitor.db"), "vllm_monitor")
 	global.AlertDB = common.LoadDBConnByPath(path.Join(global.Dir.DbDir, "alert.db"), "alert")
 
 	if _, err := os.Stat(path.Join(global.Dir.DbDir, "core.db")); err == nil {

agent/init/hook/hook.go

@@ -4,6 +4,7 @@ import (
 	"os"
 	"os/exec"
 	"strings"
+	"time"
 
 	"github.com/1Panel-dev/1Panel/agent/app/dto"
 	"github.com/1Panel-dev/1Panel/agent/app/model"
@@ -13,6 +14,7 @@ import (
 	"github.com/1Panel-dev/1Panel/agent/global"
 	"github.com/1Panel-dev/1Panel/agent/utils/alert_push"
 	"github.com/1Panel-dev/1Panel/agent/utils/xpack"
+	"gorm.io/gorm"
 )
 
 func Init() {
@@ -174,6 +176,36 @@ func initAlertTask() {
 }
 
 func initMonitorDB() {
-	_ = global.MonitorDB.AutoMigrate(&model.MonitorBase{}, &model.MonitorNetwork{}, &model.MonitorGPU{}, &model.MonitorIO{})
+	_ = global.MonitorDB.AutoMigrate(&model.MonitorBase{}, &model.MonitorNetwork{}, &model.MonitorIO{})
+	_ = global.GPUMonitorDB.AutoMigrate(&model.MonitorGPU{})
 	_ = global.TaskDB.AutoMigrate(&model.Task{})
+	// building indexes on large monitor tables can take seconds, keep it off the startup path;
+	// WAL mode leaves readers unblocked and busy_timeout covers the collector's inserts meanwhile
+	go ensureMonitorIndexes()
+}
+
+func ensureMonitorIndexes() {
+	indexes := []struct {
+		db   *gorm.DB
+		stmt string
+	}{
+		// created_at alone serves unfiltered range queries and retention cleanup
+		{global.MonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_bases_created ON monitor_bases(created_at)"},
+		{global.MonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_ios_created ON monitor_ios(created_at)"},
+		{global.MonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_networks_created ON monitor_networks(created_at)"},
+		{global.GPUMonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_gpus_created ON monitor_gpus(created_at)"},
+		// (name, created_at) serves per-device range queries and distinct name lookups
+		{global.MonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_ios_name_created ON monitor_ios(name, created_at)"},
+		{global.MonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_networks_name_created ON monitor_networks(name, created_at)"},
+		{global.GPUMonitorDB, "CREATE INDEX IF NOT EXISTS idx_monitor_gpus_product_created ON monitor_gpus(product_name, created_at)"},
+	}
+	start := time.Now()
+	for _, index := range indexes {
+		if err := index.db.Exec(index.stmt).Error; err != nil {
+			global.LOG.Warnf("create monitor index failed, stmt: %s, err: %v", index.stmt, err)
+		}
+	}
+	if elapsed := time.Since(start); elapsed > time.Second {
+		global.LOG.Infof("monitor indexes ready, took %s", elapsed)
+	}
 }

agent/init/migration/migrate.go

@@ -111,6 +111,9 @@ func agentDBMigrations() []*gormigrate.Migration {
 		migrations.SimplifyFirewallRulePolicy,
 		migrations.AddDockerPortGuardReadOnly,
 		migrations.MigrateFirewallPortWhitelistSources,
+		migrations.AddAcceleratorMetrics,
+		migrations.AddVLLMMonitor,
+		migrations.AddMonitorSettings,
 	}
 }
 

agent/init/migration/migrations/init.go

@@ -1908,3 +1908,35 @@ var AddDockerPortGuardReadOnly = &gormigrate.Migration{
 		return tx.Migrator().CreateIndex(&model.DockerPortGuardPolicy{}, "idx_docker_port_guard_endpoint")
 	},
 }
+
+var AddAcceleratorMetrics = &gormigrate.Migration{
+	ID: "20260928-accelerator-vendor-metrics",
+	Migrate: func(tx *gorm.DB) error {
+		return global.GPUMonitorDB.AutoMigrate(&model.MonitorGPU{})
+	},
+}
+
+var AddVLLMMonitor = &gormigrate.Migration{
+	ID: "20260928-vllm-monitor",
+	Migrate: func(tx *gorm.DB) error {
+		return global.VLLMMonitorDB.AutoMigrate(&model.MonitorVLLM{})
+	},
+}
+
+var AddMonitorSettings = &gormigrate.Migration{
+	ID: "20260929-monitor-settings",
+	Migrate: func(tx *gorm.DB) error {
+		for _, key := range []string{"MonitorStatus", "MonitorInterval", "MonitorStoreDays"} {
+			var setting model.Setting
+			if err := tx.Where("key = ?", key).First(&setting).Error; err != nil {
+				return err
+			}
+			for _, prefix := range []string{"GPU", "VLLM"} {
+				if err := tx.Where("key = ?", prefix+key).FirstOrCreate(&model.Setting{Key: prefix + key, Value: setting.Value}).Error; err != nil {
+					return err
+				}
+			}
+		}
+		return nil
+	},
+}

agent/router/ro_ai.go

@@ -27,6 +27,9 @@ func (a *AIToolsRouter) InitRouter(Router *gin.RouterGroup) {
 		aiToolsRouter.GET("/gpu/load", baseApi.LoadGpuInfo)
 		aiToolsRouter.POST("/gpu/search", baseApi.LoadGPUMonitor)
 		aiToolsRouter.GET("/gpu/options", baseApi.GetCPUOptions)
+		aiToolsRouter.POST("/vllm/monitor/search", baseApi.LoadVLLMMonitor)
+		aiToolsRouter.POST("/vllm/monitor/current", baseApi.LoadVLLMCurrent)
+		aiToolsRouter.POST("/vllm/monitor/clean", baseApi.CleanVLLMMonitor)
 
 		aiToolsRouter.POST("/mcp/search", baseApi.PageMcpServers)
 		aiToolsRouter.POST("/mcp/server/detail", baseApi.LoadMcpServerDetail)

agent/router/ro_container.go

@@ -77,6 +77,7 @@ func (s *ContainerRouter) InitRouter(Router *gin.RouterGroup) {
 
 		baRouter.GET("/network", baseApi.ListNetwork)
 		baRouter.POST("/network/del", baseApi.DeleteNetwork)
+		baRouter.POST("/network/clean", baseApi.CleanNetworks)
 		baRouter.POST("/network/search", baseApi.SearchNetwork)
 		baRouter.POST("/network", baseApi.CreateNetwork)
 		baRouter.GET("/volume", baseApi.ListVolume)

agent/utils/ai_tools/accelerator/accelerator.go

@@ -6,6 +6,7 @@ import (
 	"fmt"
 	"strings"
 	"sync"
+	"time"
 
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/gpu"
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/npu"
@@ -22,14 +23,6 @@ type providerResult struct {
 }
 
 func New() (bool, Client) {
-	return newClient()
-}
-
-func NewAll() (bool, Client) {
-	return New()
-}
-
-func newClient() (bool, Client) {
 	client := Client{}
 	if available, gpuClient := gpu.New(); available {
 		client.providers = append(client.providers, gpuProvider{client: gpuClient})
@@ -52,6 +45,11 @@ func (c Client) LoadInfo() (*Info, error) {
 }
 
 func (c Client) Collect(ctx context.Context) (*Snapshot, error) {
+	if err := ctx.Err(); err != nil {
+		return nil, err
+	}
+	ctx, cancel := context.WithTimeout(ctx, 15*time.Second)
+	defer cancel()
 	results := make([]providerResult, len(c.providers))
 	var wg sync.WaitGroup
 	for index, item := range c.providers {
@@ -66,8 +64,11 @@ func (c Client) Collect(ctx context.Context) (*Snapshot, error) {
 		}()
 	}
 	wg.Wait()
+	if err := ctx.Err(); err != nil {
+		return nil, err
+	}
 
-	snapshot := &Snapshot{DriverVersions: make(map[string]string)}
+	snapshot := &Snapshot{}
 	var (
 		errs       []error
 		active     []*ProviderSnapshot
@@ -78,6 +79,11 @@ func (c Client) Collect(ctx context.Context) (*Snapshot, error) {
 			errs = append(errs, result.err)
 			continue
 		}
+		if result.snapshot != nil {
+			for _, warning := range result.snapshot.Warnings {
+				errs = append(errs, errors.New(warning))
+			}
+		}
 		if result.snapshot == nil || len(result.snapshot.Devices) == 0 {
 			continue
 		}
@@ -90,11 +96,8 @@ func (c Client) Collect(ctx context.Context) (*Snapshot, error) {
 		if item.CudaVersion != "" {
 			snapshot.Info.CudaVersion = item.CudaVersion
 		}
-		if item.DriverVersion != "" {
-			snapshot.DriverVersions[item.Type] = item.DriverVersion
-			if item.Type == "xpu" {
-				xpuVersion = item.DriverVersion
-			}
+		if item.Type == "xpu" {
+			xpuVersion = item.DriverVersion
 		}
 	}
 	snapshot.Warnings = append(snapshot.Warnings, errs...)
@@ -102,6 +105,10 @@ func (c Client) Collect(ctx context.Context) (*Snapshot, error) {
 		return nil, fmt.Errorf("calling accelerator monitoring tools failed: %w", errors.Join(errs...))
 	}
 
+	snapshot.Info.CollectedAt = time.Now()
+	for _, err := range errs {
+		snapshot.Info.Warnings = append(snapshot.Info.Warnings, err.Error())
+	}
 	snapshot.Info.XPUDriverVersion = xpuVersion
 	snapshot.Info.Type, snapshot.Info.DriverVersion = mergeProviderMetadata(active)
 	return snapshot, nil

agent/utils/ai_tools/accelerator/metric.go

@@ -1,6 +1,7 @@
 package accelerator
 
 import (
+	"math"
 	"strconv"
 	"strings"
 
@@ -18,11 +19,11 @@ func metric(display, unit string) Metric {
 		return result
 	}
 	parsed, err := strconv.ParseFloat(matched[1], 64)
-	if err != nil {
+	if err != nil || math.IsNaN(parsed) || math.IsInf(parsed, 0) || (parsed < 0 && unit != "°C") {
 		return result
 	}
 	normalized, ok := convertMetricUnit(parsed, matched[2], unit)
-	if !ok {
+	if !ok || math.IsNaN(normalized) || math.IsInf(normalized, 0) || (unit == "%" && normalized > 100) {
 		return result
 	}
 	result.Value = &normalized

agent/utils/ai_tools/accelerator/normalize.go

@@ -12,26 +12,38 @@ import (
 
 func normalizeGPU(item *gpu.Device) Device {
 	metrics := Metrics{
-		Utilization: metric(item.GPUUtil, "%"),
-		Temperature: metric(item.Temperature, "°C"),
-		Power:       metric(item.PowerDraw, "W"),
-		PowerLimit:  metric(item.MaxPowerLimit, "W"),
-		MemoryUsed:  memoryMetric(item.MemUsed),
-		MemoryTotal: memoryMetric(item.MemTotal),
-		FanSpeed:    metric(item.FanSpeed, "%"),
+		MemoryActivity:     metric(item.MemoryActivity, "%"),
+		EncoderUtil:        metric(item.EncoderUtil, "%"),
+		DecoderUtil:        metric(item.DecoderUtil, "%"),
+		JPEGUtil:           metric(item.JPEGUtil, "%"),
+		OFAUtil:            metric(item.OFAUtil, "%"),
+		MediaUtil:          metric(item.MediaUtil, "%"),
+		HotspotTemperature: metric(item.HotspotTemperature, "°C"),
+		FanRPM:             metric(item.FanRPM, "RPM"),
+		MediaFrequency:     metric(item.MediaFrequency, "MHz"),
+		Utilization:        metric(item.GPUUtil, "%"),
+		Temperature:        metric(item.Temperature, "°C"),
+		MemoryTemperature:  metric(item.MemoryTemperature, "°C"),
+		Power:              metric(item.PowerDraw, "W"),
+		PowerLimit:         metric(item.PowerLimit, "W"),
+		Frequency:          metric(item.Frequency, "MHz"),
+		MemoryFrequency:    metric(item.MemoryFrequency, "MHz"),
+		MemoryUsed:         memoryMetric(item.MemUsed),
+		MemoryTotal:        memoryMetric(item.MemTotal),
+		FanSpeed:           metric(item.FanSpeed, "%"),
 	}
 	device := Device{
-		ID:      stableID(item.Type, item.BusID, strconv.FormatUint(uint64(item.Index), 10)),
-		Kind:    KindGPU,
-		Vendor:  item.Type,
-		Index:   int(item.Index),
-		Name:    item.ProductName,
-		Label:   fmt.Sprintf("%d - %s", item.Index, item.ProductName),
-		BusID:   item.BusID,
-		Metrics: metrics,
-		GPU:     item,
+		ProcessStatus: item.ProcessStatus,
+		ID:            stableID(item.Type, item.UUID, stableID("pci", item.BusID, strconv.FormatUint(uint64(item.Index), 10))),
+		Kind:          KindGPU,
+		Vendor:        item.Type,
+		Index:         int(item.Index),
+		Name:          item.ProductName,
+		Label:         fmt.Sprintf("%d - %s", item.Index, item.ProductName),
+		BusID:         item.BusID,
+		Metrics:       metrics,
+		GPU:           item,
 	}
-	device.Capabilities = capabilities(metrics)
 	for _, process := range item.Processes {
 		device.Processes = append(device.Processes, Process{
 			PID:    process.PID,
@@ -45,26 +57,36 @@ func normalizeGPU(item *gpu.Device) Device {
 
 func normalizeNPU(item *npu.Device) Device {
 	metrics := Metrics{
-		Utilization: metric(item.AICore, "%"),
-		Temperature: metric(item.Temperature, "°C"),
-		Power:       metric(item.PowerDraw, "W"),
-		MemoryUsed:  memoryMetric(item.MemUsed),
-		MemoryTotal: memoryMetric(item.MemTotal),
+		AICPUUtil:      metric(item.AICPUUtil, "%"),
+		CtrlCPUUtil:    metric(item.CtrlCPUUtil, "%"),
+		DDRBandwidth:   metric(item.DDRBandwidth, "%"),
+		HBMBandwidth:   metric(item.HBMBandwidth, "%"),
+		DDRUsed:        metric(item.MemoryUsed, "MiB"),
+		DDRTotal:       metric(item.MemoryTotal, "MiB"),
+		HBMUsed:        metric(item.HBMUsed, "MiB"),
+		HBMTotal:       metric(item.HBMTotal, "MiB"),
+		HugepagesUsed:  metric(item.HugepagesUsed, "pages"),
+		HugepagesTotal: metric(item.HugepagesTotal, "pages"),
+		Utilization:    metric(item.AICore, "%"),
+		Temperature:    metric(item.Temperature, "°C"),
+		Power:          metric(item.PowerDraw, "W"),
+		MemoryUsed:     memoryMetric(item.MemUsed),
+		MemoryTotal:    memoryMetric(item.MemTotal),
 	}
 	device := Device{
-		ID:        fmt.Sprintf("ascend:%d:%d", item.NPUIndex, item.ChipIndex),
-		Kind:      KindNPU,
-		Vendor:    "ascend",
-		Index:     int(item.Index),
-		NPUIndex:  int(item.NPUIndex),
-		ChipIndex: int(item.ChipIndex),
-		Name:      item.ProductName,
-		Label:     fmt.Sprintf("NPU %d / Chip %d - %s", item.NPUIndex, item.ChipIndex, item.ProductName),
-		BusID:     item.BusID,
-		Metrics:   metrics,
-		NPU:       item,
+		ProcessStatus: item.ProcessStatus,
+		ID:            fmt.Sprintf("%s:%d", stableID("ascend", item.BusID, strconv.FormatUint(uint64(item.NPUIndex), 10)), item.ChipIndex),
+		Kind:          KindNPU,
+		Vendor:        "ascend",
+		Index:         int(item.Index),
+		NPUIndex:      int(item.NPUIndex),
+		ChipIndex:     int(item.ChipIndex),
+		Name:          item.ProductName,
+		Label:         fmt.Sprintf("NPU %d / Chip %d - %s", item.NPUIndex, item.ChipIndex, item.ProductName),
+		BusID:         item.BusID,
+		Metrics:       metrics,
+		NPU:           item,
 	}
-	device.Capabilities = capabilities(metrics)
 	for _, process := range item.Processes {
 		device.Processes = append(device.Processes, Process{
 			PID:    process.PID,
@@ -78,26 +100,32 @@ func normalizeNPU(item *npu.Device) Device {
 
 func normalizeXPU(item *xpu.Device) Device {
 	metrics := Metrics{
-		Utilization: metric(item.Stats.GPUUtil, "%"),
-		Temperature: metric(item.Stats.Temperature, "°C"),
-		Power:       metric(item.Stats.Power, "W"),
-		MemoryUsed:  memoryMetric(item.Stats.MemoryUsed),
-		MemoryTotal: memoryMetric(item.Basic.Memory),
-		MemoryUtil:  metric(item.Stats.MemoryUtil, "%"),
-		Frequency:   metric(item.Stats.Frequency, "MHz"),
+		MediaUtil:         metric(item.Stats.MediaUtil, "%"),
+		ComputeUtil:       metric(item.Stats.ComputeUtil, "%"),
+		CopyUtil:          metric(item.Stats.CopyUtil, "%"),
+		MediaFrequency:    metric(item.Stats.MediaFrequency, "MHz"),
+		MemoryTemperature: metric(item.Stats.MemoryTemperature, "°C"),
+		MemoryBandwidth:   metric(item.Stats.MemoryBandwidthUtil, "%"),
+		Utilization:       metric(item.Stats.GPUUtil, "%"),
+		Temperature:       metric(item.Stats.Temperature, "°C"),
+		Power:             metric(item.Stats.Power, "W"),
+		MemoryUsed:        memoryMetric(item.Stats.MemoryUsed),
+		MemoryTotal:       memoryMetric(item.Basic.Memory),
+		MemoryUtil:        metric(item.Stats.MemoryUtil, "%"),
+		Frequency:         metric(item.Stats.Frequency, "MHz"),
 	}
 	device := Device{
-		ID:      stableID("xpu", item.Basic.PciBdfAddress, strconv.Itoa(item.Basic.DeviceID)),
-		Kind:    KindXPU,
-		Vendor:  item.Basic.VendorName,
-		Index:   item.Basic.DeviceID,
-		Name:    item.Basic.DeviceName,
-		Label:   fmt.Sprintf("%d - %s", item.Basic.DeviceID, item.Basic.DeviceName),
-		BusID:   item.Basic.PciBdfAddress,
-		Metrics: metrics,
-		XPU:     item,
+		ProcessStatus: item.ProcessStatus,
+		ID:            stableID("xpu", item.Basic.UUID, stableID("pci", item.Basic.PciBdfAddress, strconv.Itoa(item.Basic.DeviceID))),
+		Kind:          KindXPU,
+		Vendor:        item.Basic.VendorName,
+		Index:         item.Basic.DeviceID,
+		Name:          item.Basic.DeviceName,
+		Label:         fmt.Sprintf("%d - %s", item.Basic.DeviceID, item.Basic.DeviceName),
+		BusID:         item.Basic.PciBdfAddress,
+		Metrics:       metrics,
+		XPU:           item,
 	}
-	device.Capabilities = capabilities(metrics)
 	for _, process := range item.Processes {
 		device.Processes = append(device.Processes, Process{
 			PID:          strconv.Itoa(process.PID),
@@ -110,18 +138,6 @@ func normalizeXPU(item *xpu.Device) Device {
 	return device
 }
 
-func capabilities(metrics Metrics) Capabilities {
-	return Capabilities{
-		Utilization: metrics.Utilization.Available(),
-		Temperature: metrics.Temperature.Available(),
-		Power:       metrics.Power.Available(),
-		PowerLimit:  metrics.PowerLimit.Available(),
-		Memory:      metrics.MemoryUsed.Available() || metrics.MemoryTotal.Available(),
-		FanSpeed:    metrics.FanSpeed.Available(),
-		Frequency:   metrics.Frequency.Available(),
-	}
-}
-
 func stableID(vendor, busID, fallback string) string {
 	if busID != "" && !strings.EqualFold(busID, "N/A") {
 		return vendor + ":" + busID

agent/utils/ai_tools/accelerator/provider.go

@@ -2,6 +2,7 @@ package accelerator
 
 import (
 	"context"
+	"fmt"
 
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/gpu"
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/npu"
@@ -26,6 +27,7 @@ func (p gpuProvider) Collect(ctx context.Context) (*ProviderSnapshot, error) {
 	}
 	result := &ProviderSnapshot{
 		Type:          info.Type,
+		Warnings:      info.Warnings,
 		DriverVersion: info.DriverVersion,
 		CudaVersion:   info.CudaVersion,
 		GPUs:          info.Devices,
@@ -49,6 +51,7 @@ func (p npuProvider) Collect(ctx context.Context) (*ProviderSnapshot, error) {
 	}
 	result := &ProviderSnapshot{
 		Type:          info.Type,
+		Warnings:      info.Warnings,
 		DriverVersion: info.DriverVersion,
 		NPUs:          info.Devices,
 	}
@@ -71,11 +74,24 @@ func (p xpuProvider) Collect(ctx context.Context) (*ProviderSnapshot, error) {
 	}
 	result := &ProviderSnapshot{
 		Type:          info.Type,
+		Warnings:      info.Warnings,
 		DriverVersion: info.DriverVersion,
 		XPUs:          info.Devices,
 	}
 	for index := range result.XPUs {
-		result.Devices = append(result.Devices, normalizeXPU(&result.XPUs[index]))
+
+		parent := normalizeXPU(&result.XPUs[index])
+		result.Devices = append(result.Devices, parent)
+		for _, tile := range result.XPUs[index].Tiles {
+			raw := xpu.Device{Basic: result.XPUs[index].Basic, Stats: tile.Stats, ProcessStatus: "unavailable"}
+			raw.Basic.Memory = ""
+			raw.Basic.FreeMemory = ""
+			device := normalizeXPU(&raw)
+			device.ParentID = parent.ID
+			device.ID = fmt.Sprintf("%s:tile:%d", parent.ID, tile.TileID)
+			device.Label = fmt.Sprintf("%s / Tile %d", parent.Label, tile.TileID)
+			result.Devices = append(result.Devices, device)
+		}
 	}
 	return result, nil
 }

agent/utils/ai_tools/accelerator/types.go

@@ -2,6 +2,7 @@ package accelerator
 
 import (
 	"errors"
+	"time"
 
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/gpu"
 	"github.com/1Panel-dev/1Panel/agent/utils/ai_tools/npu"
@@ -17,6 +18,9 @@ const (
 )
 
 type Info struct {
+	CollectedAt time.Time `json:"collectedAt"`
+	Warnings    []string  `json:"warnings"`
+
 	Type             string       `json:"type"`
 	CudaVersion      string       `json:"cudaVersion"`
 	DriverVersion    string       `json:"driverVersion"`
@@ -44,25 +48,40 @@ func (m Metric) ValueOrZero() float64 {
 }
 
 type Metrics struct {
-	Utilization Metric
-	Temperature Metric
-	Power       Metric
-	PowerLimit  Metric
-	MemoryUsed  Metric
-	MemoryTotal Metric
-	MemoryUtil  Metric
-	FanSpeed    Metric
-	Frequency   Metric
-}
+	MemoryActivity     Metric
+	EncoderUtil        Metric
+	DecoderUtil        Metric
+	JPEGUtil           Metric
+	OFAUtil            Metric
+	MediaUtil          Metric
+	ComputeUtil        Metric
+	CopyUtil           Metric
+	HotspotTemperature Metric
+	FanRPM             Metric
+	AICPUUtil          Metric
+	CtrlCPUUtil        Metric
+	DDRUsed            Metric
+	DDRTotal           Metric
+	HBMUsed            Metric
+	HBMTotal           Metric
+	DDRBandwidth       Metric
+	HBMBandwidth       Metric
+	MemoryBandwidth    Metric
+	MediaFrequency     Metric
+	HugepagesUsed      Metric
+	HugepagesTotal     Metric
 
-type Capabilities struct {
-	Utilization bool
-	Temperature bool
-	Power       bool
-	PowerLimit  bool
-	Memory      bool
-	FanSpeed    bool
-	Frequency   bool
+	MemoryTemperature Metric
+	Utilization       Metric
+	Temperature       Metric
+	Power             Metric
+	PowerLimit        Metric
+	MemoryUsed        Metric
+	MemoryTotal       Metric
+	MemoryUtil        Metric
+	FanSpeed          Metric
+	Frequency         Metric
+	MemoryFrequency   Metric
 }
 
 type Process struct {
@@ -74,18 +93,19 @@ type Process struct {
 }
 
 type Device struct {
-	ID           string
-	Kind         Kind
-	Vendor       string
-	Index        int
-	NPUIndex     int
-	ChipIndex    int
-	Name         string
-	Label        string
-	BusID        string
-	Metrics      Metrics
-	Capabilities Capabilities
-	Processes    []Process
+	ParentID      string
+	ProcessStatus string
+	ID            string
+	Kind          Kind
+	Vendor        string
+	Index         int
+	NPUIndex      int
+	ChipIndex     int
+	Name          string
+	Label         string
+	BusID         string
+	Metrics       Metrics
+	Processes     []Process
 
 	GPU *gpu.Device `json:"-"`
 	NPU *npu.Device `json:"-"`
@@ -93,10 +113,9 @@ type Device struct {
 }
 
 type Snapshot struct {
-	Info           Info
-	Devices        []Device
-	DriverVersions map[string]string
-	Warnings       []error
+	Info     Info
+	Devices  []Device
+	Warnings []error
 }
 
 func (s Snapshot) Warning() error {
@@ -104,6 +123,7 @@ func (s Snapshot) Warning() error {
 }
 
 type ProviderSnapshot struct {
+	Warnings      []string
 	Type          string
 	DriverVersion string
 	CudaVersion   string

agent/utils/ai_tools/gpu/amd.go

@@ -3,12 +3,13 @@ package gpu
 import (
 	"context"
 	"fmt"
+	"math"
 	"sort"
+	"strconv"
 	"strings"
 	"sync"
 	"time"
 
-	"github.com/1Panel-dev/1Panel/agent/global"
 	"github.com/1Panel-dev/1Panel/agent/utils/cmd"
 )
 
@@ -32,16 +33,22 @@ func findAMDSMI() (string, bool) {
 
 func (a amdSMI) LoadInfo(ctx context.Context) (*Info, error) {
 	var (
-		staticData  string
-		metricData  string
-		processData string
-		staticErr   error
-		metricErr   error
-		processErr  error
-		wg          sync.WaitGroup
+		staticData   string
+		metricData   string
+		extendedData string
+		extendedErr  error
+		processData  string
+		staticErr    error
+		metricErr    error
+		processErr   error
+		wg           sync.WaitGroup
 	)
 
-	wg.Add(3)
+	wg.Add(4)
+	go func() {
+		defer wg.Done()
+		extendedData, extendedErr = runAMDSMI(ctx, a.command, "metric", "--clock", "--ecc", "--json")
+	}()
 	go func() {
 		defer wg.Done()
 		staticData, staticErr = runAMDSMI(ctx, a.command, "static", "--asic", "--bus", "--driver", "--limit", "--json")
@@ -64,16 +71,25 @@ func (a amdSMI) LoadInfo(ctx context.Context) (*Info, error) {
 		return nil, fmt.Errorf("parsing %s static output failed: %w", a.command, err)
 	}
 
+	if metricErr == nil {
+		metricErr = applyAMDMetrics(info, metricData)
+	}
 	if metricErr != nil {
-		global.LOG.Warnf("calling %s metric failed, metrics will be omitted: %v", a.command, metricErr)
-	} else if err := applyAMDMetrics(info, metricData); err != nil {
-		global.LOG.Warnf("parsing %s metric output failed, metrics will be omitted: %v", a.command, err)
+		info.Warnings = append(info.Warnings, fmt.Sprintf("%s metrics: %v", a.command, metricErr))
+	}
+	if extendedErr == nil {
+		extendedErr = applyAMDMetrics(info, extendedData)
+	}
+	if extendedErr != nil {
+		info.Warnings = append(info.Warnings, fmt.Sprintf("%s extended metrics: %v", a.command, extendedErr))
+	}
+	if processErr == nil {
+		processErr = applyAMDProcesses(info, processData)
 	}
 	if processErr != nil {
-		global.LOG.Warnf("calling %s process failed, process information will be omitted: %v", a.command, processErr)
-	} else if err := applyAMDProcesses(info, processData); err != nil {
-		global.LOG.Warnf("parsing %s process output failed, process information will be omitted: %v", a.command, err)
+		info.Warnings = append(info.Warnings, fmt.Sprintf("%s processes: %v", a.command, processErr))
 	}
+
 	return info, nil
 }
 
@@ -96,6 +112,9 @@ func parseAMDStatic(data string) (*Info, error) {
 		}
 		device := Device{
 			Type:             "amd",
+			UUID:             amdStringAt(row, "uuid", "asic.uuid"),
+			DriverVersion:    amdStringAt(row, "driver.version", "driver_version", "amdgpu_version"),
+			ProcessStatus:    "unavailable",
 			Index:            index,
 			ProductName:      amdStringAt(row, "asic.market_name", "market_name", "gpu_name"),
 			PersistenceMode:  "N/A",
@@ -111,6 +130,7 @@ func parseAMDStatic(data string) (*Info, error) {
 				"limit.max_power_limit",
 				"limit.max_power",
 			),
+			PowerLimit:  amdMetricAt(row, "W", "limit.ppt0.socket_power_limit", "limit.socket_power_limit"),
 			MemUsed:     "N/A",
 			MemTotal:    "N/A",
 			GPUUtil:     "N/A",
@@ -150,11 +170,31 @@ func applyAMDMetrics(info *Info, data string) error {
 			continue
 		}
 		setAMDMetric(&device.GPUUtil, row, "%", "usage.gfx_activity", "usage.gfx", "gfx_activity", "gfx_usage")
-		setAMDMetric(&device.Temperature, row, "°C", "temperature.hotspot", "temperature.edge", "hotspot_temperature", "gpu_temperature", "gpu_temp")
+		setAMDMetric(&device.Temperature, row, "°C", "temperature.edge", "gpu_temperature", "gpu_temp")
 		setAMDMetric(&device.PowerDraw, row, "W", "power.socket_power", "socket_power", "power_usage")
 		setAMDMetric(&device.MemUsed, row, "MB", "mem_usage.used_vram", "vram.used", "used_vram", "vram_used")
 		setAMDMetric(&device.MemTotal, row, "MB", "mem_usage.total_vram", "vram.total", "total_vram", "vram_total")
-		setAMDMetric(&device.FanSpeed, row, "%", "fan.speed", "fan_speed")
+
+		setAMDMetric(&device.FanSpeed, row, "%", "fan.usage")
+		if amdMetricAt(row, "%", "fan.usage") == "" {
+			speed, speedErr := strconv.ParseFloat(amdStringAt(row, "fan.speed"), 64)
+			maximum, maxErr := strconv.ParseFloat(amdStringAt(row, "fan.max"), 64)
+			if speedErr == nil && maxErr == nil && !math.IsNaN(speed) && !math.IsInf(speed, 0) && maximum > 0 && !math.IsInf(maximum, 0) && speed >= 0 && speed <= maximum {
+				device.FanSpeed = fmt.Sprintf("%.2f %%", speed/maximum*100)
+			}
+		}
+		setAMDMetric(&device.FanRPM, row, "RPM", "fan.rpm")
+		setAMDMetric(&device.HotspotTemperature, row, "°C", "temperature.hotspot", "hotspot_temperature")
+		setAMDMetric(&device.MemoryTemperature, row, "°C", "temperature.mem")
+		setAMDMetric(&device.MemoryActivity, row, "%", "usage.umc_activity")
+		setAMDMetric(&device.MediaUtil, row, "%", "usage.mm_activity")
+		setAMDMetric(&device.Frequency, row, "MHz", "clock.gfx_0.clk")
+		setAMDMetric(&device.MemoryFrequency, row, "MHz", "clock.mem_0.clk")
+		correctable := amdStringAt(row, "ecc.total_correctable_count", "ecc.correctable_count")
+		uncorrectable := amdStringAt(row, "ecc.total_uncorrectable_count", "ecc.uncorrectable_count")
+		if correctable != "" || uncorrectable != "" {
+			device.ECCErrors = []ECCError{{Scope: "Total", Correctable: correctable, Uncorrectable: uncorrectable}}
+		}
 		if value := amdStringAt(row, "perf_level", "performance_level"); value != "" {
 			device.PerformanceState = value
 		}
@@ -177,6 +217,7 @@ func applyAMDProcesses(info *Info, data string) error {
 		if !ok {
 			continue
 		}
+		device.ProcessStatus = "ok"
 		processList, _ := amdValueAt(row, "process_list")
 		items := amdObjectList(processList)
 		if len(items) == 0 {

agent/utils/ai_tools/gpu/client.go

@@ -64,6 +64,7 @@ func (c Client) LoadInfoContext(ctx context.Context) (*Info, error) {
 		if result.info == nil {
 			continue
 		}
+		merged.Warnings = append(merged.Warnings, result.info.Warnings...)
 		if result.info.Type != "" {
 			types = append(types, result.info.Type)
 		}
@@ -86,6 +87,9 @@ func (c Client) LoadInfoContext(ctx context.Context) (*Info, error) {
 	if len(merged.Devices) == 0 && len(errs) > 0 {
 		return nil, fmt.Errorf("calling GPU monitoring tools failed: %w", errors.Join(errs...))
 	}
+	for _, err := range errs {
+		merged.Warnings = append(merged.Warnings, err.Error())
+	}
 	return merged, nil
 }
 

agent/utils/ai_tools/gpu/nvidia.go

@@ -1,16 +1,12 @@
 package gpu
 
 import (
-	"bytes"
 	"context"
 	"encoding/xml"
-	"errors"
 	"fmt"
-	"io"
 	"strings"
 	"time"
 
-	"github.com/1Panel-dev/1Panel/agent/global"
 	"github.com/1Panel-dev/1Panel/agent/utils/cmd"
 )
 
@@ -24,41 +20,10 @@ func (n nvidiaSMI) LoadInfo(ctx context.Context) (*Info, error) {
 	if err != nil {
 		return nil, fmt.Errorf("calling %s failed: %w", nvidiaSMICommand, err)
 	}
-	data := []byte(itemData)
-	version := "v11"
-
-	buf := bytes.NewBuffer(data)
-	decoder := xml.NewDecoder(buf)
-	for {
-		token, err := decoder.Token()
-		if err != nil {
-			if errors.Is(err, io.EOF) {
-				break
-			}
-			return nil, fmt.Errorf("reading token failed: %w", err)
-		}
-		d, ok := token.(xml.Directive)
-		if !ok {
-			continue
-		}
-		directive := string(d)
-		if !strings.HasPrefix(directive, "DOCTYPE") {
-			continue
-		}
-		parts := strings.Split(directive, " ")
-		s := strings.Trim(parts[len(parts)-1], "\" ")
-		if strings.HasPrefix(s, "nvsmi_device_") && strings.HasSuffix(s, ".dtd") {
-			version = strings.TrimSuffix(strings.TrimPrefix(s, "nvsmi_device_"), ".dtd")
-		} else {
-			global.LOG.Debugf("Cannot find schema version in %q", directive)
-		}
-		break
-	}
-
-	return parseNvidiaSMI(data, version)
+	return parseNvidiaSMI([]byte(itemData))
 }
 
-func parseNvidiaSMI(buf []byte, version string) (*Info, error) {
+func parseNvidiaSMI(buf []byte) (*Info, error) {
 	var (
 		s    nvidiaSMIResponse
 		info Info
@@ -72,42 +37,84 @@ func parseNvidiaSMI(buf []byte, version string) (*Info, error) {
 	info.DriverVersion = s.DriverVersion
 	for i := range s.Gpu {
 		gpuItem := Device{
-			Type:             "nvidia",
-			Index:            uint(i),
-			ProductName:      s.Gpu[i].ProductName,
-			PersistenceMode:  s.Gpu[i].PersistenceMode,
-			BusID:            s.Gpu[i].ID,
-			DisplayActive:    s.Gpu[i].DisplayActive,
-			ECC:              s.Gpu[i].EccErrors.Volatile.DramUncorrectable,
-			FanSpeed:         s.Gpu[i].FanSpeed,
-			Temperature:      s.Gpu[i].Temperature.GpuTemp,
-			PerformanceState: s.Gpu[i].PerformanceState,
-			MemUsed:          s.Gpu[i].FbMemoryUsage.Used,
-			MemTotal:         s.Gpu[i].FbMemoryUsage.Total,
-			GPUUtil:          s.Gpu[i].Utilization.GpuUtil,
-			ComputeMode:      s.Gpu[i].ComputeMode,
-			MigMode:          s.Gpu[i].MigMode.CurrentMig,
-		}
-		if version == "v12" || version == "v13" {
-			gpuItem.PowerDraw = s.Gpu[i].GpuPowerReadings.PowerDraw
-			if gpuItem.PowerDraw == "" {
-				gpuItem.PowerDraw = s.Gpu[i].GpuPowerReadings.InstantPowerDraw
-			}
-			gpuItem.MaxPowerLimit = s.Gpu[i].GpuPowerReadings.CurrentPowerLimit
-		} else {
-			gpuItem.PowerDraw = s.Gpu[i].PowerReadings.PowerDraw
-			gpuItem.MaxPowerLimit = s.Gpu[i].PowerReadings.MaxPowerLimit
+			Type:              "nvidia",
+			MemoryActivity:    s.Gpu[i].Utilization.MemoryUtil,
+			EncoderUtil:       s.Gpu[i].Utilization.EncoderUtil,
+			DecoderUtil:       s.Gpu[i].Utilization.DecoderUtil,
+			JPEGUtil:          s.Gpu[i].Utilization.JpegUtil,
+			OFAUtil:           s.Gpu[i].Utilization.OfaUtil,
+			MediaFrequency:    s.Gpu[i].Clocks.VideoClock,
+			UUID:              s.Gpu[i].UUID,
+			DriverVersion:     s.DriverVersion,
+			Architecture:      s.Gpu[i].ProductArchitecture,
+			Frequency:         s.Gpu[i].Clocks.GraphicsClock,
+			MemoryFrequency:   s.Gpu[i].Clocks.MemClock,
+			MemoryTemperature: s.Gpu[i].Temperature.MemoryTemp,
+			MemoryFree:        s.Gpu[i].FbMemoryUsage.Free,
+			MemoryReserved:    s.Gpu[i].FbMemoryUsage.Reserved,
+			PCIeGeneration:    firstSMIValue(s.Gpu[i].Pci.PciGpuLinkInfo.PcieGen.DeviceCurrentLinkGen, s.Gpu[i].Pci.PciGpuLinkInfo.PcieGen.CurrentLinkGen),
+			PCIeMaxGeneration: firstSMIValue(s.Gpu[i].Pci.PciGpuLinkInfo.PcieGen.MaxDeviceLinkGen, s.Gpu[i].Pci.PciGpuLinkInfo.PcieGen.MaxLinkGen),
+			PCIeWidth:         s.Gpu[i].Pci.PciGpuLinkInfo.LinkWidths.CurrentLinkWidth,
+			PCIeMaxWidth:      s.Gpu[i].Pci.PciGpuLinkInfo.LinkWidths.MaxLinkWidth,
+			ProcessStatus:     "unavailable",
+			Index:             uint(i),
+			ProductName:       s.Gpu[i].ProductName,
+			PersistenceMode:   s.Gpu[i].PersistenceMode,
+			BusID:             s.Gpu[i].ID,
+			DisplayActive:     s.Gpu[i].DisplayActive,
+			ECC:               s.Gpu[i].EccMode.CurrentEcc,
+			FanSpeed:          s.Gpu[i].FanSpeed,
+			Temperature:       s.Gpu[i].Temperature.GpuTemp,
+			PerformanceState:  s.Gpu[i].PerformanceState,
+			MemUsed:           s.Gpu[i].FbMemoryUsage.Used,
+			MemTotal:          s.Gpu[i].FbMemoryUsage.Total,
+			GPUUtil:           s.Gpu[i].Utilization.GpuUtil,
+			ComputeMode:       s.Gpu[i].ComputeMode,
+			MigMode:           s.Gpu[i].MigMode.CurrentMig,
 		}
+		gpuItem.ECCPending = s.Gpu[i].EccMode.PendingEcc
+		gpuItem.ECCErrors = []ECCError{
+			{Scope: "Volatile Total", Correctable: s.Gpu[i].EccErrors.Volatile.SingleBit.Total, Uncorrectable: s.Gpu[i].EccErrors.Volatile.DoubleBit.Total},
+			{Scope: "Aggregate Total", Correctable: s.Gpu[i].EccErrors.Aggregate.SingleBit.Total, Uncorrectable: s.Gpu[i].EccErrors.Aggregate.DoubleBit.Total},
 
-		for _, process := range s.Gpu[i].Processes.ProcessInfo {
-			gpuItem.Processes = append(gpuItem.Processes, Process{
-				PID:         process.Pid,
-				Type:        process.Type,
-				ProcessName: process.ProcessName,
-				UsedMemory:  process.UsedMemory,
-			})
+			{Scope: "Volatile DRAM", Correctable: s.Gpu[i].EccErrors.Volatile.DramCorrectable, Uncorrectable: s.Gpu[i].EccErrors.Volatile.DramUncorrectable},
+			{Scope: "Volatile SRAM", Correctable: s.Gpu[i].EccErrors.Volatile.SramCorrectable, Uncorrectable: s.Gpu[i].EccErrors.Volatile.SramUncorrectable},
+			{Scope: "Aggregate DRAM", Correctable: s.Gpu[i].EccErrors.Aggregate.DramCorrectable, Uncorrectable: s.Gpu[i].EccErrors.Aggregate.DramUncorrectable},
+			{Scope: "Aggregate SRAM", Correctable: s.Gpu[i].EccErrors.Aggregate.SramCorrectable, Uncorrectable: s.Gpu[i].EccErrors.Aggregate.SramUncorrectable},
+		}
+		gpuItem.PowerDraw = firstSMIValue(s.Gpu[i].GpuPowerReadings.PowerDraw, s.Gpu[i].GpuPowerReadings.InstantPowerDraw, s.Gpu[i].PowerReadings.PowerDraw)
+		gpuItem.PowerLimit = firstSMIValue(s.Gpu[i].GpuPowerReadings.CurrentPowerLimit, s.Gpu[i].PowerReadings.EnforcedPowerLimit, s.Gpu[i].PowerReadings.PowerLimit)
+		gpuItem.MaxPowerLimit = firstSMIValue(s.Gpu[i].GpuPowerReadings.MaxPowerLimit, s.Gpu[i].PowerReadings.MaxPowerLimit)
+		gpuItem.DefaultPowerLimit = firstSMIValue(s.Gpu[i].GpuPowerReadings.DefaultPowerLimit, s.Gpu[i].PowerReadings.DefaultPowerLimit)
+		for _, event := range append(s.Gpu[i].ClocksEventReasons.Reasons, s.Gpu[i].ClocksThrottleReasons.Reasons...) {
+			if strings.EqualFold(strings.TrimSpace(event.Value), "Active") {
+				gpuItem.ClockEvents = append(gpuItem.ClockEvents, strings.TrimPrefix(strings.TrimPrefix(event.XMLName.Local, "clocks_event_reason_"), "clocks_throttle_reason_"))
+			}
+		}
+		if s.Gpu[i].Processes != nil && strings.TrimSpace(s.Gpu[i].Processes.Text) == "" {
+			gpuItem.ProcessStatus = "ok"
+		}
+		if s.Gpu[i].Processes != nil {
+			for _, process := range s.Gpu[i].Processes.ProcessInfo {
+				gpuItem.Processes = append(gpuItem.Processes, Process{
+					PID:         process.Pid,
+					Type:        process.Type,
+					ProcessName: process.ProcessName,
+					UsedMemory:  process.UsedMemory,
+				})
+			}
 		}
 		info.Devices = append(info.Devices, gpuItem)
 	}
 	return &info, nil
 }
+
+func firstSMIValue(values ...string) string {
+	for _, value := range values {
+		value = strings.TrimSpace(value)
+		if value != "" && !strings.EqualFold(value, "N/A") && !strings.EqualFold(value, "Not Supported") {
+			return value
+		}
+	}
+	return ""
+}

agent/utils/ai_tools/gpu/nvidia_types.go

@@ -1,9 +1,12 @@
 package gpu
 
+import "encoding/xml"
+
 type nvidiaSMIResponse struct {
-	AttachedGpus  string `xml:"attached_gpus"`
-	CudaVersion   string `xml:"cuda_version"`
-	DriverVersion string `xml:"driver_version"`
+	XMLName       xml.Name `xml:"nvidia_smi_log"`
+	AttachedGpus  string   `xml:"attached_gpus"`
+	CudaVersion   string   `xml:"cuda_version"`
+	DriverVersion string   `xml:"driver_version"`
 	Gpu           []struct {
 		ID                       string   `xml:"id,attr"`
 		AccountedProcesses       struct{} `xml:"accounted_processes"`
@@ -37,16 +40,17 @@ type nvidiaSMIResponse struct {
 			VideoClock    string `xml:"video_clock"`
 		} `xml:"clocks"`
 		ClocksEventReasons struct {
-			ClocksEventReasonApplicationsClocksSetting string `xml:"clocks_event_reason_applications_clocks_setting"`
-			ClocksEventReasonDisplayClocksSetting      string `xml:"clocks_event_reason_display_clocks_setting"`
-			ClocksEventReasonGpuIdle                   string `xml:"clocks_event_reason_gpu_idle"`
-			ClocksEventReasonHwPowerBrakeSlowdown      string `xml:"clocks_event_reason_hw_power_brake_slowdown"`
-			ClocksEventReasonHwSlowdown                string `xml:"clocks_event_reason_hw_slowdown"`
-			ClocksEventReasonHwThermalSlowdown         string `xml:"clocks_event_reason_hw_thermal_slowdown"`
-			ClocksEventReasonSwPowerCap                string `xml:"clocks_event_reason_sw_power_cap"`
-			ClocksEventReasonSwThermalSlowdown         string `xml:"clocks_event_reason_sw_thermal_slowdown"`
-			ClocksEventReasonSyncBoost                 string `xml:"clocks_event_reason_sync_boost"`
+			Reasons []struct {
+				XMLName xml.Name
+				Value   string `xml:",chardata"`
+			} `xml:",any"`
 		} `xml:"clocks_event_reasons"`
+		ClocksThrottleReasons struct {
+			Reasons []struct {
+				XMLName xml.Name
+				Value   string `xml:",chardata"`
+			} `xml:",any"`
+		} `xml:"clocks_throttle_reasons"`
 		ComputeMode               string `xml:"compute_mode"`
 		DefaultApplicationsClocks struct {
 			GraphicsClock string `xml:"graphics_clock"`
@@ -63,12 +67,24 @@ type nvidiaSMIResponse struct {
 		} `xml:"driver_model"`
 		EccErrors struct {
 			Aggregate struct {
+				SingleBit struct {
+					Total string `xml:"total"`
+				} `xml:"single_bit"`
+				DoubleBit struct {
+					Total string `xml:"total"`
+				} `xml:"double_bit"`
 				DramCorrectable   string `xml:"dram_correctable"`
 				DramUncorrectable string `xml:"dram_uncorrectable"`
 				SramCorrectable   string `xml:"sram_correctable"`
 				SramUncorrectable string `xml:"sram_uncorrectable"`
 			} `xml:"aggregate"`
 			Volatile struct {
+				SingleBit struct {
+					Total string `xml:"total"`
+				} `xml:"single_bit"`
+				DoubleBit struct {
+					Total string `xml:"total"`
+				} `xml:"double_bit"`
 				DramCorrectable   string `xml:"dram_correctable"`
 				DramUncorrectable string `xml:"dram_uncorrectable"`
 				SramCorrectable   string `xml:"sram_correctable"`
@@ -226,7 +242,8 @@ type nvidiaSMIResponse struct {
 			MinPowerLimit      string `xml:"min_power_limit"`
 			MaxPowerLimit      string `xml:"max_power_limit"`
 		} `xml:"power_readings"`
-		Processes struct {
+		Processes *struct {
+			Text        string `xml:",chardata"`
 			ProcessInfo []struct {
 				Pid         string `xml:"pid"`
 				Type        string `xml:"type"`

agent/utils/ai_tools/gpu/types.go

@@ -1,6 +1,8 @@
 package gpu
 
 type Info struct {
+	Warnings []string `json:"warnings"`
+
 	CudaVersion   string `json:"cudaVersion"`
 	DriverVersion string `json:"driverVersion"`
 	Type          string `json:"type"`
@@ -8,7 +10,42 @@ type Info struct {
 	Devices []Device `json:"gpu"`
 }
 
+type ECCError struct {
+	Scope         string `json:"scope"`
+	Correctable   string `json:"correctable"`
+	Uncorrectable string `json:"uncorrectable"`
+}
+
 type Device struct {
+	ECCPending         string     `json:"eccPending"`
+	ECCErrors          []ECCError `json:"eccErrors"`
+	MemoryActivity     string     `json:"memoryActivity"`
+	EncoderUtil        string     `json:"encoderUtil"`
+	DecoderUtil        string     `json:"decoderUtil"`
+	JPEGUtil           string     `json:"jpegUtil"`
+	OFAUtil            string     `json:"ofaUtil"`
+	MediaUtil          string     `json:"mediaUtil"`
+	HotspotTemperature string     `json:"hotspotTemperature"`
+	FanRPM             string     `json:"fanRPM"`
+	MediaFrequency     string     `json:"mediaFrequency"`
+
+	UUID              string   `json:"uuid"`
+	DriverVersion     string   `json:"driverVersion"`
+	Architecture      string   `json:"architecture"`
+	Frequency         string   `json:"frequency"`
+	MemoryFrequency   string   `json:"memoryFrequency"`
+	MemoryTemperature string   `json:"memoryTemperature"`
+	MemoryFree        string   `json:"memoryFree"`
+	MemoryReserved    string   `json:"memoryReserved"`
+	PowerLimit        string   `json:"powerLimit"`
+	DefaultPowerLimit string   `json:"defaultPowerLimit"`
+	PCIeGeneration    string   `json:"pcieGeneration"`
+	PCIeMaxGeneration string   `json:"pcieMaxGeneration"`
+	PCIeWidth         string   `json:"pcieWidth"`
+	PCIeMaxWidth      string   `json:"pcieMaxWidth"`
+	ProcessStatus     string   `json:"processStatus"`
+	ClockEvents       []string `json:"clockEvents"`
+
 	Type            string `json:"type"`
 	Index           uint   `json:"index"`
 	ProductName     string `json:"productName"`

agent/utils/ai_tools/npu/npu.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"strconv"
 	"strings"
+	"sync"
 	"time"
 
 	"github.com/1Panel-dev/1Panel/agent/utils/cmd"
@@ -34,7 +35,31 @@ func (c Client) LoadInfoContext(ctx context.Context) (*Info, error) {
 	if err != nil {
 		return nil, fmt.Errorf("calling %s failed: %w", ascendSMICommand, err)
 	}
-	return parseAscendSMI(itemData), nil
+
+	info := parseAscendSMI(itemData)
+	var wg sync.WaitGroup
+	warnings := make([]string, len(info.Devices))
+	for i := range info.Devices {
+		wg.Add(1)
+		go func(index int) {
+			defer wg.Done()
+			device := &info.Devices[index]
+			mgr := cmd.NewCommandMgr(cmd.WithContext(ctx), cmd.WithTimeout(5*time.Second))
+			data, err := mgr.RunWithStdout(ascendSMICommand, "info", "-t", "usages", "-i", strconv.FormatUint(uint64(device.NPUIndex), 10), "-c", strconv.FormatUint(uint64(device.ChipIndex), 10))
+			if err != nil {
+				warnings[index] = fmt.Sprintf("npu-smi usages %d/%d: %v", device.NPUIndex, device.ChipIndex, err)
+				return
+			}
+			applyAscendUsages(device, data)
+		}(i)
+	}
+	wg.Wait()
+	for _, warning := range warnings {
+		if warning != "" {
+			info.Warnings = append(info.Warnings, warning)
+		}
+	}
+	return info, nil
 }
 
 func parseAscendSMI(data string) *Info {
@@ -158,6 +183,12 @@ func parseAscendSMI(data string) *Info {
 		pending = nil
 	}
 
+	for i := range info.Devices {
+		info.Devices[i].ProcessStatus = "unavailable"
+		if processSection {
+			info.Devices[i].ProcessStatus = "ok"
+		}
+	}
 	return info
 }
 
@@ -208,6 +239,9 @@ func ascendMemoryPools(value, header string) (string, string, string, string) {
 	}
 
 	hbm := usage[len(usage)-1]
+	if !strings.Contains(normalizedHeader, "MEMORYUSAGE") {
+		memoryUsed, memoryTotal = "", ""
+	}
 	return memoryUsed, memoryTotal, hbm[0] + " MB", hbm[1] + " MB"
 }
 
@@ -233,3 +267,29 @@ func ascendValueWithUnit(value, unit string) string {
 	}
 	return value + " " + unit
 }
+
+func applyAscendUsages(device *Device, data string) {
+	for _, line := range strings.Split(data, "\n") {
+		key, value, ok := strings.Cut(line, ":")
+		if !ok {
+			continue
+		}
+		key = strings.ToLower(strings.Join(strings.Fields(key), ""))
+		value = strings.TrimSpace(value)
+		switch key {
+		case "aicoreusagerate(%)":
+			usage, err := strconv.ParseFloat(strings.TrimSpace(strings.TrimSuffix(value, "%")), 64)
+			if err == nil && usage >= 0 && usage <= 100 {
+				device.AICore = ascendValueWithUnit(value, "%")
+			}
+		case "aicpuusagerate(%)":
+			device.AICPUUtil = ascendValueWithUnit(value, "%")
+		case "ctrlcpuusagerate(%)":
+			device.CtrlCPUUtil = ascendValueWithUnit(value, "%")
+		case "ddrbandwidthusagerate(%)", "memorybandwidthusagerate(%)":
+			device.DDRBandwidth = ascendValueWithUnit(value, "%")
+		case "hbmbandwidthusagerate(%)":
+			device.HBMBandwidth = ascendValueWithUnit(value, "%")
+		}
+	}
+}

agent/utils/ai_tools/npu/types.go

@@ -1,12 +1,20 @@
 package npu
 
 type Info struct {
+	Warnings      []string `json:"warnings"`
 	Type          string   `json:"type"`
 	DriverVersion string   `json:"driverVersion"`
 	Devices       []Device `json:"npu"`
 }
 
 type Device struct {
+	AICPUUtil    string `json:"aiCPUUtil"`
+	CtrlCPUUtil  string `json:"ctrlCPUUtil"`
+	DDRBandwidth string `json:"ddrBandwidth"`
+	HBMBandwidth string `json:"hbmBandwidth"`
+
+	ProcessStatus string `json:"processStatus"`
+
 	Type        string `json:"type"`
 	Index       uint   `json:"index"`
 	NPUIndex    uint   `json:"npuIndex"`

agent/utils/ai_tools/vllm/metrics.go

@@ -0,0 +1,335 @@
+package vllm
+
+import (
+	"bufio"
+	"encoding/json"
+	"fmt"
+	"io"
+	"math"
+	"sort"
+	"strconv"
+	"strings"
+
+	"github.com/1Panel-dev/1Panel/agent/app/model"
+	"github.com/1Panel-dev/1Panel/agent/utils/re"
+)
+
+type Metrics map[string]map[string]float64
+
+type histogram struct {
+	Sum     float64            `json:"sum"`
+	Count   float64            `json:"count"`
+	Buckets map[string]float64 `json:"buckets,omitempty"`
+}
+
+func Parse(reader io.Reader) (Metrics, error) {
+	metrics := Metrics{}
+	scanner := bufio.NewScanner(reader)
+	scanner.Buffer(make([]byte, 4096), 1024*1024)
+	for scanner.Scan() {
+		line := strings.TrimSpace(scanner.Text())
+		if !strings.HasPrefix(line, "vllm:") {
+			continue
+		}
+		match := re.GetRegex(re.VLLMMetricSamplePattern).FindStringSubmatch(line)
+		if match == nil {
+			return nil, fmt.Errorf("invalid vLLM metric sample")
+		}
+		name := strings.TrimPrefix(match[1], "vllm:")
+		switch name {
+		case "num_requests_running", "num_requests_waiting", "kv_cache_usage_perc", "gpu_cache_usage_perc", "prompt_tokens_total", "generation_tokens_total", "request_success_total":
+		default:
+			base := strings.TrimSuffix(strings.TrimSuffix(strings.TrimSuffix(name, "_sum"), "_count"), "_bucket")
+			if base == name {
+				continue
+			}
+			switch base {
+			case "time_to_first_token_seconds", "time_per_output_token_seconds", "inter_token_latency_seconds", "e2e_request_latency_seconds", "request_prefill_time_seconds", "request_decode_time_seconds":
+			default:
+				continue
+			}
+		}
+		value, err := strconv.ParseFloat(match[3], 64)
+		if err != nil {
+			return nil, fmt.Errorf("invalid vLLM metric value: %w", err)
+		}
+		if math.IsNaN(value) || math.IsInf(value, 0) || value < 0 {
+			continue
+		}
+		if metrics[name] == nil {
+			metrics[name] = map[string]float64{}
+		}
+		labels := re.GetRegex(re.VLLMMetricLabelPattern).FindAllStringSubmatch(match[2], -1)
+		parts := make([]string, 0, len(labels))
+		for _, label := range labels {
+			parts = append(parts, label[1]+"="+label[2])
+		}
+		sort.Strings(parts)
+		metrics[name]["{"+strings.Join(parts, ",")+"}"] = value
+	}
+	if err := scanner.Err(); err != nil {
+		return nil, err
+	}
+	if len(metrics) == 0 {
+		return nil, fmt.Errorf("no supported vLLM metrics")
+	}
+	return metrics, nil
+}
+
+func Calculate(current, previous Metrics, elapsed float64) (model.MonitorVLLM, error) {
+	result := model.MonitorVLLM{}
+	for name, target := range map[string]**float64{
+		"num_requests_running": &result.Running,
+		"num_requests_waiting": &result.Waiting,
+	} {
+		if len(current[name]) == 0 {
+			continue
+		}
+		value := 0.0
+		for _, sample := range current[name] {
+			value += sample
+		}
+		*target = &value
+	}
+	cache := current["kv_cache_usage_perc"]
+	if len(cache) == 0 {
+		cache = current["gpu_cache_usage_perc"]
+	}
+	if len(cache) > 0 {
+		value := 0.0
+		for _, sample := range cache {
+			value += sample
+		}
+		value = value / float64(len(cache)) * 100
+		result.CacheUsage = &value
+	}
+	if elapsed <= 0 {
+		return result, nil
+	}
+	for name, target := range map[string]**float64{
+		"prompt_tokens_total":     &result.PromptThroughput,
+		"generation_tokens_total": &result.GenerationThroughput,
+		"request_success_total":   &result.RequestThroughput,
+	} {
+		if value := counterDelta(current[name], previous[name]); value != nil {
+			rate := *value / elapsed
+			*target = &rate
+		}
+	}
+	histograms := map[string]histogram{}
+	for _, name := range []string{"time_to_first_token_seconds", "inter_token_latency_seconds", "e2e_request_latency_seconds", "request_prefill_time_seconds", "request_decode_time_seconds"} {
+		source := name
+		if name == "inter_token_latency_seconds" && len(current[name+"_count"]) == 0 {
+			source = "time_per_output_token_seconds"
+		}
+		sum := counterDelta(current[source+"_sum"], previous[source+"_sum"])
+		count := counterDelta(current[source+"_count"], previous[source+"_count"])
+		if sum == nil || count == nil {
+			continue
+		}
+		histograms[name] = histogram{Sum: *sum, Count: *count, Buckets: bucketDeltas(current[source+"_bucket"], previous[source+"_bucket"], *count)}
+	}
+	applyHistograms(&result, histograms)
+	encoded, err := json.Marshal(histograms)
+	if err != nil {
+		return model.MonitorVLLM{}, err
+	}
+	result.HistogramDeltas = string(encoded)
+	return result, nil
+}
+
+func AggregateHistograms(encoded string, result *model.MonitorVLLM) error {
+	var samples []map[string]histogram
+	if err := json.Unmarshal([]byte(encoded), &samples); err != nil {
+		return err
+	}
+	merged := map[string]histogram{}
+	for _, sample := range samples {
+		for name, item := range sample {
+			if item.Count == 0 {
+				continue
+			}
+			total, exists := merged[name]
+			if !exists {
+				merged[name] = item
+				continue
+			}
+			total.Sum += item.Sum
+			total.Count += item.Count
+			if len(total.Buckets) != len(item.Buckets) {
+				total.Buckets = nil
+			}
+			for bound, count := range total.Buckets {
+				addition, ok := item.Buckets[bound]
+				if !ok {
+					total.Buckets = nil
+					break
+				}
+				total.Buckets[bound] = count + addition
+			}
+			merged[name] = total
+		}
+	}
+	applyHistograms(result, merged)
+	return nil
+}
+
+func applyHistograms(result *model.MonitorVLLM, histograms map[string]histogram) {
+	for _, item := range []struct {
+		name      string
+		mean      **float64
+		quantiles []**float64
+	}{
+		{"time_to_first_token_seconds", &result.TimeToFirstToken, []**float64{&result.TimeToFirstTokenP50, &result.TimeToFirstTokenP90, &result.TimeToFirstTokenP95, &result.TimeToFirstTokenP99}},
+		{"inter_token_latency_seconds", &result.TimePerOutputToken, []**float64{&result.TimePerOutputTokenP50, &result.TimePerOutputTokenP90, &result.TimePerOutputTokenP95, &result.TimePerOutputTokenP99}},
+		{"e2e_request_latency_seconds", &result.RequestLatency, []**float64{&result.RequestLatencyP50, &result.RequestLatencyP90, &result.RequestLatencyP95, &result.RequestLatencyP99}},
+		{"request_prefill_time_seconds", &result.PrefillTime, nil},
+		{"request_decode_time_seconds", &result.DecodeTime, nil},
+	} {
+		histogram, ok := histograms[item.name]
+		if !ok || histogram.Count <= 0 {
+			continue
+		}
+		mean := histogram.Sum / histogram.Count
+		*item.mean = &mean
+		if len(item.quantiles) == 0 {
+			continue
+		}
+		quantiles := histogramQuantiles(histogram.Buckets)
+		for i, target := range item.quantiles {
+			*target = quantiles[i]
+		}
+	}
+}
+
+func counterDelta(current, previous map[string]float64) *float64 {
+	if len(current) == 0 || len(current) != len(previous) {
+		return nil
+	}
+	value := 0.0
+	for label, sample := range current {
+		before, ok := previous[label]
+		if !ok || sample < before {
+			return nil
+		}
+		value += sample - before
+	}
+	if math.IsInf(value, 0) || math.IsNaN(value) {
+		return nil
+	}
+	return &value
+}
+
+func bucketDeltas(current, previous map[string]float64, count float64) map[string]float64 {
+	if len(current) == 0 || len(current) != len(previous) {
+		return nil
+	}
+	groups := map[string]map[string]float64{}
+	for labels, sample := range current {
+		before, ok := previous[labels]
+		if !ok || sample < before {
+			return nil
+		}
+		bound := ""
+		var identity []string
+		for _, label := range re.GetRegex(re.VLLMMetricLabelPattern).FindAllStringSubmatch(labels, -1) {
+			if label[1] == "le" {
+				raw, err := strconv.Unquote(label[2])
+				if err != nil {
+					return nil
+				}
+				upper, err := strconv.ParseFloat(raw, 64)
+				if err != nil || math.IsNaN(upper) || upper < 0 {
+					return nil
+				}
+				bound = strconv.FormatFloat(upper, 'g', -1, 64)
+			} else {
+				identity = append(identity, label[1]+"="+label[2])
+			}
+		}
+		if bound == "" {
+			return nil
+		}
+		group := strings.Join(identity, ",")
+		if groups[group] == nil {
+			groups[group] = map[string]float64{}
+		}
+		if _, exists := groups[group][bound]; exists {
+			return nil
+		}
+		groups[group][bound] = sample - before
+	}
+	result := map[string]float64{}
+	for _, buckets := range groups {
+		if _, ok := buckets["+Inf"]; !ok || len(buckets) < 2 {
+			return nil
+		}
+		if len(result) > 0 && len(result) != len(buckets) {
+			return nil
+		}
+		for bound := range result {
+			if _, ok := buckets[bound]; !ok {
+				return nil
+			}
+		}
+		bounds := make([]float64, 0, len(buckets))
+		for bound := range buckets {
+			value, _ := strconv.ParseFloat(bound, 64)
+			bounds = append(bounds, value)
+		}
+		sort.Float64s(bounds)
+		previousCount := 0.0
+		for _, bound := range bounds {
+			key := strconv.FormatFloat(bound, 'g', -1, 64)
+			value := buckets[key]
+			if value < previousCount {
+				if previousCount-value > 1e-12*(previousCount+value) {
+					return nil
+				}
+				value = previousCount
+			}
+			result[key] += value
+			previousCount = value
+		}
+	}
+	if math.Abs(result["+Inf"]-count) > 1e-12*(result["+Inf"]+count) {
+		return nil
+	}
+	return result
+}
+
+func histogramQuantiles(buckets map[string]float64) [4]*float64 {
+	var quantiles [4]*float64
+	if len(buckets) < 2 || buckets["+Inf"] <= 0 {
+		return quantiles
+	}
+	bounds := make([]float64, 0, len(buckets)-1)
+	for bound := range buckets {
+		value, err := strconv.ParseFloat(bound, 64)
+		if err != nil || math.IsNaN(value) || value < 0 {
+			return quantiles
+		}
+		if !math.IsInf(value, 1) {
+			bounds = append(bounds, value)
+		}
+	}
+	if len(bounds) == 0 {
+		return quantiles
+	}
+	sort.Float64s(bounds)
+	for i, q := range [...]float64{0.5, 0.9, 0.95, 0.99} {
+		rank := q * buckets["+Inf"]
+		lower, before := 0.0, 0.0
+		value := bounds[len(bounds)-1]
+		for _, upper := range bounds {
+			count := buckets[strconv.FormatFloat(upper, 'g', -1, 64)]
+			if count >= rank && count > before {
+				value = lower + (upper-lower)*(rank-before)/(count-before)
+				break
+			}
+			lower, before = upper, count
+		}
+		quantiles[i] = &value
+	}
+	return quantiles
+}

agent/utils/ai_tools/xpu/types.go

@@ -33,16 +33,29 @@ type discoveryInfo struct {
 }
 
 type DeviceLevelMetric struct {
-	MetricsType string  `json:"metrics_type"`
-	Value       float64 `json:"value"`
+	MetricsType string   `json:"metrics_type"`
+	Value       *float64 `json:"value"`
+}
+
+type TileMetrics struct {
+	TileID   int                 `json:"tile_id"`
+	DataList []DeviceLevelMetric `json:"data_list"`
+}
+
+type TileStats struct {
+	TileID int   `json:"tileID"`
+	Stats  Stats `json:"stats"`
 }
 
 type DeviceStats struct {
+	TileLevel   []TileMetrics       `json:"tile_level"`
 	DeviceID    int                 `json:"device_id"`
 	DeviceLevel []DeviceLevelMetric `json:"device_level"`
 }
 
 type Info struct {
+	Warnings []string `json:"warnings"`
+
 	Type          string `json:"type"`
 	DriverVersion string `json:"driverVersion"`
 
@@ -50,12 +63,17 @@ type Info struct {
 }
 
 type Device struct {
+	Tiles         []TileStats `json:"tiles"`
+	ProcessStatus string      `json:"processStatus"`
+
 	Basic     Basic     `json:"basic"`
 	Stats     Stats     `json:"stats"`
 	Processes []Process `json:"processes"`
 }
 
 type Basic struct {
+	UUID string `json:"uuid"`
+
 	DeviceID      int    `json:"deviceID"`
 	DeviceName    string `json:"deviceName"`
 	VendorName    string `json:"vendorName"`
@@ -66,6 +84,14 @@ type Basic struct {
 }
 
 type Stats struct {
+	MediaUtil         string `json:"mediaUtil"`
+	ComputeUtil       string `json:"computeUtil"`
+	CopyUtil          string `json:"copyUtil"`
+	MediaFrequency    string `json:"mediaFrequency"`
+	MemoryTemperature string `json:"memoryTemperature"`
+
+	MemoryBandwidthUtil string `json:"memoryBandwidthUtil"`
+
 	Power       string `json:"power"`
 	GPUUtil     string `json:"gpuUtil"`
 	Frequency   string `json:"frequency"`

agent/utils/ai_tools/xpu/xpu.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
+	"math"
 	"sort"
 	"strconv"
 	"sync"
@@ -51,12 +52,15 @@ func (c Client) LoadInfoContext(ctx context.Context) (*Info, error) {
 
 	processData, err := cmdMgr.RunWithStdout(xpuSMICommand, "ps", "-j")
 	if err != nil {
-		global.LOG.Warnf("calling xpu-smi ps failed, process information will be omitted: %v", err)
+		res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi ps: %v", err))
 	} else {
 		var psList DeviceUtilByProcList
 		if err := json.Unmarshal([]byte(processData), &psList); err != nil {
-			global.LOG.Warnf("processData json unmarshal failed, process information will be omitted: %v", err)
+			res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi ps output: %v", err))
 		} else {
+			for i := range res.Devices {
+				res.Devices[i].ProcessStatus = "ok"
+			}
 			for _, ps := range psList.DeviceUtilByProcList {
 				process := Process{
 					PID:     ps.ProcessID,
@@ -87,7 +91,9 @@ func (c Client) loadDeviceInfo(ctx context.Context, device discoveryDevice, wg *
 	defer wg.Done()
 
 	xpu := Device{
+		ProcessStatus: "unavailable",
 		Basic: Basic{
+			UUID:          device.UUID,
 			DeviceID:      device.DeviceID,
 			DeviceName:    device.DeviceName,
 			VendorName:    device.VendorName,
@@ -115,13 +121,19 @@ func (c Client) loadDeviceInfo(ctx context.Context, device discoveryDevice, wg *
 	wgCmd.Wait()
 
 	if xpuErr != nil {
-		global.LOG.Errorf("calling xpu-smi discovery failed for device %d, %v", device.DeviceID, xpuErr)
+		mu.Lock()
+		res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi device %d: %v", device.DeviceID, xpuErr))
+		res.Devices = append(res.Devices, xpu)
+		mu.Unlock()
 		return
 	}
 
 	var info discoveryDevice
 	if err := json.Unmarshal([]byte(xpuData), &info); err != nil {
-		global.LOG.Errorf("xpuData json unmarshal failed for device %d, err: %v", device.DeviceID, err)
+		mu.Lock()
+		res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi device %d output: %v", device.DeviceID, err))
+		res.Devices = append(res.Devices, xpu)
+		mu.Unlock()
 		return
 	}
 
@@ -142,13 +154,22 @@ func (c Client) loadDeviceInfo(ctx context.Context, device discoveryDevice, wg *
 	}
 
 	if statsErr != nil {
-		global.LOG.Warnf("calling xpu-smi stats failed for device %d, metrics will be omitted: %v", device.DeviceID, statsErr)
+		mu.Lock()
+		res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi stats %d: %v", device.DeviceID, statsErr))
+		mu.Unlock()
 	} else {
 		var stats DeviceStats
 		if err := json.Unmarshal([]byte(statsData), &stats); err != nil {
-			global.LOG.Warnf("statsData json unmarshal failed for device %d, metrics will be omitted: %v", device.DeviceID, err)
+			mu.Lock()
+			res.Warnings = append(res.Warnings, fmt.Sprintf("xpu-smi stats %d output: %v", device.DeviceID, err))
+			mu.Unlock()
 		} else {
 			loadStats(&xpu.Stats, stats.DeviceLevel)
+			for _, tile := range stats.TileLevel {
+				item := TileStats{TileID: tile.TileID}
+				loadStats(&item.Stats, tile.DataList)
+				xpu.Tiles = append(xpu.Tiles, item)
+			}
 		}
 	}
 
@@ -162,19 +183,38 @@ func (c Client) loadDeviceInfo(ctx context.Context, device discoveryDevice, wg *
 
 func loadStats(stats *Stats, metrics []DeviceLevelMetric) {
 	for _, stat := range metrics {
+		if stat.Value == nil || math.IsNaN(*stat.Value) || math.IsInf(*stat.Value, 0) {
+			continue
+		}
+		value := *stat.Value
+		if value < 0 && stat.MetricsType != "XPUM_STATS_GPU_CORE_TEMPERATURE" && stat.MetricsType != "XPUM_STATS_MEMORY_TEMPERATURE" {
+			continue
+		}
 		switch stat.MetricsType {
+		case "XPUM_STATS_MEMORY_TEMPERATURE":
+			stats.MemoryTemperature = fmt.Sprintf("%.1f°C", value)
+		case "XPUM_STATS_ENGINE_GROUP_COMPUTE_ALL_UTILIZATION":
+			stats.ComputeUtil = fmt.Sprintf("%.1f%%", value)
+		case "XPUM_STATS_ENGINE_GROUP_MEDIA_ALL_UTILIZATION":
+			stats.MediaUtil = fmt.Sprintf("%.1f%%", value)
+		case "XPUM_STATS_ENGINE_GROUP_COPY_ALL_UTILIZATION":
+			stats.CopyUtil = fmt.Sprintf("%.1f%%", value)
+		case "XPUM_STATS_MEDIA_ENGINE_FREQUENCY":
+			stats.MediaFrequency = fmt.Sprintf("%.1fMHz", value)
 		case "XPUM_STATS_POWER":
-			stats.Power = fmt.Sprintf("%.1fW", stat.Value)
+			stats.Power = fmt.Sprintf("%.1fW", value)
 		case "XPUM_STATS_GPU_UTILIZATION":
-			stats.GPUUtil = fmt.Sprintf("%.1f%%", stat.Value)
+			stats.GPUUtil = fmt.Sprintf("%.1f%%", value)
 		case "XPUM_STATS_GPU_FREQUENCY":
-			stats.Frequency = fmt.Sprintf("%.1fMHz", stat.Value)
+			stats.Frequency = fmt.Sprintf("%.1fMHz", value)
 		case "XPUM_STATS_GPU_CORE_TEMPERATURE":
-			stats.Temperature = fmt.Sprintf("%.1f°C", stat.Value)
+			stats.Temperature = fmt.Sprintf("%.1f°C", value)
 		case "XPUM_STATS_MEMORY_USED":
-			stats.MemoryUsed = fmt.Sprintf("%.1f MiB", stat.Value)
-		case "XPUM_STATS_MEMORY_UTILIZATION", "XPUM_STATS_MEMORY_BANDWIDTH", "XPUM_STATS_MEMORY_BANDWIDTH_UTILIZATION":
-			stats.MemoryUtil = fmt.Sprintf("%.1f%%", stat.Value)
+			stats.MemoryUsed = fmt.Sprintf("%.1f MiB", value)
+		case "XPUM_STATS_MEMORY_UTILIZATION":
+			stats.MemoryUtil = fmt.Sprintf("%.1f%%", value)
+		case "XPUM_STATS_MEMORY_BANDWIDTH", "XPUM_STATS_MEMORY_BANDWIDTH_UTILIZATION":
+			stats.MemoryBandwidthUtil = fmt.Sprintf("%.1f%%", value)
 		}
 	}
 }

agent/utils/alert/alert.go

@@ -24,10 +24,10 @@ import (
 	"github.com/jinzhu/copier"
 )
 
-var cronJobAlertTypes = []string{"shell", "app", "website", "database", "directory", "log", "snapshot", "curl", "cutWebsiteLog", "clean", "ntp"}
+var cronJobAlertTypes = []string{"shell", "app", "website", "database", "directory", "log", "snapshot", "curl", "cutWebsiteLog", "clean", "ntp", "syncIpGroup", "cleanLog"}
 
 func CreateTaskScanEmailAlertLog(alert dto.AlertDTO, create dto.AlertLogCreate, pushAlert dto.PushAlert, method string, transport *http.Transport, agentInfo *dto.AgentInfo, emailConfig model.AlertConfig) error {
-	params := CreateAlertParams(GetCronJobTypeName(pushAlert.Param))
+	params := CreateTaskAlertParams(pushAlert)
 	alertDetail := ProcessAlertDetail(alert, pushAlert.TaskName, params, method)
 	alertRule := ProcessAlertRule(alert)
 	create.AlertRule = alertRule
@@ -76,14 +76,14 @@ func CreateEmailAlertLog(create dto.AlertLogCreate, alert dto.AlertDTO, params [
 			Encryption: emailInfo.Encryption,
 			Recipient:  emailInfo.Recipient,
 		}
-		content := GetSendContent(alert.Type, params, agentInfo)
+		content := GetAlertLogContent(create, alert, params, agentInfo)
 		if content == "" {
 			content = i18n.GetMsgWithMap("CommonAlert", map[string]interface{}{"msg": alert.Title})
 		}
 		msg := email.EmailMessage{
 			Subject: i18n.GetMsgByKey("PanelAlertTitle"),
 			Body:    content,
-			IsHTML:  true,
+			IsHTML:  GetCronJobType(alert.Type) != "cronJob",
 		}
 
 		if err = email.SendMail(smtpConfig, msg, transport); err != nil {
@@ -110,7 +110,7 @@ func CreateBarkAlertLog(create dto.AlertLogCreate, alert dto.AlertDTO, params []
 		return SaveAlertLog(create, &alertLog)
 	}
 
-	content := GetSendContent(alert.Type, params, agentInfo)
+	content := GetAlertLogContent(create, alert, params, agentInfo)
 	if content == "" {
 		content = i18n.GetMsgWithMap("CommonAlert", map[string]interface{}{"msg": alert.Title})
 	}
@@ -268,7 +268,7 @@ func ProcessAlertDetail(alert dto.AlertDTO, project string, params []dto.Param,
 	alertDetail := dto.AlertDetail{
 		Type:    GetCronJobType(alert.Type),
 		SubType: alert.Type,
-		Title:   alert.Title,
+		Title:   TaskAlertTitle(alert.Type, alert.Title, project, params),
 		Method:  method,
 		Project: project,
 		Params:  params,
@@ -281,6 +281,14 @@ func ProcessAlertDetail(alert dto.AlertDTO, project string, params []dto.Param,
 	return string(marshal)
 }
 
+func TaskAlertTitle(alertType, title, project string, params []dto.Param) string {
+	if GetCronJobType(alertType) != "cronJob" || CronJobAlertResultFromParams(params) != CronJobAlertSuccess {
+		return title
+	}
+	name := cronJobTaskName(project, params)
+	return i18n.GetMsgWithMap("TaskSuccess", map[string]interface{}{"name": name})
+}
+
 func ProcessAlertRule(alert dto.AlertDTO) string {
 	marshal, err := json.Marshal(alert)
 	if err != nil {
@@ -324,6 +332,10 @@ func GetCronJobTypeName(cronJobType string) string {
 		module = "系统快照"
 	case "ntp":
 		module = "同步服务器时间"
+	case "syncIpGroup":
+		module = "同步 IP 组"
+	case "cleanLog":
+		module = "清理日志"
 	default:
 	}
 	return module
@@ -444,6 +456,30 @@ func isWithinTimeRange(savedTimeString string) bool {
 }
 
 func GetSendContent(alertType string, params []dto.Param, agentInfo *dto.AgentInfo) string {
+	return GetAlertDetailContent(dto.AlertDetail{Type: alertType, Params: params}, agentInfo)
+}
+
+func GetAlertLogContent(create dto.AlertLogCreate, alert dto.AlertDTO, params []dto.Param, agentInfo *dto.AgentInfo) string {
+	detail := dto.AlertDetail{Type: alert.Type, Params: params}
+	var stored dto.AlertDetail
+	if json.Unmarshal([]byte(create.AlertDetail), &stored) == nil {
+		detail = stored
+		if detail.Type == "" {
+			detail.Type = alert.Type
+		}
+		if detail.Params == nil {
+			detail.Params = params
+		}
+	}
+	return GetAlertDetailContent(detail, agentInfo)
+}
+
+func GetAlertDetailContent(detail dto.AlertDetail, agentInfo *dto.AgentInfo) string {
+	alertType := detail.SubType
+	if alertType == "" {
+		alertType = detail.Type
+	}
+	params := detail.Params
 	switch GetCronJobType(alertType) {
 	case "ssl":
 		return i18n.GetMsgWithMap("SSLAlert", map[string]interface{}{"num": getValueByIndex(params, "1"), "day": getValueByIndex(params, "2"), "node": getNodeName(agentInfo), "ip": getNodeIp(agentInfo)})
@@ -464,7 +500,12 @@ func GetSendContent(alertType string, params []dto.Param, agentInfo *dto.AgentIn
 	case "disk":
 		return i18n.GetMsgWithMap("DiskUsedAlert", map[string]interface{}{"name": getValueByIndex(params, "1"), "used": getValueByIndex(params, "2"), "node": getNodeName(agentInfo), "ip": getNodeIp(agentInfo)})
 	case "cronJob":
-		return i18n.GetMsgWithMap("CronJobFailedAlert", map[string]interface{}{"name": getValueByIndex(params, "1"), "node": getNodeName(agentInfo), "ip": getNodeIp(agentInfo)})
+		messageKey := "CronJobFailedAlert"
+		if CronJobAlertResultFromParams(params) == CronJobAlertSuccess {
+			messageKey = "CronJobSuccessAlert"
+		}
+		name := cronJobTaskName(detail.Project, params)
+		return i18n.GetMsgWithMap(messageKey, map[string]interface{}{"name": name, "node": getNodeName(agentInfo), "ip": getNodeIp(agentInfo)})
 	case "clams":
 		return i18n.GetMsgWithMap("ClamAlert", map[string]interface{}{"num": getValueByIndex(params, "1"), "node": getNodeName(agentInfo), "ip": getNodeIp(agentInfo)})
 	case "panelLogin":

agent/utils/alert/cronjob.go

@@ -0,0 +1,130 @@
+package alert
+
+import (
+	"encoding/json"
+	"fmt"
+	"strings"
+
+	"github.com/1Panel-dev/1Panel/agent/app/dto"
+)
+
+const (
+	CronJobAlertFailed  = "failed"
+	CronJobAlertSuccess = "success"
+	CronJobAlertBoth    = "both"
+)
+
+func cronJobAlertParams(advanced string) (map[string]json.RawMessage, error) {
+	params := make(map[string]json.RawMessage)
+	if strings.TrimSpace(advanced) != "" {
+		if err := json.Unmarshal([]byte(advanced), &params); err != nil {
+			return nil, fmt.Errorf("invalid cronjob alert advanced parameters: %w", err)
+		}
+	}
+	if params == nil {
+		params = make(map[string]json.RawMessage)
+	}
+	return params, nil
+}
+
+func CronJobAlertTriggerMode(advanced string) (string, error) {
+	params, err := cronJobAlertParams(advanced)
+	if err != nil {
+		return CronJobAlertFailed, err
+	}
+	mode := CronJobAlertFailed
+	if raw, ok := params["alertTriggerMode"]; ok {
+		if err := json.Unmarshal(raw, &mode); err != nil {
+			return CronJobAlertFailed, fmt.Errorf("invalid cronjob alert trigger mode: %w", err)
+		}
+	}
+	switch mode {
+	case CronJobAlertFailed, CronJobAlertSuccess, CronJobAlertBoth:
+		return mode, nil
+	default:
+		return CronJobAlertFailed, fmt.Errorf("invalid cronjob alert trigger mode %q", mode)
+	}
+}
+
+func MergeCronJobAlertParams(previous, incoming string) (string, error) {
+	params, err := cronJobAlertParams(previous)
+	if err != nil {
+		return "", err
+	}
+	updates, err := cronJobAlertParams(incoming)
+	if err != nil {
+		return "", err
+	}
+	for key, value := range updates {
+		params[key] = value
+	}
+	if _, ok := params["alertTriggerMode"]; !ok {
+		params["alertTriggerMode"] = json.RawMessage(`"failed"`)
+	}
+	data, err := json.Marshal(params)
+	if err != nil {
+		return "", err
+	}
+	if _, err := CronJobAlertTriggerMode(string(data)); err != nil {
+		return "", err
+	}
+	return string(data), nil
+}
+
+func MatchCronJobAlertResult(advanced, result string) bool {
+	mode, err := CronJobAlertTriggerMode(advanced)
+	if err != nil {
+		return false
+	}
+	if result == "" {
+		result = CronJobAlertFailed
+	}
+	if result != CronJobAlertFailed && result != CronJobAlertSuccess {
+		return false
+	}
+	return mode == CronJobAlertBoth || mode == result
+}
+
+func CreateTaskAlertParams(pushAlert dto.PushAlert) []dto.Param {
+	params := CreateAlertParams(GetCronJobTypeName(pushAlert.Param))
+	if GetCronJobType(pushAlert.AlertType) != "cronJob" {
+		return params
+	}
+	params = append(params, CreateCronJobResultParam(pushAlert.Result))
+	return params
+}
+
+func cronJobTaskName(project string, params []dto.Param) string {
+	if project != "" {
+		return project
+	}
+	if name := getValueByIndex(params, "taskName"); name != "" {
+		return name
+	}
+	return getValueByIndex(params, "1")
+}
+
+func CreateCronJobResultParam(result string) dto.Param {
+	value := "失败"
+	if result == CronJobAlertSuccess {
+		value = "成功"
+	}
+	return dto.Param{Index: "2", Key: "result", Value: value}
+}
+
+func CronJobAlertResultFromParams(params []dto.Param) string {
+	for _, param := range params {
+		if param.Index == "result" {
+			if param.Value == CronJobAlertSuccess {
+				return CronJobAlertSuccess
+			}
+			return CronJobAlertFailed
+		}
+	}
+	for _, param := range params {
+		if param.Index == "2" && param.Key == "result" && param.Value == "成功" {
+			return CronJobAlertSuccess
+		}
+	}
+	return CronJobAlertFailed
+}

agent/utils/alert/custom_webhook.go

@@ -46,7 +46,7 @@ func CreateTaskScanCustomWebhookAlertLog(
 	transport *http.Transport,
 	agentInfo *dto.AgentInfo,
 ) error {
-	params := CreateAlertParams(GetCronJobTypeName(pushAlert.Param))
+	params := CreateTaskAlertParams(pushAlert)
 	alertInfo := info
 	alertInfo.Type = alertType
 	create.Type = GetCronJobType(alertType)
@@ -99,7 +99,7 @@ func customWebhookTemplateData(rawDetail string, agentInfo *dto.AgentInfo, occur
 	if businessType == "" {
 		return webhook_sender.TemplateData{}, errors.New("resolve custom webhook alert detail failed")
 	}
-	content := GetSendContent(businessType, detail.Params, agentInfo)
+	content := GetAlertDetailContent(detail, agentInfo)
 	if content == "" {
 		content = i18n.GetMsgWithMap("CommonAlert", map[string]interface{}{"msg": detail.Title})
 	}

agent/utils/alert_push/alert_push.go

@@ -26,6 +26,9 @@ func PushAlert(pushAlert dto.PushAlert) error {
 	}
 	var alert dto.AlertDTO
 	_ = copier.Copy(&alert, &alertInfo)
+	if alertUtil.GetCronJobType(pushAlert.AlertType) == "cronJob" && !alertUtil.MatchCronJobAlertResult(alert.AdvancedParams, pushAlert.Result) {
+		return nil
+	}
 
 	methods := strings.Split(alert.Method, ",")
 	for _, m := range methods {
@@ -126,7 +129,7 @@ func sendAlert(alertRepo repo.IAlertRepo, alert dto.AlertDTO, pushAlert dto.Push
 		}
 		transport := xpack.MultiNodeProvider.LoadRequestTransport()
 		agentInfo, _ := xpack.MultiNodeProvider.GetAgentInfo()
-		params := alertUtil.CreateAlertParams(alertUtil.GetCronJobTypeName(pushAlert.Param))
+		params := alertUtil.CreateTaskAlertParams(pushAlert)
 		alertDetail := alertUtil.ProcessAlertDetail(alert, pushAlert.TaskName, params, constant.Bark)
 		alertRule := alertUtil.ProcessAlertRule(alert)
 		create.AlertRule = alertRule

agent/utils/cloud_storage/client/sftp.go

@@ -97,7 +97,7 @@ func (s sftpClient) Upload(ctx context.Context, src, target string) (bool, error
 	}()
 	defer close(done)
 
-	client, err := sftp.NewClient(sshClient)
+	client, err := sftp.NewClient(sshClient, sftp.UseConcurrentWrites(true))
 	if err != nil {
 		return false, err
 	}
@@ -127,7 +127,15 @@ func (s sftpClient) Upload(ctx context.Context, src, target string) (bool, error
 	}
 	defer dstFile.Close()
 
-	if _, err := io.Copy(dstFile, srcFile); err != nil {
+	// ReadFrom uses the pipelined write path of pkg/sftp. A plain io.Copy would
+	// pick os.File.WriteTo (the source is an *os.File), which issues one
+	// synchronous SFTP WRITE per packet and waits for the reply every time, so
+	// the throughput is bounded by packetSize/RTT.
+	written, err := dstFile.ReadFrom(srcFile)
+	if err != nil {
+		// A failed concurrent write may leave holes in the remote file, so cut
+		// it back to the number of bytes actually written.
+		_ = dstFile.Truncate(written)
 		if ctxErr := ctx.Err(); ctxErr != nil {
 			return false, ctxErr
 		}

agent/utils/docker/network_cleanup.go

@@ -0,0 +1,108 @@
+package docker
+
+import (
+	"context"
+	"sort"
+
+	"github.com/1Panel-dev/1Panel/agent/app/dto"
+	"github.com/docker/docker/api/types/container"
+	"github.com/docker/docker/api/types/network"
+	"github.com/docker/docker/errdefs"
+)
+
+type NetworkCleanupClient interface {
+	NetworkList(context.Context, network.ListOptions) ([]network.Inspect, error)
+	ContainerList(context.Context, container.ListOptions) ([]container.Summary, error)
+	NetworkInspect(context.Context, string, network.InspectOptions) (network.Inspect, error)
+	NetworkRemove(context.Context, string) error
+}
+
+func CleanUnusedNetworks(ctx context.Context, cli NetworkCleanupClient, onResult ...func(string, dto.NetworkCleanupItem)) (*dto.NetworkCleanupReport, error) {
+	networks, err := cli.NetworkList(ctx, network.ListOptions{})
+	if err != nil {
+		return nil, err
+	}
+	containers, err := cli.ContainerList(ctx, container.ListOptions{All: true})
+	if err != nil {
+		return nil, err
+	}
+	used := make(map[string]bool)
+	for _, c := range containers {
+		if c.NetworkSettings == nil {
+			continue
+		}
+		for name, endpoint := range c.NetworkSettings.Networks {
+			used[name] = true
+			if endpoint != nil {
+				used[endpoint.NetworkID] = true
+			}
+		}
+	}
+	report := &dto.NetworkCleanupReport{Deleted: []dto.NetworkCleanupItem{}, Skipped: []dto.NetworkCleanupItem{}, Failed: []dto.NetworkCleanupItem{}}
+	record := func(status string, item dto.NetworkCleanupItem) {
+		switch status {
+		case "deleted":
+			report.Deleted = append(report.Deleted, item)
+		case "skipped":
+			report.Skipped = append(report.Skipped, item)
+		case "failed":
+			report.Failed = append(report.Failed, item)
+		}
+		for _, notify := range onResult {
+			if notify != nil {
+				notify(status, item)
+			}
+		}
+	}
+	sort.Slice(networks, func(i, j int) bool { return networks[i].Name < networks[j].Name })
+	for _, n := range networks {
+		if err := ctx.Err(); err != nil {
+			return report, err
+		}
+		item := dto.NetworkCleanupItem{ID: n.ID, Name: n.Name}
+		switch {
+		case n.Name == "none" || n.Name == "host" || n.Name == "bridge" || n.Name == "1panel-network":
+			item.Reason = "protected"
+		case n.Scope != "local" || n.Ingress || n.ConfigOnly:
+			item.Reason = "unsupported_network"
+		case used[n.Name] || used[n.ID]:
+			item.Reason = "container_connected"
+		}
+		if item.Reason != "" {
+			record("skipped", item)
+			continue
+		}
+		inspected, err := cli.NetworkInspect(ctx, n.ID, network.InspectOptions{})
+		if err != nil {
+			if errdefs.IsNotFound(err) {
+				item.Reason = "already_removed"
+				record("skipped", item)
+			} else {
+				item.Reason = "inspect_failed"
+				record("failed", item)
+			}
+			continue
+		}
+		if len(inspected.Containers) > 0 {
+			item.Reason = "container_connected"
+			record("skipped", item)
+			continue
+		}
+		if err := cli.NetworkRemove(ctx, n.ID); err != nil {
+			switch {
+			case errdefs.IsNotFound(err):
+				item.Reason = "already_removed"
+				record("skipped", item)
+			case errdefs.IsConflict(err):
+				item.Reason = "network_in_use"
+				record("skipped", item)
+			default:
+				item.Reason = "remove_failed"
+				record("failed", item)
+			}
+			continue
+		}
+		record("deleted", item)
+	}
+	return report, nil
+}

agent/utils/docker/network_cleanup_test.go

@@ -0,0 +1,72 @@
+package docker
+
+import (
+	"context"
+	"errors"
+	"reflect"
+	"testing"
+
+	"github.com/docker/docker/api/types/container"
+	"github.com/docker/docker/api/types/network"
+	"github.com/docker/docker/errdefs"
+)
+
+type cleanupFake struct {
+	networks                    []network.Inspect
+	containers                  []container.Summary
+	inspected                   map[string]network.Inspect
+	inspectErrors, removeErrors map[string]error
+	listError                   error
+	removed                     []string
+}
+
+func (f *cleanupFake) NetworkList(context.Context, network.ListOptions) ([]network.Inspect, error) {
+	return f.networks, nil
+}
+func (f *cleanupFake) ContainerList(_ context.Context, o container.ListOptions) ([]container.Summary, error) {
+	if !o.All {
+		panic("must include stopped containers")
+	}
+	return f.containers, f.listError
+}
+func (f *cleanupFake) NetworkInspect(_ context.Context, id string, _ network.InspectOptions) (network.Inspect, error) {
+	return f.inspected[id], f.inspectErrors[id]
+}
+func (f *cleanupFake) NetworkRemove(_ context.Context, id string) error {
+	f.removed = append(f.removed, id)
+	return f.removeErrors[id]
+}
+
+func TestNetworkCleanupProtectsAndReports(t *testing.T) {
+	f := &cleanupFake{inspected: map[string]network.Inspect{"attached": {Containers: map[string]network.EndpointResource{"container": {}}}}, inspectErrors: map[string]error{"unknown": errors.New("inspect failure")}, removeErrors: map[string]error{"race": errdefs.Conflict(errors.New("has active endpoints")), "gone": errdefs.NotFound(errors.New("gone")), "failed": errors.New("denied")}}
+	for _, name := range []string{"none", "host", "bridge", "1panel-network", "configured", "attached", "stopped", "free", "unknown", "race", "gone", "failed"} {
+		f.networks = append(f.networks, network.Inspect{ID: name, Name: name, Scope: "local"})
+	}
+	// Compose ownership labels do not protect an otherwise unused network.
+	for i := range f.networks {
+		if f.networks[i].Name == "configured" {
+			f.networks[i].Labels = map[string]string{"com.docker.compose.project": "demo", "com.docker.compose.network": "default"}
+		}
+	}
+	f.containers = []container.Summary{{State: "exited", NetworkSettings: &container.NetworkSettingsSummary{Networks: map[string]*network.EndpointSettings{"stopped": {NetworkID: "stopped"}}}}}
+	report, err := CleanUnusedNetworks(context.Background(), f)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !reflect.DeepEqual(f.removed, []string{"configured", "failed", "free", "gone", "race"}) {
+		t.Fatal(f.removed)
+	}
+	if len(report.Deleted) != 2 || report.Deleted[0].Name != "configured" || report.Deleted[1].Name != "free" || len(report.Failed) != 2 || len(report.Skipped) != 8 {
+		t.Fatalf("%+v", report)
+	}
+}
+
+func TestNetworkCleanupDiscoveryFailureDeletesNothing(t *testing.T) {
+	f := &cleanupFake{networks: []network.Inspect{{ID: "free", Name: "free", Scope: "local"}}, listError: errors.New("cannot list containers")}
+	if _, err := CleanUnusedNetworks(context.Background(), f); err == nil {
+		t.Fatal("expected error")
+	}
+	if len(f.removed) > 0 {
+		t.Fatal(f.removed)
+	}
+}

agent/utils/re/re.go

@@ -6,6 +6,8 @@ import (
 )
 
 const (
+	VLLMMetricSamplePattern            = `^([a-zA-Z_:][a-zA-Z0-9_:]*)(\{(?:[^"\\]|"(?:\\.|[^"\\])*")*\})?\s+(\S+)(?:\s+.*)?$`
+	VLLMMetricLabelPattern             = `([a-zA-Z_][a-zA-Z0-9_]*)\s*=\s*("(?:[^"\\]|\\.)*")`
 	NumberAlphaPattern                 = `(\d+)([A-Za-z]+)`
 	ComposeDisallowedCharsPattern      = `[^a-z0-9_-]+`
 	ComposeNamePattern                 = `^[a-z0-9][a-z0-9_-]{0,255}$`
@@ -61,6 +63,8 @@ var regexMap = make(map[string]*regexp.Regexp)
 
 func Init() {
 	patterns := []string{
+		VLLMMetricSamplePattern,
+		VLLMMetricLabelPattern,
 		NumberAlphaPattern,
 		ComposeDisallowedCharsPattern,
 		ComposeNamePattern,

core/app/service/upgrade.go

@@ -10,6 +10,7 @@ import (
 	"strconv"
 	"strings"
 	"syscall"
+	"time"
 
 	"github.com/1Panel-dev/1Panel/core/app/dto"
 	"github.com/1Panel-dev/1Panel/core/app/model"
@@ -25,6 +26,7 @@ import (
 	"github.com/1Panel-dev/1Panel/core/utils/req_helper"
 	upgradeUtil "github.com/1Panel-dev/1Panel/core/utils/upgrade"
 	"github.com/1Panel-dev/1Panel/core/utils/xpack"
+	"golang.org/x/net/html"
 )
 
 type serviceInfo struct {
@@ -299,7 +301,7 @@ func (u *UpgradeService) LoadRelease() ([]dto.ReleasesNotes, error) {
 	docSource, _ := settingRepo.GetValueByKey("DocSource")
 	lang, _ := settingRepo.GetValueByKey("Language")
 	var notes []dto.ReleasesNotes
-	url := "https://1panel.cn/docs/v2/search/search_index.json"
+	url := "https://docs.fit2cloud.com/1panel/changelog/"
 	useIntlDocs := false
 	lang = strings.ToLower(strings.TrimSpace(lang))
 	if docSource == "withByRegion" {
@@ -315,6 +317,12 @@ func (u *UpgradeService) LoadRelease() ([]dto.ReleasesNotes, error) {
 		return notes, err
 	}
 	defer resp.Body.Close()
+	if !useIntlDocs {
+		if resp.StatusCode != http.StatusOK {
+			return notes, fmt.Errorf("load release notes failed: HTTP %d", resp.StatusCode)
+		}
+		return parseReleaseHTML(resp.Body)
+	}
 	body, err := io.ReadAll(resp.Body)
 	if err != nil {
 		return notes, err
@@ -336,6 +344,89 @@ func (u *UpgradeService) LoadRelease() ([]dto.ReleasesNotes, error) {
 	return notes, nil
 }
 
+func parseReleaseHTML(reader io.Reader) ([]dto.ReleasesNotes, error) {
+	doc, err := html.Parse(reader)
+	if err != nil {
+		return nil, err
+	}
+	var article *html.Node
+	for node := range doc.Descendants() {
+		if node.Type == html.ElementNode && node.Data == "article" {
+			article = node
+			break
+		}
+	}
+	if article == nil {
+		return nil, fmt.Errorf("release notes article not found")
+	}
+	textContent := func(node *html.Node) string {
+		var text strings.Builder
+		for child := range node.Descendants() {
+			if child.Type == html.TextNode {
+				text.WriteString(child.Data)
+			}
+		}
+		return strings.TrimSpace(strings.ReplaceAll(text.String(), "\u200b", ""))
+	}
+	var notes []dto.ReleasesNotes
+	for heading := range article.Descendants() {
+		if heading.Type != html.ElementNode || heading.Data != "h3" {
+			continue
+		}
+		version := textContent(heading)
+		if !strings.HasPrefix(version, "v") {
+			continue
+		}
+		item := dto.ReleasesNotes{Version: version}
+		var content strings.Builder
+		section := ""
+		for node := heading.NextSibling; node != nil; node = node.NextSibling {
+			if node.Type != html.ElementNode {
+				continue
+			}
+			if node.Data == "h1" || node.Data == "h2" || node.Data == "h3" {
+				break
+			}
+			if item.CreatedAt == "" {
+				date := textContent(node)
+				if _, err := time.Parse("2006年1月2日", date); node.Data != "p" || err != nil {
+					return nil, fmt.Errorf("release date not found for %s", version)
+				}
+				item.CreatedAt = date
+				continue
+			}
+			if node.Data == "p" {
+				section = textContent(node)
+			}
+			for child := range node.Descendants() {
+				if child.Type != html.ElementNode || child.Data != "li" {
+					continue
+				}
+				switch section {
+				case "新增功能":
+					item.NewCount++
+				case "功能优化":
+					item.OptimizationCount++
+				case "问题修复":
+					item.FixCount++
+				}
+			}
+			if err := html.Render(&content, node); err != nil {
+				return nil, err
+			}
+		}
+		if item.CreatedAt == "" || content.Len() == 0 {
+			return nil, fmt.Errorf("release notes content not found for %s", version)
+		}
+		item.Content = content.String()
+		notes = append(notes, item)
+	}
+	if len(notes) == 0 {
+		return nil, fmt.Errorf("release notes versions not found")
+	}
+	return notes, nil
+}
+
 func analyzeDoc(version, content string) dto.ReleasesNotes {
 	var item dto.ReleasesNotes
 	parts := strings.Split(content, "<p>")

core/cmd/server/docs/x-log.json

@@ -644,6 +644,13 @@
 		"formatZH": "创建容器网络 name",
 		"formatEN": "create container network [name]"
 	},
+	"/containers/network/clean": {
+		"bodyKeys": [],
+		"paramKeys": [],
+		"beforeFunctions": [],
+		"formatZH": "清理未使用的容器网络",
+		"formatEN": "Clean unused container networks"
+	},
 	"/containers/network/del": {
 		"bodyKeys": [
 			"names"
@@ -2486,6 +2493,17 @@
 		"formatZH": "删除虚拟机 [name]",
 		"formatEN": "delete VM [name]"
 	},
+	"/core/xpack/vms/disks/resize": {
+		"bodyKeys": [
+			"name",
+			"target",
+			"sizeGiB"
+		],
+		"paramKeys": [],
+		"beforeFunctions": [],
+		"formatZH": "扩容虚拟机 [name] 磁盘 [target] 至 [sizeGiB] GiB",
+		"formatEN": "expand VM [name] disk [target] to [sizeGiB] GiB"
+	},
 	"/core/xpack/vms/environment/enable": {
 		"bodyKeys": [],
 		"paramKeys": [],
@@ -4028,12 +4046,21 @@
 	},
 	"/runtimes/operate": {
 		"bodyKeys": [
-			"id"
+			"ID"
 		],
 		"paramKeys": [],
-		"beforeFunctions": [],
-		"formatZH": "操作运行环境 [id]",
-		"formatEN": "Operate runtime [id]"
+		"beforeFunctions": [
+			{
+				"input_column": "id",
+				"input_value": "ID",
+				"isList": false,
+				"db": "runtimes",
+				"output_column": "name",
+				"output_value": "name"
+			}
+		],
+		"formatZH": "操作运行环境 [name]",
+		"formatEN": "Operate runtime [name]"
 	},
 	"/runtimes/php/config": {
 		"bodyKeys": [

core/go.mod

@@ -1,6 +1,6 @@
 module github.com/1Panel-dev/1Panel/core
 
-go 1.26.1
+go 1.26.6
 
 require (
 	github.com/beevik/etree v1.7.0
@@ -39,8 +39,8 @@ require (
 	github.com/swaggo/files/v2 v2.0.2
 	github.com/swaggo/swag v1.16.6
 	github.com/xlzd/gotp v0.1.0
-	golang.org/x/crypto v0.55.0
-	golang.org/x/image v0.41.0
+	golang.org/x/crypto v0.56.0
+	golang.org/x/image v0.45.0
 	golang.org/x/net v0.58.0
 	golang.org/x/oauth2 v0.36.0
 	golang.org/x/sys v0.47.0
@@ -129,9 +129,9 @@ require (
 	go.mongodb.org/mongo-driver/v2 v2.5.1 // indirect
 	go.yaml.in/yaml/v3 v3.0.5 // indirect
 	golang.org/x/arch v0.26.0 // indirect
-	golang.org/x/mod v0.38.0 // indirect
+	golang.org/x/mod v0.40.0 // indirect
 	golang.org/x/sync v0.22.0 // indirect
-	golang.org/x/tools v0.48.0 // indirect
+	golang.org/x/tools v0.49.0 // indirect
 	google.golang.org/protobuf v1.36.11 // indirect
 	modernc.org/libc v1.72.0 // indirect
 	modernc.org/mathutil v1.7.1 // indirect

core/go.sum

@@ -124,8 +124,6 @@ github.com/google/go-tpm v0.9.8/go.mod h1:h9jEsEECg7gtLis0upRBQU+GhYVH6jMjrFxI8u
 github.com/google/go-tpm-tools v0.3.13-0.20230620182252-4639ecce2aba h1:qJEJcuLzH5KDR0gKc0zcktin6KSAwL7+jWKBYceddTc=
 github.com/google/go-tpm-tools v0.3.13-0.20230620182252-4639ecce2aba/go.mod h1:EFYHy8/1y2KfgTAsx7Luu7NGhoxtuVHnNo8jE7FikKc=
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
-github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e h1:ijClszYn+mADRFY17kjQEVQ1XRhq2/JR1M3sGqeJoxs=
-github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e/go.mod h1:boTsfXsheKC2y+lKOCMpSfarhxDeIzfZG1jqGcPl3cA=
 github.com/google/pprof v0.0.0-20250820193118-f64d9cf942d6 h1:EEHtgt9IwisQ2AZ4pIsMjahcegHh6rmhqxzIRQIyepY=
 github.com/google/pprof v0.0.0-20250820193118-f64d9cf942d6/go.mod h1:I6V7YzU0XDpsHqbsyrghnFZLO1gwK6NPTNvmetQIk9U=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
@@ -304,13 +302,13 @@ golang.org/x/arch v0.26.0 h1:jZ6dpec5haP/fUv1kLCbuJy6dnRrfX6iVK08lZBFpk4=
 golang.org/x/arch v0.26.0/go.mod h1:0X+GdSIP+kL5wPmpK7sdkEVTt2XoYP0cSjQSbZBwOi8=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
 golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
-golang.org/x/crypto v0.55.0 h1:+KWHjbgOaAQ66dh/YlkZKHlz9ZUlq61AFirAR9ntP8M=
-golang.org/x/crypto v0.55.0/go.mod h1:uq0V9dE/fzQuJtbnL+2EhWOE63vo164FY8xqEnV9xis=
+golang.org/x/crypto v0.56.0 h1:GUh5Ii4J5jtcseSMiRqr1jXCNHoxjeV9Fmekc2oLy6Y=
+golang.org/x/crypto v0.56.0/go.mod h1:OMW5y6CY9l38uPLmxU6l6pwcXp1obtLo3e6gT7gQR2I=
 golang.org/x/image v0.0.0-20190910094157-69e4b8554b2a/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0=
-golang.org/x/image v0.41.0 h1:8wS72eGJMJaBxK6okTzd4WaXumUlTVlb753MlsSvTCo=
-golang.org/x/image v0.41.0/go.mod h1:uIc348UZMSvS5Z65CVZ7iDPaNobNFEPeJ4kbqTOszmA=
-golang.org/x/mod v0.38.0 h1:MECBjubtXD7yj4HrhIUcywNaGeNVUdfVnxmPajOk4yk=
-golang.org/x/mod v0.38.0/go.mod h1:V6Xz0pq8TQ3dGqVQ1FVHuelZpAL0uNhSkk9ogYP3c40=
+golang.org/x/image v0.45.0 h1:FMb1nTbH5H9vF55SriQHgFw5GnNL9Jg6L25BwXKzhB0=
+golang.org/x/image v0.45.0/go.mod h1:n62x/7RqlwXDvGsSU4u6IUTUf6KghUZ9Bt7cG/T9Fx4=
+golang.org/x/mod v0.40.0 h1:hUv+3cXcdRHz08UmSiOob7sadHig73uo5bkXxQ/tvUs=
+golang.org/x/mod v0.40.0/go.mod h1:0/weTWkPWGBikyTWAX3dkjVztMmBA5hM0DH6BElSupE=
 golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
 golang.org/x/net v0.0.0-20201202161906-c7110b5ffcbb/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
 golang.org/x/net v0.58.0 h1:ynWG7rqYi4ccpTEuPZ2QGWHktVEM9DMCj9yzDE0Q7To=
@@ -335,8 +333,8 @@ golang.org/x/text v0.41.0/go.mod h1:jvf1O8ajNzZqhSrQBPbutR/EB83Cc0CFrezNQIwbb5M=
 golang.org/x/time v0.12.0 h1:ScB/8o8olJvc+CQPWrK3fPZNfh7qgwCrY0zJmoEQLSE=
 golang.org/x/time v0.12.0/go.mod h1:CDIdPxbZBQxdj6cxyCIdrNogrJKMJ7pr37NYpMcMDSg=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/tools v0.48.0 h1:3+hClM1aLL5mjMKm5ovokw9epgRXPuu2tILgismM6RE=
-golang.org/x/tools v0.48.0/go.mod h1:08xX0orndb/F7jJxGDicx061tyd5pcMto75YMAXr6lk=
+golang.org/x/tools v0.49.0 h1:3NI7VXzL9+1WZD52Dx2ttoPwD5DWrFGpl9mFZDlmisI=
+golang.org/x/tools v0.49.0/go.mod h1:SJNXV9DBKT0UbdttsQjbfJlAE/q+y36++zo3uL3N0Oo=
 golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 google.golang.org/protobuf v1.36.11 h1:fV6ZwhNocDyBLK0dj+fg8ektcVegBBuEolpbTQyBNVE=
 google.golang.org/protobuf v1.36.11/go.mod h1:HTf+CrKn2C3g5S8VImy6tdcUvCska2kB7j23XfzDpco=

core/i18n/lang/en.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "Virtual machine network does not exist: {{ .detail }}"
 ErrVMInvalidPath: "Invalid virtual machine path: {{ .detail }}"
 ErrVMInvalidInput: "Invalid virtual machine parameter: {{ .detail }}"
 ErrVMCommandFailed: "Virtual machine command failed: {{ .err }}"
+ErrVMDiskResizeBlocked: "The disk does not meet expansion requirements or is referenced by another device. Check the disk status."
+ErrVMDiskReferenced: "This virtual disk is used by another virtual machine or device and cannot be expanded yet."
+ErrVMDiskReferenceCheck: "Cannot complete virtual disk reference checks. Expansion is unavailable; check the disks of other virtual machines."
+ErrVMSnapshotMetadataSync: "The snapshot was restored, but disk information could not be synced. Refresh and check; do not restore the snapshot again."
 VMCreateTask: "Create virtual machine [{{ .name }}]"
 VMCreateFromTemplateTask: "Create virtual machine [{{ .name }}] from template"
 VMUpdateTask: "Update virtual machine [{{ .name }}]"
+VMDiskResizeTask: "Expand disk [{{ .disk }}] of virtual machine [{{ .name }}]"
 VMRenameTask: "Rename virtual machine [{{ .name }}] to [{{ .newName }}]"
 VMDeleteTask: "Delete virtual machine [{{ .name }}]"
 VMTemplateCreateTask: "Save virtual machine [{{ .name }}] as template [{{ .template }}]"

core/i18n/lang/es-ES.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "La red de la máquina virtual no existe: {{ .detail }}"
 ErrVMInvalidPath: "Ruta de máquina virtual no válida: {{ .detail }}"
 ErrVMInvalidInput: "Parámetro de máquina virtual no válido: {{ .detail }}"
 ErrVMCommandFailed: "Error al ejecutar el comando de la máquina virtual: {{ .err }}"
+ErrVMDiskResizeBlocked: "El disco no cumple los requisitos de ampliación o está referenciado por otro dispositivo. Compruebe su estado."
+ErrVMDiskReferenced: "Otro equipo virtual o dispositivo utiliza este disco virtual; la ampliación aún no se admite."
+ErrVMDiskReferenceCheck: "No se pueden comprobar las referencias al disco virtual. La ampliación no está disponible; revise los discos de otras máquinas virtuales."
+ErrVMSnapshotMetadataSync: "La instantánea se restauró, pero falló la sincronización de los datos del disco. Actualice y compruebe; no restaure de nuevo la instantánea."
 VMCreateTask: "Crear máquina virtual [{{ .name }}]"
 VMCreateFromTemplateTask: "Crear máquina virtual [{{ .name }}] desde plantilla"
 VMUpdateTask: "Actualizar máquina virtual [{{ .name }}]"
+VMDiskResizeTask: "Ampliar disco [{{ .disk }}] de la máquina virtual [{{ .name }}]"
 VMRenameTask: "Renombrar máquina virtual [{{ .name }}] a [{{ .newName }}]"
 VMDeleteTask: "Eliminar máquina virtual [{{ .name }}]"
 VMTemplateCreateTask: "Guardar máquina virtual [{{ .name }}] como plantilla [{{ .template }}]"

core/i18n/lang/fa.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "شبکه ماشین مجازی وجود ندارد: {{ .d
 ErrVMInvalidPath: "مسیر ماشین مجازی نامعتبر است: {{ .detail }}"
 ErrVMInvalidInput: "پارامتر ماشین مجازی نامعتبر است: {{ .detail }}"
 ErrVMCommandFailed: "اجرای دستور ماشین مجازی ناموفق بود: {{ .err }}"
+ErrVMDiskResizeBlocked: "دیسک شرایط افزایش ظرفیت را ندارد یا دستگاه دیگری به آن ارجاع می‌دهد. وضعیت دیسک را بررسی کنید."
+ErrVMDiskReferenced: "این دیسک مجازی توسط ماشین مجازی یا دستگاه دیگری استفاده می‌شود؛ افزایش ظرفیت فعلاً پشتیبانی نمی‌شود."
+ErrVMDiskReferenceCheck: "بررسی ارجاع‌های دیسک مجازی کامل نشد. افزایش ظرفیت ممکن نیست؛ دیسک‌های سایر ماشین‌های مجازی را بررسی کنید."
+ErrVMSnapshotMetadataSync: "اسنپ‌شات بازیابی شد، اما همگام‌سازی اطلاعات دیسک ناموفق بود. تازه‌سازی و بررسی کنید؛ نیازی به بازیابی دوباره اسنپ‌شات نیست."
 VMCreateTask: "ایجاد ماشین مجازی [{{ .name }}]"
 VMCreateFromTemplateTask: "ایجاد ماشین مجازی [{{ .name }}] از قالب"
 VMUpdateTask: "به‌روزرسانی ماشین مجازی [{{ .name }}]"
+VMDiskResizeTask: "افزایش ظرفیت دیسک [{{ .disk }}] ماشین مجازی [{{ .name }}]"
 VMRenameTask: "تغییر نام ماشین مجازی [{{ .name }}] به [{{ .newName }}]"
 VMDeleteTask: "حذف ماشین مجازی [{{ .name }}]"
 VMTemplateCreateTask: "ذخیره ماشین مجازی [{{ .name }}] به عنوان قالب [{{ .template }}]"

core/i18n/lang/ja.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "仮想マシンネットワークが存在しません: {
 ErrVMInvalidPath: "仮想マシンのパスが無効です: {{ .detail }}"
 ErrVMInvalidInput: "仮想マシンのパラメータが無効です: {{ .detail }}"
 ErrVMCommandFailed: "仮想マシンコマンドの実行に失敗しました: {{ .err }}"
+ErrVMDiskResizeBlocked: "ディスクが拡張条件を満たしていないか、他のデバイスから参照されています。ディスクの状態を確認してください。"
+ErrVMDiskReferenced: "この仮想ディスクは他の仮想マシンまたはデバイスで使用されているため、現在は拡張できません。"
+ErrVMDiskReferenceCheck: "仮想ディスクの参照を確認できないため拡張できません。他の仮想マシンのディスク状態を確認してください。"
+ErrVMSnapshotMetadataSync: "スナップショットは復元されましたが、ディスク情報の同期に失敗しました。更新して確認してください。再度復元する必要はありません。"
 VMCreateTask: "仮想マシン [{{ .name }}] を作成"
 VMCreateFromTemplateTask: "テンプレートから仮想マシン [{{ .name }}] を作成"
 VMUpdateTask: "仮想マシン [{{ .name }}] を更新"
+VMDiskResizeTask: "仮想マシン [{{ .name }}] のディスク [{{ .disk }}] を拡張"
 VMRenameTask: "仮想マシン [{{ .name }}] を [{{ .newName }}] にリネーム"
 VMDeleteTask: "仮想マシン [{{ .name }}] を削除"
 VMTemplateCreateTask: "仮想マシン [{{ .name }}] をテンプレート [{{ .template }}] として保存"

core/i18n/lang/ko.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "가상 머신 네트워크가 존재하지 않습니다:
 ErrVMInvalidPath: "가상 머신 경로가 유효하지 않습니다: {{ .detail }}"
 ErrVMInvalidInput: "가상 머신 매개변수가 유효하지 않습니다: {{ .detail }}"
 ErrVMCommandFailed: "가상 머신 명령 실행 실패: {{ .err }}"
+ErrVMDiskResizeBlocked: "디스크가 확장 조건을 충족하지 않거나 다른 장치에서 참조하고 있습니다. 디스크 상태를 확인하세요."
+ErrVMDiskReferenced: "다른 가상 머신 또는 장치가 이 가상 디스크를 사용 중이므로 현재 확장할 수 없습니다."
+ErrVMDiskReferenceCheck: "가상 디스크 참조를 확인할 수 없어 확장할 수 없습니다. 다른 가상 머신의 디스크 상태를 확인하세요."
+ErrVMSnapshotMetadataSync: "스냅샷은 복원되었지만 디스크 정보 동기화에 실패했습니다. 새로 고쳐 확인하세요. 스냅샷을 다시 복원할 필요는 없습니다."
 VMCreateTask: "가상 머신 [{{ .name }}] 생성"
 VMCreateFromTemplateTask: "템플릿에서 가상 머신 [{{ .name }}] 생성"
 VMUpdateTask: "가상 머신 [{{ .name }}] 업데이트"
+VMDiskResizeTask: "가상 머신 [{{ .name }}]의 디스크 [{{ .disk }}] 확장"
 VMRenameTask: "가상 머신 [{{ .name }}]을(를) [{{ .newName }}](으)로 이름 변경"
 VMDeleteTask: "가상 머신 [{{ .name }}] 삭제"
 VMTemplateCreateTask: "가상 머신 [{{ .name }}]을(를) 템플릿 [{{ .template }}](으)로 저장"

core/i18n/lang/lo.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "ບໍ່ມີເຄືອຂ່າຍເຄື່ອງ
 ErrVMInvalidPath: "ເສັ້ນທາງເຄື່ອງສະເໝືອນບໍ່ຖືກຕ້ອງ: {{ .detail }}"
 ErrVMInvalidInput: "ພາລາມິເຕີເຄື່ອງສະເໝືອນບໍ່ຖືກຕ້ອງ: {{ .detail }}"
 ErrVMCommandFailed: "ຄຳສັ່ງເຄື່ອງສະເໝືອນລົ້ມເຫຼວ: {{ .err }}"
+ErrVMDiskResizeBlocked: "ດິສກ໌ບໍ່ຕອບສະໜອງເງື່ອນໄຂການຂະຫຍາຍ ຫຼື ຖືກອ້າງອີງໂດຍອຸປະກອນອື່ນ. ກະລຸນາກວດສອບສະຖານະດິສກ໌."
+ErrVMDiskReferenced: "ດິສກ໌ເສມືອນນີ້ຖືກໃຊ້ໂດຍເຄື່ອງເສມືອນ ຫຼື ອຸປະກອນອື່ນ; ຍັງບໍ່ຮອງຮັບການຂະຫຍາຍ."
+ErrVMDiskReferenceCheck: "ບໍ່ສາມາດກວດສອບການອ້າງອີງດິສກ໌ເສມືອນໄດ້. ຍັງຂະຫຍາຍບໍ່ໄດ້; ກະລຸນາກວດສອບດິສກ໌ຂອງເຄື່ອງເສມືອນອື່ນ."
+ErrVMSnapshotMetadataSync: "ກູ້ຄືນສະແນັບຊັອດແລ້ວ ແຕ່ຊິງຂໍ້ມູນດິສກ໌ບໍ່ສຳເລັດ. ກະລຸນາໂຫຼດໃໝ່ ແລະ ກວດສອບ; ບໍ່ຈຳເປັນຕ້ອງກູ້ຄືນສະແນັບຊັອດອີກ."
 VMCreateTask: "ສ້າງເຄື່ອງສະເໝືອນ [{{ .name }}]"
 VMCreateFromTemplateTask: "ສ້າງເຄື່ອງສະເໝືອນ [{{ .name }}] ຈາກແມ່ແບບ"
 VMUpdateTask: "ອັບເດດເຄື່ອງສະເໝືອນ [{{ .name }}]"
+VMDiskResizeTask: "ຂະຫຍາຍດິສກ໌ [{{ .disk }}] ຂອງເຄື່ອງເສມືອນ [{{ .name }}]"
 VMRenameTask: "ປ່ຽນຊື່ເຄື່ອງສະເໝືອນ [{{ .name }}] ເປັນ [{{ .newName }}]"
 VMDeleteTask: "ລຶບເຄື່ອງສະເໝືອນ [{{ .name }}]"
 VMTemplateCreateTask: "ບັນທຶກເຄື່ອງສະເໝືອນ [{{ .name }}] ເປັນແມ່ແບບ [{{ .template }}]"

core/i18n/lang/ms.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "Rangkaian mesin maya tidak wujud: {{ .detail }}"
 ErrVMInvalidPath: "Laluan mesin maya tidak sah: {{ .detail }}"
 ErrVMInvalidInput: "Parameter mesin maya tidak sah: {{ .detail }}"
 ErrVMCommandFailed: "Arahan mesin maya gagal: {{ .err }}"
+ErrVMDiskResizeBlocked: "Cakera tidak memenuhi syarat pembesaran atau dirujuk oleh peranti lain. Semak status cakera."
+ErrVMDiskReferenced: "Cakera maya ini digunakan oleh mesin maya atau peranti lain; pembesaran belum disokong."
+ErrVMDiskReferenceCheck: "Rujukan cakera maya tidak dapat disahkan. Pembesaran tidak tersedia; semak cakera mesin maya lain."
+ErrVMSnapshotMetadataSync: "Syot kilat telah dipulihkan, tetapi maklumat cakera gagal disegerakkan. Muat semula dan semak; tidak perlu memulihkan syot kilat sekali lagi."
 VMCreateTask: "Cipta mesin maya [{{ .name }}]"
 VMCreateFromTemplateTask: "Cipta mesin maya [{{ .name }}] daripada templat"
 VMUpdateTask: "Kemas kini mesin maya [{{ .name }}]"
+VMDiskResizeTask: "Besarkan cakera [{{ .disk }}] mesin maya [{{ .name }}]"
 VMRenameTask: "Namakan semula mesin maya [{{ .name }}] kepada [{{ .newName }}]"
 VMDeleteTask: "Padam mesin maya [{{ .name }}]"
 VMTemplateCreateTask: "Simpan mesin maya [{{ .name }}] sebagai templat [{{ .template }}]"

core/i18n/lang/pt-BR.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "Rede da máquina virtual não existe: {{ .detail }}"
 ErrVMInvalidPath: "Caminho da máquina virtual inválido: {{ .detail }}"
 ErrVMInvalidInput: "Parâmetro da máquina virtual inválido: {{ .detail }}"
 ErrVMCommandFailed: "Falha no comando da máquina virtual: {{ .err }}"
+ErrVMDiskResizeBlocked: "O disco não atende aos requisitos de expansão ou é referenciado por outro dispositivo. Verifique seu estado."
+ErrVMDiskReferenced: "Este disco virtual é usado por outra máquina virtual ou dispositivo e ainda não pode ser expandido."
+ErrVMDiskReferenceCheck: "Não foi possível verificar as referências ao disco virtual. A expansão está indisponível; verifique os discos das outras máquinas virtuais."
+ErrVMSnapshotMetadataSync: "O snapshot foi restaurado, mas a sincronização dos dados do disco falhou. Atualize e verifique; não restaure o snapshot novamente."
 VMCreateTask: "Criar máquina virtual [{{ .name }}]"
 VMCreateFromTemplateTask: "Criar máquina virtual [{{ .name }}] a partir do modelo"
 VMUpdateTask: "Atualizar máquina virtual [{{ .name }}]"
+VMDiskResizeTask: "Expandir disco [{{ .disk }}] da máquina virtual [{{ .name }}]"
 VMRenameTask: "Renomear máquina virtual [{{ .name }}] para [{{ .newName }}]"
 VMDeleteTask: "Excluir máquina virtual [{{ .name }}]"
 VMTemplateCreateTask: "Salvar máquina virtual [{{ .name }}] como modelo [{{ .template }}]"

core/i18n/lang/ru.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "Сеть виртуальной машины не сущ
 ErrVMInvalidPath: "Недопустимый путь виртуальной машины: {{ .detail }}"
 ErrVMInvalidInput: "Недопустимые параметры виртуальной машины: {{ .detail }}"
 ErrVMCommandFailed: "Ошибка выполнения команды виртуальной машины: {{ .err }}"
+ErrVMDiskResizeBlocked: "Диск не соответствует условиям расширения или используется другим устройством. Проверьте состояние диска."
+ErrVMDiskReferenced: "Этот виртуальный диск используется другой виртуальной машиной или устройством; расширение пока недоступно."
+ErrVMDiskReferenceCheck: "Не удалось проверить ссылки на виртуальный диск. Расширение недоступно; проверьте диски других виртуальных машин."
+ErrVMSnapshotMetadataSync: "Снимок восстановлен, но синхронизация данных диска не удалась. Обновите данные и проверьте их; повторно восстанавливать снимок не нужно."
 VMCreateTask: "Создать виртуальную машину [{{ .name }}]"
 VMCreateFromTemplateTask: "Создать виртуальную машину [{{ .name }}] из шаблона"
 VMUpdateTask: "Обновить виртуальную машину [{{ .name }}]"
+VMDiskResizeTask: "Расширить диск [{{ .disk }}] виртуальной машины [{{ .name }}]"
 VMRenameTask: "Переименовать виртуальную машину [{{ .name }}] в [{{ .newName }}]"
 VMDeleteTask: "Удалить виртуальную машину [{{ .name }}]"
 VMTemplateCreateTask: "Сохранить виртуальную машину [{{ .name }}] как шаблон [{{ .template }}]"

core/i18n/lang/tr.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "Sanal makine ağı mevcut değil: {{ .detail }}"
 ErrVMInvalidPath: "Geçersiz sanal makine yolu: {{ .detail }}"
 ErrVMInvalidInput: "Geçersiz sanal makine parametresi: {{ .detail }}"
 ErrVMCommandFailed: "Sanal makine komutu başarısız: {{ .err }}"
+ErrVMDiskResizeBlocked: "Disk genişletme koşullarını karşılamıyor veya başka bir aygıt tarafından kullanılıyor. Disk durumunu kontrol edin."
+ErrVMDiskReferenced: "Bu sanal disk başka bir sanal makine veya aygıt tarafından kullanılıyor; genişletme henüz desteklenmiyor."
+ErrVMDiskReferenceCheck: "Sanal disk başvuruları doğrulanamadı. Genişletme kullanılamıyor; diğer sanal makinelerin disklerini kontrol edin."
+ErrVMSnapshotMetadataSync: "Anlık görüntü geri yüklendi ancak disk bilgileri eşitlenemedi. Yenileyip kontrol edin; anlık görüntüyü tekrar geri yüklemeyin."
 VMCreateTask: "Sanal makine [{{ .name }}] oluştur"
 VMCreateFromTemplateTask: "Şablondan sanal makine [{{ .name }}] oluştur"
 VMUpdateTask: "Sanal makine [{{ .name }}] güncelle"
+VMDiskResizeTask: "[{{ .name }}] sanal makinesinin [{{ .disk }}] diskini genişlet"
 VMRenameTask: "Sanal makine [{{ .name }}] adını [{{ .newName }}] olarak değiştir"
 VMDeleteTask: "Sanal makine [{{ .name }}] sil"
 VMTemplateCreateTask: "Sanal makine [{{ .name }}] öğesini şablon [{{ .template }}] olarak kaydet"

core/i18n/lang/zh-Hant.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "虛擬機網路不存在: {{ .detail }}"
 ErrVMInvalidPath: "虛擬機路徑非法: {{ .detail }}"
 ErrVMInvalidInput: "虛擬機參數非法: {{ .detail }}"
 ErrVMCommandFailed: "虛擬機指令執行失敗: {{ .err }}"
+ErrVMDiskResizeBlocked: "磁碟不符合擴容條件，或被其他裝置引用，請檢查磁碟狀態"
+ErrVMDiskReferenced: "此虛擬硬碟被其他虛擬機或裝置使用，暫不支援擴容"
+ErrVMDiskReferenceCheck: "無法完成虛擬硬碟引用檢查，暫不支援擴容，請檢查其他虛擬機的磁碟狀態"
+ErrVMSnapshotMetadataSync: "快照已還原，但磁碟資訊同步失敗，請重新整理檢查，無需重複還原快照"
 VMCreateTask: "建立虛擬機 [{{ .name }}]"
 VMCreateFromTemplateTask: "從範本建立虛擬機 [{{ .name }}]"
 VMUpdateTask: "更新虛擬機 [{{ .name }}]"
+VMDiskResizeTask: "擴容虛擬機 [{{ .name }}] 磁碟 [{{ .disk }}]"
 VMRenameTask: "重新命名虛擬機 [{{ .name }}] 為 [{{ .newName }}]"
 VMDeleteTask: "刪除虛擬機 [{{ .name }}]"
 VMTemplateCreateTask: "轉存虛擬機 [{{ .name }}] 為範本 [{{ .template }}]"

core/i18n/lang/zh.yaml

@@ -152,9 +152,14 @@ ErrVMNetworkNotFound: "虚拟机网络不存在: {{ .detail }}"
 ErrVMInvalidPath: "虚拟机路径非法: {{ .detail }}"
 ErrVMInvalidInput: "虚拟机参数非法: {{ .detail }}"
 ErrVMCommandFailed: "虚拟机命令执行失败: {{ .err }}"
+ErrVMDiskResizeBlocked: "磁盘不符合扩容条件，或被其他设备引用，请检查磁盘状态"
+ErrVMDiskReferenced: "该虚拟硬盘被其他虚拟机或设备使用，暂不支持扩容"
+ErrVMDiskReferenceCheck: "无法完成虚拟硬盘引用检查，暂不支持扩容，请检查其他虚拟机的磁盘状态"
+ErrVMSnapshotMetadataSync: "快照已恢复，但磁盘信息同步失败，请刷新检查，无需重复恢复快照"
 VMCreateTask: "创建虚拟机 [{{ .name }}]"
 VMCreateFromTemplateTask: "从模板创建虚拟机 [{{ .name }}]"
 VMUpdateTask: "更新虚拟机 [{{ .name }}]"
+VMDiskResizeTask: "扩容虚拟机 [{{ .name }}] 磁盘 [{{ .disk }}]"
 VMRenameTask: "重命名虚拟机 [{{ .name }}] 为 [{{ .newName }}]"
 VMDeleteTask: "删除虚拟机 [{{ .name }}]"
 VMTemplateCreateTask: "转存虚拟机 [{{ .name }}] 为模板 [{{ .template }}]"

core/init/router/router.go

@@ -87,6 +87,7 @@ func Routers() *gin.Engine {
 	}
 
 	Router.Use(middleware.FrontendFallback())
+	Router.Use(middleware.WebSocketOriginGuard())
 	Router.Use(middleware.OperationLog())
 	Router.Use(middleware.GlobalLoading())
 	Router.Use(xpack.AuthProvider.CoreAPIAuthMiddleware())

core/middleware/demo_handle.go

@@ -27,6 +27,11 @@ var demoAllowedRoutes = map[demoRoute]struct{}{
 	{http.MethodPost, "/api/v2/toolbox/clam/base"}:             {},
 	{http.MethodPost, "/api/v2/backups/record/size"}:           {},
 
+	{http.MethodPost, "/api/v2/hosts/firewall/base"}:                {},
+	{http.MethodPost, "/api/v2/hosts/firewall/forward/base"}:        {},
+	{http.MethodPost, "/api/v2/hosts/firewall/rules/native/detail"}: {},
+	{http.MethodPost, "/api/v2/hosts/firewall/rules/sync/preview"}:  {},
+
 	{http.MethodPost, "/api/v2/core/auth/login"}:     {},
 	{http.MethodPost, "/api/v2/core/logs/login"}:     {},
 	{http.MethodPost, "/api/v2/core/logs/operation"}: {},

core/middleware/websocket_origin.go

@@ -0,0 +1,41 @@
+package middleware
+
+import (
+	"net/http"
+	"net/url"
+	"strings"
+
+	"github.com/gin-gonic/gin"
+	"github.com/gorilla/websocket"
+)
+
+func WebSocketOriginGuard() gin.HandlerFunc {
+	return func(c *gin.Context) {
+		if !websocket.IsWebSocketUpgrade(c.Request) {
+			c.Next()
+			return
+		}
+
+		if site := c.GetHeader("Sec-Fetch-Site"); site != "" && site != "same-origin" {
+			c.AbortWithStatus(http.StatusForbidden)
+			return
+		}
+		origins := c.Request.Header.Values("Origin")
+		if len(origins) == 0 {
+			c.Next()
+			return
+		}
+		if len(origins) != 1 {
+			c.AbortWithStatus(http.StatusForbidden)
+			return
+		}
+		origin, err := url.Parse(origins[0])
+		if err != nil || (origin.Scheme != "http" && origin.Scheme != "https") ||
+			origin.Host == "" || !strings.EqualFold(origin.Host, c.Request.Host) ||
+			origin.User != nil || origin.Path != "" || origin.RawQuery != "" || origin.ForceQuery || origin.Fragment != "" {
+			c.AbortWithStatus(http.StatusForbidden)
+			return
+		}
+		c.Next()
+	}
+}

docs/README.zh-Hans.md

@@ -53,11 +53,11 @@
 
 $ bash -c "$(curl -sSL https://resource.fit2cloud.com/1panel/package/v2/quick_start.sh)"
 
-如果是用于离线环境，推荐使用 [安装包方式](https://1panel.cn/docs/v2/installation/enterprise_installation/) 进行安装部署。
+如果是用于离线环境，推荐使用 [安装包方式](https://docs.fit2cloud.com/1panel/installation/enterprise-installation/) 进行安装部署。
 
 **学习资料**
 
-- [在线文档](https://1panel.cn/docs/)
+- [在线文档](https://docs.fit2cloud.com/1panel/)
 - [社区论坛](https://bbs.fit2cloud.com/c/1p/7)
 
 ## 版本对比

docs/README.zh-Hant.md

@@ -55,11 +55,11 @@
 curl -sSL https://resource.fit2cloud.com/1panel/package/quick_start.sh -o quick_start.sh && sudo bash quick_start.sh
 ```
 
-如果是用於離線環境，推薦使用 [安裝包方式](https://1panel.cn/docs/installation/package_installation/) 進行安裝部署。
+如果是用於離線環境，推薦使用 [安裝包方式](https://docs.fit2cloud.com/1panel/installation/enterprise-installation/) 進行安裝部署。
 
 **學習資料**
 
-- [線上文件](https://1panel.cn/docs/)
+- [線上文件](https://docs.fit2cloud.com/1panel/)
 - [社區論壇](https://bbs.fit2cloud.com/c/1p/7)
 - [如何加入微信交流群?](https://bbs.fit2cloud.com/t/topic/2147)
 

docs/container-network-cleanup.md

@@ -0,0 +1,33 @@
+# 批量清理未使用的容器网络
+
+`POST /api/v2/containers/network/clean`，使用现有 1Panel 认证，不需要请求体。
+这是异步接口，立即返回 `data.taskID`。实际结果在任务中心和任务日志查看，HTTP 请求结束不会取消任务。
+
+规则：
+
+- 永远保留 `none`、`host`、`bridge`、`1panel-network`。
+- 保留所有容器（包括停止容器）引用的网络，删除前再通过 NetworkInspect 检查连接。
+- 不检查 Compose 文件或应用配置引用；仅被这些配置引用、没有容器连接的普通网络也会被删除。
+- 当前仅清理本地普通网络，跳过 Swarm、ingress 和 config-only 网络。
+- Docker 删除时发现新连接会拒绝删除；不会强制断开容器。
+
+成功响应示例：
+
+```json
+{"code":200,"data":{"taskID":"服务端生成的任务 UUID"}}
+```
+
+任务类型为 Container，操作为 TaskClean。网络页“清理网络”按钮调用此接口，拿到 taskID 后立即打开任务日志，并刷新任务中心。
+每个网络处理完成后写入名称、ID 和结果；有容器连接或 Docker 返回占用冲突时，明确记录“未删除”。最后记录已删除、跳过和失败数量。
+跳过在用网络属于正常结果；检查/删除失败会将任务标记为失败，已删除的网络不回滚。
+
+日志调用链：
+
+- `agent/app/service/container_network_cleanup.go`：NewTaskWithOps 创建任务，AddSubTask 注册清理步骤，Execute 执行任务，t.Log 写逐项日志。
+- `agent/app/task/task.go`：任务持久化，维护状态，写入开始、结束标记；日志文件为 `<global.Dir.TaskDir>/Container/<taskID>.log`。
+- `frontend/src/views/container/network/index.vue`：请求成功后调用 openTaskLog(taskID)。
+- `frontend/src/components/log/task/index.vue`：openWithTaskID 打开日志弹窗。
+- `frontend/src/components/log/file/index.vue`：通过 `/logs/tasks/read` 按行读取任务日志。
+
+范围：不检查 Compose 文件或应用配置引用；仅清理本地普通网络。原 `/containers/prune` 接口仍保持原逻辑。
+测试使用假 Docker 客户端和临时任务数据库，不执行真实网络删除。

frontend/package.json

@@ -57,18 +57,18 @@
         "vue-codemirror": "^6.1.1",
         "vue-demi": "^0.14.6",
         "vue-i18n": "^11.4.4",
-        "vue-router": "^5.2.0"
+        "vue-router": "^5.3.1"
     },
     "devDependencies": {
         "@tailwindcss/postcss": "^4.3.3",
-        "@types/node": "^26.4.0",
+        "@types/node": "^26.6.3",
         "@types/uuid": "^11.0.0",
         "@typescript-eslint/eslint-plugin": "^8.65.0",
         "@typescript-eslint/parser": "^8.68.0",
         "@vitejs/plugin-vue": "^6.0.7",
         "@vitejs/plugin-vue-jsx": "^5.1.6",
         "@vue/compiler-sfc": "^3.5.42",
-        "autoprefixer": "^10.5.4",
+        "autoprefixer": "^10.6.1",
         "esbuild": "^0.28.2",
         "eslint": "^10.8.0",
         "eslint-config-prettier": "^10.1.8",
@@ -79,7 +79,7 @@
         "postcss-html": "^2.0.0",
         "prettier": "^3.9.6",
         "rollup-plugin-visualizer": "^7.0.1",
-        "sass": "^1.103.1",
+        "sass": "^1.105.0",
         "standard-version": "^9.5.0",
         "tailwindcss": "^4.3.3",
         "typescript": "^6.0.3",
@@ -88,7 +88,7 @@
         "vite": "^8.2.0",
         "vite-plugin-compression": "^0.5.1",
         "vite-plugin-eslint2": "^5.3.0",
-        "vite-svg-loader": "^5.1.2",
+        "vite-svg-loader": "^5.1.3",
         "vue-eslint-parser": "^10.4.1",
         "vue-tsc": "^3.3.9"
     },

frontend/src/api/interface/ai.ts

@@ -22,6 +22,8 @@ export namespace AI {
     }
 
     export interface Info {
+        collectedAt?: string;
+        warnings?: string[];
         cudaVersion: string;
         driverVersion: string;
         type: string;
@@ -31,6 +33,34 @@ export namespace AI {
         xpu: XpuInfo['xpu'];
     }
     export interface GPU {
+        eccPending?: string;
+        eccErrors?: { scope: string; correctable: string; uncorrectable: string }[];
+        memoryActivity?: string;
+        encoderUtil?: string;
+        decoderUtil?: string;
+        jpegUtil?: string;
+        ofaUtil?: string;
+        mediaUtil?: string;
+        hotspotTemperature?: string;
+        fanRPM?: string;
+        mediaFrequency?: string;
+
+        uuid?: string;
+        driverVersion?: string;
+        architecture?: string;
+        frequency?: string;
+        memoryFrequency?: string;
+        memoryTemperature?: string;
+        memoryFree?: string;
+        memoryReserved?: string;
+        powerLimit?: string;
+        defaultPowerLimit?: string;
+        pcieGeneration?: string;
+        pcieMaxGeneration?: string;
+        pcieWidth?: string;
+        pcieMaxWidth?: string;
+        clockEvents?: string[];
+        processStatus?: string;
         type: 'nvidia' | 'amd';
         index: number;
         productName: string;
@@ -52,6 +82,12 @@ export namespace AI {
         processes: Process[];
     }
     export interface NPU {
+        aiCPUUtil?: string;
+        ctrlCPUUtil?: string;
+        ddrBandwidth?: string;
+        hbmBandwidth?: string;
+
+        processStatus?: string;
         type: 'ascend';
         index: number;
         npuIndex: number;
@@ -73,18 +109,24 @@ export namespace AI {
         processes: NPUProcess[];
     }
     export interface MonitorGPUSearch {
+        aggregation?: 'avg' | 'max';
+        deviceID?: string;
+        legacy?: boolean;
         productName: string;
         startTime: Date;
         endTime: Date;
     }
     export interface MonitorGPUOptions {
+        supported: boolean;
         gpuType: string;
         options: Array<string>;
         chartHide: Array<ChartHide>;
     }
     export interface ChartHide {
+        deviceID?: string;
+        legacy?: boolean;
         productName: string;
-        type?: 'gpu' | 'xpu';
+        type?: 'gpu' | 'npu' | 'xpu';
         process: boolean;
         gpu: boolean;
         memory: boolean;
@@ -94,17 +136,45 @@ export namespace AI {
         speed: boolean;
     }
     export interface MonitorGPUData {
+        memoryActivity: Array<number | null>;
+        encoderUtil: Array<number | null>;
+        decoderUtil: Array<number | null>;
+        jpegUtil: Array<number | null>;
+        ofaUtil: Array<number | null>;
+        mediaUtil: Array<number | null>;
+        computeUtil: Array<number | null>;
+        copyUtil: Array<number | null>;
+        hotspotTemperature: Array<number | null>;
+        fanRPM: Array<number | null>;
+        aiCPUUtil: Array<number | null>;
+        ctrlCPUUtil: Array<number | null>;
+        ddrUsed: Array<number | null>;
+        ddrTotal: Array<number | null>;
+        hbmUsed: Array<number | null>;
+        hbmTotal: Array<number | null>;
+        ddrBandwidth: Array<number | null>;
+        hbmBandwidth: Array<number | null>;
+        memoryBandwidth: Array<number | null>;
+        mediaFrequency: Array<number | null>;
+        hugepagesUsed: Array<number | null>;
+        hugepagesTotal: Array<number | null>;
+
+        bucketSeconds: number;
+        sampleCount: number;
+        memoryTemperatureValue: Array<number | null>;
+        frequencyValue: Array<number | null>;
+        memoryFrequencyValue: Array<number | null>;
         date: Array<Date>;
-        gpuValue: Array<number>;
-        temperatureValue: Array<number>;
-        powerTotal: Array<number>;
-        powerUsed: Array<number>;
-        powerPercent: Array<number>;
-        memoryTotal: Array<number>;
-        memoryUsed: Array<number>;
-        memoryPercent: Array<number>;
-        speedValue: Array<number>;
-        processCount: Array<number>;
+        gpuValue: Array<number | null>;
+        temperatureValue: Array<number | null>;
+        powerTotal: Array<number | null>;
+        powerUsed: Array<number | null>;
+        powerPercent: Array<number | null>;
+        memoryTotal: Array<number | null>;
+        memoryUsed: Array<number | null>;
+        memoryPercent: Array<number | null>;
+        speedValue: Array<number | null>;
+        processCount: Array<number | null>;
         gpuProcesses: Array<Array<GPUProcess>>;
     }
     export interface GPUProcess {
@@ -132,12 +202,15 @@ export namespace AI {
     }
 
     interface Xpu {
+        tiles?: { tileID: number; stats: Stats }[];
+        processStatus?: string;
         basic: Basic;
         stats: Stats;
         processes: XpuProcess[];
     }
 
     interface Basic {
+        uuid?: string;
         deviceID: number;
         deviceName: string;
         vendorName: string;
@@ -148,6 +221,13 @@ export namespace AI {
     }
 
     interface Stats {
+        mediaUtil?: string;
+        computeUtil?: string;
+        copyUtil?: string;
+        mediaFrequency?: string;
+        memoryTemperature?: string;
+
+        memoryBandwidthUtil?: string;
         power: string;
         gpuUtil: string;
         frequency: string;

frontend/src/api/interface/alert.ts

@@ -1,4 +1,5 @@
 import { CommonModel, ReqPage } from '@/api/interface';
+import type { CronjobAlertTriggerMode } from '@/utils/cronjob-alert';
 
 export namespace Alert {
     export interface AlertInfo extends CommonModel {
@@ -9,22 +10,25 @@ export namespace Alert {
         interval: number;
         method: string;
         title: string;
+        taskName?: string;
         project: string;
         status: string;
         sendCount: number;
         sendMethod: string[];
         advancedParams: string;
+        alertTriggerMode?: CronjobAlertTriggerMode;
         createUser?: string;
         updateUser?: string;
     }
 
     export interface AlertDetail {
         type: string;
+        subType?: string;
         licenseId: string;
         title: string;
         project: string;
         method: string;
-        params: string;
+        params: { index: string; key: string; value: string }[] | string;
     }
 
     export interface AlertUpdateStatusReq {
@@ -74,6 +78,7 @@ export namespace Alert {
         project: string;
         status: string;
         sendCount: number;
+        advancedParams?: string;
     }
 
     export interface AlertUpdateReq extends CommonModel {
@@ -86,6 +91,7 @@ export namespace Alert {
         project: string;
         status: string;
         sendCount: number;
+        advancedParams?: string;
     }
 
     export interface DelReq {